Landscape in Language: Transdisciplinary perspectives

Landscape in Language

Culture and Language Use Studies in Anthropological Linguistics CLU-SAL publishes monographs and edited collections, culturally oriented grammars and dictionaries in the cross- and interdisciplinary domain of anthropological linguistics or linguistic anthropology. The series offers a forum for anthropological research based on knowledge of the native languages of the people being studied and that linguistic research and grammatical studies must be based on a deep understanding of the function of speech forms in the speech community under study. For an overview of all books published in this series, please see http://benjamins.com/catalog/clu

Editor Gunter Senft

Max Planck Institute for Psycholinguistics, Nijmegen

Volume 4 Landscape in Language. Transdisciplinary perspectives Edited by David M. Mark, Andrew G. Turk, Niclas Burenhult and David Stea

Landscape in Language Transdisciplinary perspectives Edited by

David M. Mark University at Buffalo (SUNY)

Andrew G. Turk Murdoch University

Niclas Burenhult Max Planck Institute for Psycholinguistics, Nijmegen and Lund University

David Stea Texas State University

John Benjamins Publishing Company Amsterdamâ•›/â•›Philadelphia

8

TM

The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984.

Library of Congress Cataloging-in-Publication Data Landscape in language : transdisciplinary perspectives / edited by David M. Mark ... [et al.].        p. cm. (Culture and Language Use, issn 1879-5838 ; v. 4) Includes bibliographical references and index. 1.  Language and languages--Variation. 2.  Semantics. 3.  Geographical perception. 4.  Language and culture.  I. Mark, David M. P120.V37L25â•…â•… 2011 910’.02014--dc22 2011003203 isbn 978 90 272 0286 4 (Hb ; alk. paper) isbn 978 90 272 8704 5 (Eb)

© 2011 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Co. · P.O. Box 36224 · 1020 me Amsterdam · The Netherlands John Benjamins North America · P.O. Box 27519 · Philadelphia pa 19118-0519 · usa

Table of contents

Foreword

ix

Preface 

xi

Landscape in language: An introduction David M. Mark, Andrew G. Turk, Niclas Burenhult and David Stea Ethnophysiography Andrew G. Turk, David M. Mark and David Stea

1

25

Exploring philosophy of place: Potential for synergy between  phenomenology and ethnophysiography Andrew G. Turk

47

Embedded in place: ‘Mirror knowledge’ and ‘simultaneous  landscapes’ among Māori Brian Murton

73

Philosophical issues in ethnophysiography: Landform terms,  disciplinarity, and the question of method Bruce B. Janz

101

‘Land’ and life: Ethnoecology and ethnogeography as complementary  approaches to the analyses of landscape perception  Chris S. Duvall

121

Landscape in Western Pantar, a Papuan outlier of southern Indonesia Gary Holton

143

vi

Landscape in Language: Transdisciplinary Perspectives

Hawaiian storied place names: Re-placing cultural meaning Renee Pualani Louis

167

Between the trees and the tides: Inuit ways of discriminating space  in a coastal and boreal landscape Scott A. Heyes

187

Differing conceptualizations of the same landscape: The Athabaskan  and Eskimo language boundary in Alaska Gary Holton

225

A case study in Ahtna Athabascan geographic knowledge James Kari

239

Revitalizing place names through stories and songs Susan Paskvan

261

Language and landscape among the Tlingit Thomas F. Thornton

275

Language, landscape and ethnoecology, reflections  from northwestern Canada Leslie Main Johnson

291

Landscape embedded in language: The Navajo of Canyon de Chelly,  Arizona, and their named places  Stephen C. Jett

327

Navajo landscape and its contexts Carmelita Topaha

343

Navigating regional landscapes with Jicarilla personal narrative Elizabeth M. Lynch

353



Table of contents vii

Ontology of landscape in language Werner Kuhn The role of geospatial technologies for integrating landscape  in language: Geographic Information Systems and the Cree  of northern Quebec Renée Sieber and Christopher Wellen

369

381

Classifying landscape character Lars Brabyn and David M. Mark

395

Perspectives on the ethical conduct of landscape in language research Andrew G. Turk and David M. Mark

411

Notes on contributors

435

Index

443

Foreword

This book substantially advances a nascent interdisciplinary field focused on the human cognition of the natural world we inhabit, as reflected especially though language, but also through other kinds of action. Here, archaeology, geography, psychology, anthropology and linguistics all intersect in mutually informing€ways. This book is not just another “Language and X” book, where X is some other random, perhaps trivial, domain. Perhaps the term ‘Landscape’ doesn’t help here: according to the OED, it came into English at the end of the sixteenth century from Middle Dutch, hitch-hiking on the small easel paintings produced for the newly formed urban bourgeois market in such things. From there, it was rapidly generalized to views and vistas, and then more slowly to the Romantic landscape-appreciation and garden making of the eighteenth century. That sentiment born of a vanishing countryside is not the subject matter here. Instead the focus of this book is on our Umwelt, the terrain and water worlds we inhabit and exploit. As yet we have no better widely-accepted term, however, that captures this interdisciplinary domain. Our relationships with our ecologies are obviously of fundamental importance, and may yet determine the fate of the species on this planet. But on the surface it is quite unclear that concepts of landscape and ecology are systematically reflected in language. Many domains important to human existence have poor lexicalization in the world’s languages, and show little or no grammatical reflex (consider e.g. the limited terms for disease or internal organs in many unwritten languages). A moment’s reflection though shows that landscape is different. For a start, there are just two great systems of proper nouns in just about all languages: namely personal names, and toponyms. Proper nouns are names, which usually have special grammatical reflexes (as in English where determiners are usually disallowed, as in the oddity of the London, the Mount Everest, etc.), even if ultimately what is a proper noun is determined pragmatically by usage. This property of nameability or properhood reveals a deep psychological reality. Persons on the one hand and places on the other are our two great mental index systems – they are the two coordinate systems we use to plot our social and ecological spaces. Naturally, the two systems intersect: we think of places in terms of persons, and persons in terms of places. Both systems are underpinned by specialized neural

˘

Landscape in Language: Transdisciplinary Perspectives

circuitry. Both derive their cognitive power from the fact that they name nodes in great networks – a person is joined by kinship or association links to a field of other persons, and a place is connected by pathways to a network of other places. So finding or naming a node in the network releases the computational power afforded by the network – all the inferences about connections, shortest routes, triangulations and so forth. Toponyms have been the focus of extensive study, especially by those interested in the history they reveal due to their extraordinary stability through time€– a stability of course that is derived from their cognitive centrality. Proper nouns, names, are merely a way of picking out specific persons or locations – a shorthand instruction to find a unique particular of a certain class. Normally, then, toponyms ride on a classification of the underlying landforms: Mount Everest is a unique exemplar of the class of mountains, the Thames of the class of rivers, Lake Michigan of the class of lakes, etc. Toponyms thus presuppose cognitive categories, normally expressible in common nouns (although the relation is sometimes more complex). This book focuses on that underlying cognitive classification, as reflected (perhaps imperfectly) in language. This topic – unlike the study of toponyms in the major Western languages – has been relatively very neglected, and it is only in the years immediately preceding this volume that it has come to the fore as a major focus of interdisciplinary work. This book samples this recent work right across the languages and the varied ecologies of the world, from Maori concepts of landscape in New Zealand to Inuit conceptions of the arctic world. It offers a wonderful panorama of languages, landscape types and methods of analysis, from linguistics to ethnography, geography and GIS to philosophy. It also demonstrates the special added value of interdisciplinary work: none of these studies would have the rich, descriptive ‘bite’ they have without being deeply informed by parallel work in other disciplines. One of the signal attractions of this domain is that scientific inquiry is still in the nascent stages. There are still fundamental discoveries to be made, and it will be years before we have achieved a thorough synthesis of all this new material. There are many fundamental open questions about the underlying nature of the cognitive categories involved, how universal they are or how molded by local ecologies or local linguistic structures, and how closely language, cognition, action and use interact in this domain. This book will inspire further research on these issues, while remaining a landmark (forgive the inevitable metaphor) for years to come.   

Stephen C. Levinson Nijmegen, The Netherlands 18 December 2010

Preface

This book is an outcome of an international collaborative effort that began more than eight years ago, and resulted in a transdisciplinary workshop held in New Mexico and Arizona (USA) in October and November 2008. The idea for the workshop followed from the interaction of two interrelated research projects: a research program on space and landscape, coordinated by Stephen C. Levinson and Niclas Burenhult, in the Language and Cognition Group of the Max Planck Institute for Psycholinguistics (MPI) in Nijmegen, the Netherlands; and the “Ethnophysiography Project” of David Mark (University at Buffalo, USA), Andrew Turk (Murdoch University, Australia), and David Stea (Center for Global Justice, San Miguel de Allende, Mexico). A place names project at MPI was launched in 2001, and Mark visited MPI during May 2002. Then Mark worked with Turk to launch ethnophysiographic fieldwork in Northwestern Australia in October 2002. Stea joined the Ethnophysiography Project in 2003 to coinitiate a case study on the Navajo language. Turk, who had already visited the Navajo Nation, visited MPI in October 2005 to update researchers there on progress of the ethnophysiography case studies and to discuss enhanced collaboration. The researchers from MPI conducted studies in several languages, leading up to a special issue of Language Sciences (Burenhult 2008). Mark and Turk assisted with this publication and visited MPI together in June 2007, at which time the idea of having an international workshop on the topic of landscape in language was discussed and confirmed. In April 2008, the National Science Foundation awarded a grant (BCS-0753737, David Mark, PI) to support the workshop. A call for participants was published internationally, and twenty-eight people were invited to the workshop. Participants gathered in Albuquerque, New Mexico, on Sunday, October 26, 2008, to launch the workshop. Monday morning started with introductory remarks by the organizers, followed by seven presentations. The next day, participants boarded a chartered bus to travel west to the Navajo Reservation. The group stopped for a while in Window Rock, capital of the Navajo Nation, where participants learned about the activities of the Navajo Nation Cultural Preservation Office and their Geographic Information Systems office. The group then traveled

xii Landscape in Language: Transdisciplinary Perspectives

north to Canyon de Chelly, near Chinle, Arizona, still on the Navajo Reservation, where the remainder of the workshop was held at historic Thunderbird Lodge. During all of Wednesday and Thursday morning, more presentations were given and the issues they raised were discussed. Thursday afternoon was spent touring the magnificent Canyon de Chelly, visiting ancient cliff dwellings situated on the red sandstone cliffs, and meeting Navajo artists. After Friday morning’s final presentations, the afternoon was occupied by three sessions of panel and audience discussions of critical topics: Ethical Issues in Indigenous Landscape Research; Methods and Theory/Methods Interaction; and Priorities for Future Research. On Saturday the group traveled back to Albuquerque, with a long stop in Newcomb, New Mexico, for a lunch and cultural celebration with members of the local Navajo community. This book presents chapters written by many of the workshop participants. All participants were invited to submit chapters and contributions to the volume were edited by members of the workshop organizing committee. The editors are pleased that the book is being published by John Benjamins as part of their new series, “Culture and Language Use: Studies in Anthropological Linguistics (CLUSAL),” edited by MPI’s Gunter Senft, who participated in the workshop. Many people and organizations contributed to the success of the workshop and the production of this book. The editors wish to thank the U.S. National Science Foundation for awarding grant BCS-0753737 to support the workshop and the preparation of this book. We are also grateful to the Max Planck Society for funding the participation of some MPI staff. Linda Doerfler and Sunita Gupta in the offices of NCGIA-Buffalo assisted with the paperwork and arrangements for the workshop, from the proposal to the final travel reimbursements. Silke Lambert copy-edited all of the chapters, bringing them into compliance with the publisher’s format guidelines and detecting a few logical or grammatical inconsistencies; her contribution to the production of the book is greatly appreciated. We especially wish to thank Ron Maldonado, of the Historic Preservation Office, Navajo Nation, who helped arrange our visit to Navajo Nation offices and participated in the workshop, and his colleagues who spoke to our group in Window Rock. We also thank Larry King, who conducted a Navajo blessing of our trip into Canyon de Chelly, and Carmelita Topaha, who organized our visit to Newcomb chapter. The workshop participants who did not contribute chapters for this book nevertheless played an important role in the discussions and presentations. In alphabetical order, these are: Shonto Begay, Clair Hill, Jay Johnson, Karen Kemp, Larry King, Asifa Majid, Ron Maldonado, Carolyn O’Meara, and Gunter Senft. Several of these people, along with many of the chapter authors, also reviewed other book



Preface xiii

chapters. We wish to thank all the workshop participants for the free and open sharing of knowledge and opinions, much of which appears in this book.     

David M. Mark, Amherst, New York, USA Andrew G. Turk, Perth, Western Australia Niclas Burenhult, Lund, Sweden David Stea, San Miguel de Allende, Guanajuato, México February 2011

chapter 1

Landscape in language An introduction David M. Mark, Andrew G. Turk,  Niclas Burenhult and David Stea

1.

Introduction

The relationships that people have with landscape, individually and collectively, have long formed an important research theme in several disciplines, notably geography, anthropology, philosophy, and psychology. A “man-land tradition,” now often re-labeled as a “human-environment tradition,” was one of the four traditions of geography identified in Pattison’s (1964) classic article on the intellectual core of geography. This tradition and its successors have focused on land use and land-based activities. Anthropologists also examine the relations of people to their environments, mainly emphasizing cultural aspects, and an anthropology of landscape has developed (see especially Ingold 2000). The fundamental relations between culture and landscape, and attachment to landscape, have been discussed in geography (Tuan’s 1974 idea of topophilia) and in anthropology (for example, Keith Basso’s 1996 book Wisdom Sits in Places). Philosophers of place have often taken a phenomenological approach to the relationship of people to ‘lived-in’ landscape (e.g. Casey 1996; Malpas 1999, 2007). However, until recently, there has been relatively little scholarly research on how landscape is conceptualized, that is, how a continuous land surface, a landscape, becomes cognitive entities, and how those entities are classified and represented in language and in thought. Toponyms, the proper names given to geographic features, have certainly been studied, but the relation of the generic parts of such names to geographic categories has received considerably less attention (although see Zelinsky 1955). There has been even less work on cross-cultural and cross-linguistic variations and similarities in delimitation, classification, and naming of geographic features. The definition of “landscape” itself is not necessarily simple. Granö (1997) suggested that landscape consists of the more distant parts of the human environment, and used the term proximity for the part of the environment close to the

˘

David M. Mark et al.

observer. Granö claimed that landscape is perceived mainly through the visual sense. While intuitively appealing, this definition is not unproblematic. For example, senses other than vision may also be significant: Feld (1996) has drawn attention to the importance of sound in the nature of some places, and similar claims could be made for odor (e.g. the smell of vegetation types, water) and skin sensation (e.g. temperature, wind). Multi-sensory perception of landscape is discussed further in the chapter by Louis in this volume. Furthermore, the proximity/landscape dichotomy is itself difficult to operationalize, and people may have different ideas as to whether they see landscape and its features as a distant backdrop or a surrounding environment that necessarily contains them (cf. Burenhult & Levinson 2008). There is the possibility that whether a landscape feature or scene is considered to be (as distinguished from “perceived”, in the psychological sense) nearby or distant may be related to the relative intimacy of attachment to that feature or scene. As with “environment,” there is always the question of the role that is played by the human. Does the person perceive him/herself as necessarily a part of the landscape, or apart from it, as observer, listener, etc.? In other words, is the observer active or passive, and is the landscape “distant” in any relative or absolute sense? On the cognitive level, what is the relation of landscape to spaces too large or complex to be apprehended from a single vantage point? Such spaces are termed “transperceptual spaces” (Downs & Stea 1977) because, to construct the overall concept of the place, a series of direct perceptual experiences would need to be integrated over time. Despite the apparent lack of clarity in the “proximal/distal” distinction regarding landscape, this distinction has been accepted in the psychology of perception for more than a century. For instance, it is implicit in Gibson’s later work (Gibson 1979, 1987; Gibson & Bridgeman 1987; Reed & Jones 1982), in his use of “contours” and “textural gradients,” for example. However, a definition relying on common understanding of the proximal/ distal distinction may be difficult to reconcile with notions of emotional and cultural attachment to landscapes, landscape features, and places (Basso 1996; Tuan 1974). For example, Ingold’s (2000) approach to anthropology of landscape involves concepts such as dwelling and livelihood, and does not recognize a distinction between natural and human-modified landscapes. There is debate in the literature concerning the sources and meanings of the English term landscape and its historical and conceptual links to similar terms in associated languages, such as the German Landschaft (Cosgrove 2004; Ingold 2010; Olwig 2008). Ingold (2010) discusses the visual bias in most accounts of landscape and suggests that: “… we might do well to return to an earlier understanding of landscape – one that is closer to the ground, more haptic than optical” (p. 17). Cosgrove (2004) notes that: “Spatially, landscape was constructed as a



Landscape in language

bounded and measured area, an absolute space, represented through the scientific techniques of measured distance, geometrical survey, and linear perspective” (p. 62). He goes on to make the case for definitions of landscape and Landschaft which are not so bound to measurement orthodoxy but incorporate a sense of social construction: Much recent scholarship has sought to unmask and denaturalize landscape, paying as much attention to its pictorial and literary representations as to material spaces themselves. In refusing to take landscape “at face value,” such landscape study moves beyond Landschaft in its original Germanic sense, beyond the pictorial English sense of landscape as an aesthetically unified space, and beyond the traditional geographical sense of landscape as an expression of ecological relations between land and life. It draws upon and contributes to the revised ways of conceptualizing space with which I opened this discussion, regarding space as a function of natural and social processes, but also as an outcome that in turn has social agency, able to create and transform the material world. Landscape’s revival within contemporary geography derives from those aspects embedded in its conceptual history that allow it to transcend the modernist dualism (perhaps dialectic) of nature and culture.  (p. 68)

This sentiment is echoed by Smith (1989:â•›109): “With a dialectical conception of geography more rooted today, and the mutual interdependence of social theory and geography increasingly evident, the time has proven ripe for a more serious and long-overdue reexamination of landscape that moves beyond narrow descriptive, aesthetic, and idealistic confines.” The Landscape in Language Workshop, and this book, open up new perspectives on this discussion of the relationship between physical attributes of land and the meanings attributed to it by those who dwell in that place. To explain all of these issues would involve a lengthy analysis of relevant aspects of cultural geography, phenomenology, semiotics, linguistics, human ecology, environmental psychology, etc., impractical to cover adequately in this chapter. However, some discussion of these issues is included below and in other chapters of this volume.

2.

The linguistics of landscape

The language sciences have no tradition of research on landscape. The only branch of linguistics with a vested interest in landscape is onomastics (the study of names), but typically the study of place names (toponymy) has been approached in a way totally divorced from the actual referential subject matter.

˘

˘

David M. Mark et al.

In fact, landscape has several characteristics that give it potential to exert considerable influence on theorization of key problems throughout linguistics. How do languages select geographic entities as objects to be labeled (‘mountain,’ ‘river,’ ‘valley’)? Are there universal categories? What is the formal and referential relationship between common nouns (landscape terms) and proper nouns (toponyms or place names)? Are terms for landforms easily translatable across languages (mountain–montagne–Berg)? What are the ontological principles behind such terms? Do they involve structured sets of lexicon, semantic fields and relations, with possible repercussions in grammar? How much variation exists in categorial strategies across languages and speakers? What are the factors that drive such variation? Does variation in linguistic representation have resonance in cognition? Questions such as these indicate that linguistics as a discipline may have a lot to gain from in-depth exploration of the landscape domain. Language and linguistics also have great potential to be of help in studying various aspects of human-landscape relationships (Levinson 2003; Levinson & Wilkins 2006; Majid et al. 2004). Language reflects a range of cultural and cognitive preoccupations, and linguistics has tools and models for identifying, describing and explaining representations of landscape of key concern to other branches of science. Thus, landscape opens up important links between linguistics and disciplines with a longer tradition of interest in the domain that usually do not have a major focus on language, such as anthropology, archaeology, environmental psychology, philosophy, and cognitive geography (e.g. Bell et al. 1996; Bender 1993; Hirsch & O’Hanlon 1995; Tilley 1994; Mark et al. 1999; Smith & Mark 2001). Here, linguistic attention to the domain is certain to unleash a variety of new questions and perspectives of inquiry. An indication of the potential of linguistic exploration of landscape was provided by a recent pilot study at the Max Planck Institute for Psycholinguistics, initiated by its director Stephen C. Levinson in 2004 and coordinated by Niclas Burenhult. The study – whose results were published as a special issue of the journal Language Sciences (Burenhult 2008a) – was a cross-cultural inquiry into linguistic categorization of the landscape domain. Case studies of language-specific systems were carried out with uniform methods by language experts across a sample of nine geographically and genealogically diverse languages. The following were the general research questions pursued by the study: 1. How is landscape divided into categories, and how are these categories named? Are there cross-linguistic differences in how landscape is divided into categories? Which are the main determinants of landscape categorization?



Landscape in language

2. How do we formally identify place names? To what places do place names refer? How are places semantically construed for this purpose? 3. What is the denotational relation between landscape terms and place names? Research topics and methods of investigating them were detailed in a questionnaire formulated by Bohnemeyer et al. (2004) and taken to the field by language experts for first-hand fieldwork. The language settings included in the study were Tzeltal (Mayan, Mesoamerica, Brown 2008), Jahai (Austroasiatic, Malay Peninsula, Burenhult 2008b), Marquesan (Austronesian, Polynesia, Cablitz 2008), Lao (Tai, Mainland Southeast Asia, Enfield 2008), Yélî Dnye (isolate, Island Melanesia, Levinson 2008), Lowland Chontal (isolate, Mesoamerica, O’Connor & Kroefges 2008), Seri (isolate, Mesoamerica, O’Meara & Bohnemeyer 2008), KiliÂ� vila (Austronesian, Island Melanesia, Senft 2008), and ≠Akhoe Hai//om (Khoisan, southwestern Africa, Widlok 2008). The introductory article of the special issue of Language Sciences mentioned above outlined the main results of the study and the theoretical relevance of landscape to the discipline of linguistics (Burenhult & Levinson 2008). A major finding was that languages are extremely diverse in how they categorize landscape features and name places, and in how the two ontological categories of landforms and place names are related. Languages were shown to vary along the following dimensions: 1. The denotation of landform terms (‘mountain,’ ‘river,’ ‘valley’): different languages carve out and classify similar landforms in different ways and indeed differ also in which features get labeled. For example, terms for which the English translation ‘mountain’ is used vary considerably in how they extend their meaning with respect to the magnitude, shape, substance, and boundedness of referents; also, equivalents of some presumably basic landform terms are not universally present, notably ‘valley.’ 2. Whether landscape can be considered a discrete semantic domain: in some languages there is structural evidence for claiming the existence of such a distinct domain, in others there is evidence against it. 3. Semantic themes: mapping and labeling of landforms draw on a variety of ontological design principles, including metaphor and analogy involving body, animacy, agency, containment, and so on.

. These results find clear parallels in the recent work by Mark, Turk and Stea on the Yindjibarndi language, discussed in Chapter 2.

˘

˘

David M. Mark et al.

4. How proper nouns (i.e. place names) are linguistically generated: the formal tools used (e.g. form classes, syntactic units, morphological complexity) vary across languages, as do the lexical source domains from which names are drawn (landscape features, body parts, animals, plants, objects, people, proper names, activities, etc.); there is also a great deal of variation within and between languages as to the degree to which the meaning of place names can be analyzed, since they frequently have archaic characteristics. 5. Which geographical categories get named, e.g. whether names map neatly onto generic landform categories or if they represent an ontologically distinct system; both extremes were present in the language sample cited, some languages naming only features with a corresponding generic label and others systematically not doing so. The study identified a number of linguistic phenomena which apply particularly to the landscape domain, or find in landscape their main locus operandi, and which have considerable relevance in a wider linguistic, cultural and cognitive context. One such phenomenon consists of sets of semantic distinctions in a particular language that structure the lexicon across more than one word class, so that, for example, the same set of semantic oppositions recur in landscape nouns, place names, motion verbs and directional terms. The discovery of this descriptively and theoretically overlooked phenomenon, dubbed ‘semplates’ (a blend of “semantic templates”) and in fact not restricted to landscape, was possible thanks to in-depth analysis of a domain whose fundamental character is likely to be crucial for such structures to develop. The concept of semplates and its theoretical significance is elaborated in Levinson and Burenhult (2009), and a field guide for their identification and exploration is provided in Burenhult and Levinson (2009). Another ontological puzzle addressed by Burenhult and Levinson (2009) was the tendency of languages to treat landscape terms as systematically ambiguous regarding whether their referents are considered to be objects or places, a phenomenon noted by Lyons (1977) and followed up by Smith and Mark (2001) and Cablitz (2008). Size and perceptual (non)discreteness are factors that influence to which ontological category a landscape term is assigned in a given context, with considerable syntactic ramifications (see e.g. Cablitz 2008; cf. English expressions such as ‘the big forest’ vs. ‘a house in the forest’). Although fundamental, landscape and its features therefore have a dual nature, with challenging consequences for human formal representational systems. This and other facts turn landscape, surprisingly, into a center stage for the study of grammar. Yet another set of ontological problems concerns the relationship between categorization of landscape and categorization of the human body and its parts.



Landscape in language

The two can be thought of as parallel domains offering similar categorial affordances (Gibson 1979) and challenges, with diffuse natural borders, part-whole relationships, omnipresence etc. Languages frequently use the lexicon of one of these domains to map and name the other in more or less systematic and pervasive ways through analogy and metaphor (Brown 2008; Burenhult 2008b; Cablitz 2008). To summarize, even a limited exploratory study based on a handful of languages has managed to uncover not only interesting patterns of linguistic diversity and illuminating examples of classical linguistic problems, but also previously unknown linguistic phenomena of considerable theoretical interest. Thus, landscape shows all the signs of being a fascinating domain for studying the dynamics and resourcefulness of language.

3.

Issues in ethnophysiography

Mark and Turk (2003a, b, 2004) coined the term ‘ethnophysiography’ to cover their research on the topic of cultural and linguistic aspects of landscape concepts, in the context of their ongoing case study of the Australian indigenous language Yindjibarndi. To some extent this research was a natural extension of earlier work on geographical ontology, and especially on geographic categories, in some European languages (Smith & Mark 2001). The core of ethnophysiography is the investigation (for any particular language) of categories of landscape features, especially those denoted by common words (usually nouns or noun phrases). Those terms and their definitions form a research topic of considerable importance in its own right. But an understanding of the landscape vocabulary also provides foundations for understanding other important dimensions of ethnophysiography, including the study of knowledge systems, beliefs, and customs of a people concerning landforms and landscapes. Thus, ethnophysiography is related to the study of ‘place,’ ‘sense of place,’ and ‘place attachment.’ Ethnophysiography examines how these significances are tied into traditional beliefs, such as those embedded in creation stories, which help to make sense of the world, of its physiographic entities, and of the relationship of such entities to everyday activities, including traditional cultural practices (ceremonies, music, art, etc.). An initial set of broad research questions includes: – Do all people, from different cultural/linguistic groups, think about landscape in more or less the same way, or are there significant cross-cultural and crosslinguistic differences in the ways human beings conceptualize their environments at landscape scales?

˘

˘

David M. Mark et al.

– How important is the nature of the particular landscape that provides the environment for a speech community, and especially the range of forms in that landscape, in the development of the particular landscape category system and terms used by a speech community? – How influential are the culture and lifestyle of the people on their conceptions of landscape; i.e., what is the impact of human interaction with the landscape? – In what ways does the nature of the language itself (its grammar etc.) contribute to the way terms for landscape are structured? – Can alternative worldviews, as expressed in representations of landforms, be utilized to produce appropriate, culturally specific geographic information systems (GIS)? The use of proper names (toponyms) for individual landscape features, such as “Golden Hill,” as against generic terms for classes of features (and instances of this class – e.g. any hill), is also an important aspect of ethnophysiography. Why do some landscape features acquire proper names while others do not? Are proper names more frequent for some classes of landscape features and less common for other classes? Are there cultures that rely predominantly on proper names, rather than a more extensive set of generic terms, and vice versa? Turk et al. (Chapter 2, this volume) have suggested that a way of making some of these research questions more tractable is to construct a descriptive model of factors relevant to ethnophysiography. There are various ways (dimensions) in which the attributes of places and peoples differ. Also, there are numerous common ways for languages to vary in their ways of referring to landscape. Researchers in ethnophysiography are seeking to understand these factors through examination of case studies. Mark, Turk and Stea are involved in work with the Yindjibarndi people in northwestern Australia and Navajos in southwestern USA (Mark et al. 2003; Mark & Turk 2003b, 2004; Mark et al. 2007, 2010; Turk et al. Chapter 2, this volume). A detailed description of the methodologies used in these ethnophysiography case studies is provided in Turk et al. (in press). As discussed above, researchers at the Max Planck Institute for Psycholinguistics have recently concluded a set of case studies of landscape terms and place names in nine languages in a wide variety of geographic locations (Burenhult 2008a). This work has extended very significantly the range of ethnophysiography-related case studies and strengthened its linguistic basis. In the introduction to that collection of studies, Burenhult and Levinson (2008:â•›1) discuss the theoretical basis of the work and its relationship to ethnophysiography. They review the results of the case studies and state that “The data point to considerable variation within and across languages in how systems of landscape terms and place names



Landscape in language

are ontologised. This has important implications for practical applications from international law to modern navigation systems.” The case studies to date provide strong support for the basic ethnophysiography hypothesis – i.e. that people from different language groups/cultures have different ways of conceptualizing landscape, as evidenced by different terminology and ways of talking about and naming landscape features. However, much more collaborative research on this topic is required. Effective research in this field requires an integrated approach by researchers from many disciplines. Research approaches of increasing levels of integration may be termed: disciplinary; multidisciplinary; interdisciplinary; transdisciplinary; and postdisciplinary (Booth et al. 2000; see also Schunn et al. 2006). The last of these is difficult to achieve other than in individual or small group projects; hence the objective is to develop a transdisciplinary approach to ethnophysiography research. This goal was significantly advanced by the workshop conducted in 2008 that led to this book, and the results of this cross-disciplinary debate permeate the chapters in this volume.

4.

Landscape terms and categories Of all the countless possible ways of dividing entities of the world into categories, why do members of a culture use some groupings and not use others? What is it about the nature of the human mind and the way that it interacts with the nature of the world that gives rise to the categories that are used?  (Malt 1995:â•›85)

Barbara Malt went on from those questions to present an excellent review of research issues regarding categories. The issues raised by Malt frame the terminological aspects of ethnophysiography. For geographic entities and categories, the problem of defining the sets of instances is even more complicated than for domains commonly studied, since to some extent instances of landforms and many other geographic features are themselves contingent on the definitions of the available types. Thus, while a bird might be a robin, but otherwise is still a bird, an elevated area of land might be a “hill” in one speech community but not a coherent entity at all under different landform definitions. Even within a single language, different words may be used to refer to particular landscape features, depending upon frames/scales of reference: e.g. a landform in an area of moderate topographic variation might be spoken of as a “mountain” by locals, but be . The word “ontologised” was intended to refer to the way any particular language uses generic landscape terms as a way of breaking the landscape systematically into parts (forming an ontology of landscape features) and the impact of this on place naming practices. (Note: This footnote was added by the present authors, and was not in Burenhult & Levinson 2008.)

˘

10

David M. Mark et al.

referred to as a “hill” by people from areas of more extreme topography. Some have claimed that the existence and classification of landforms and other landscape elements are not facts of a mind-independent reality, either as objects or as categories (Smith & Mark 1998, 2001, 2003). Thus landscape constitutes an excellent domain for the investigation of whether categorization of natural entities is universal, or whether categorizations differ significantly across languages and cultures. Landscapes vary considerably, making this universally relevant domain particularly problematic and fascinating for research. An understanding of landscape calls for universal attention and maximal representational plasticity at the same time. Related research has been carried out regarding the potential impact of categorization on the design of GIS (Kuhn 2002, 2003, 2005, 2007; Mark 1993; Mark & Smith 2004). There are many contributions in the published literature that are implicitly ‘ethnophysiography,’ although the works either predate the creation of this term, or their authors were not aware of its having been coined, or chose not to use the term. Terms and categories for landscape features have been addressed in general works on linguistics and in ethnoecology. In their classic work on the problem of lexical choice in the Hopi language, Voegelin and Voegelin (1957) introduced the idea of eliciting vocabulary by systematically going through semantic domains. The domain of topography was the very first domain that they discussed in the monograph; other landscape terms were included in the water and vegetation domains. Several books and papers on ethnoecology (Hunn with Selam and family 1990; Taller de Tradición Oral del CEPEC & Beaucage 1996; Beaucage & Taller de Tradición Oral del CEPEC 1997; Johnson 1999, 2000, 2002; Johnson & Hargus 2006) have included brief systematic discussions of ecological features and assemblages at landscape scales. Likewise, some studies in other aspects of cultural anthropology have included accounts of important landscape terms and their referents, including Pinxten et al.’s (1983) exploration of Navajo natural philosophy and landscape in the USA, Myers’ (1986) work on the Pintupi country and culture in Australia, and Kofod’s (2003) report on language, identity, and land in the East Kimberley region of Australia.

5.

Place names

Place names (toponyms) have long fascinated geographers, cartographers, and others, and there are many books about the origins of place names that have been written for a general audience. In many languages, including English, most proper names for geographic features contain a generic part that designates the feature type of the named entity: Mount Washington, Niagara River, and Gulf of



Landscape in language

Mexico, for example. However, this is not the practice in all languages, and the generic parts of place names do not always indicate what kind of thing current residents believe the feature to be. Both the generic and specific parts of place names, if present, may contain clues about the cultural or linguistic history of a place, and may be a source of prototypes for learning the meaning of generic landscape terms. Zelinsky’s (1955) paper on the distribution of generic terms in the place names of the northeastern United States is a classic example of a study connecting place names to cultural history. Zelinsky mapped the generic parts of names for features such as watercourses, and showed that the distributions of terms for smaller watercourses, such as ‘creek,’ ‘brook,’ ‘kill,’ and ‘branch,’ were often localized and reflected the naming practices of different cultural groups. Other authors have followed up this theme, including investigations of Navajo place naming practices (Jett 1997, 2001, 2006; Jett this volume).

6.

Cultural and spiritual significance of places Time will tell what other cultural constructions await the ethnographer bent on an interest in place. But that such constructions are everywhere to be found – in deserts and savannas, mountains and rain forests, cities and towns – is altogether certain. We should begin to explore them with all deliberate speed, and not, I would emphasize, solely for the purpose of enlarging our knowledge of particular social groups. For as surely as place is an elemental existential fact, sense of place is a universal genre of experience, and therefore, as more and more work gets done, it may be found to exhibit transcultural qualities.  (Keith Basso, 1996, Wisdom Sits in Places, pp. 147–148)

The preceding sections of this chapter concentrated implicitly on realist aspects of topography and of cognitive responses. However, for many peoples conceptualization of landscape elements also incorporates what could be termed beliefs, cultural considerations, or spirituality; aspects of what makes a “space” into a “place.” A review of some literature on how lived experience (engagement with social and physical worlds) influences conceptions of place and topophilia (Tuan 1974) is provided in Mark et al. (2010). In their COSIT03 paper, Mark and Turk (2003a) noted that for “Indigenous Australians, including the Yindjibarndi people, spirituality and topography are inseparable (e.g. all yinda have warlu)” (p. 45) – for Yindjibarndi people, each permanent water hole (yinda) has a spirit (warlu) which determines how one should act at that place. Yu (2002) has described similar beliefs and associated practices among the Karajarri people, an indigenous group living several hundred

11

12

David M. Mark et al.

kilometers northeast of the Yindjibarndi. Malpas (1999:â•›95) emphasizes the critical importance of considering the observer (agent) in the process: Understanding an agent, understanding oneself, as engaged in some activity is a matter both of understanding the agent as standing in certain causal and spatial relations to objects and of grasping the agent as having certain attitudes – notably certain relevant beliefs and desires – about the objects concerned.

Mark and Turk (2003a) discussed the difficulty of integrating such aspects of indigenous worldviews with Western Philosophy (in the ‘realist’ tradition) and indicated that further attempts would be made to undertake this task. Andrew Turk has commenced a major study (Turk 2007) of the relationship between these aspects of ethnophysiography and phenomenology (Malpas 2007; Mohanty 1997; Patocka 1996; Schutz 1977; Smith 2007). Chapters in this volume by Turk and Janz explore these issues.

7.

Cultural differences and Geographic Information Systems

The questions raised above are fundamental to many academic disciplines. But now, in the information age, they also have important new practical implications. Are the spatial data infrastructures being compiled by various countries and international agencies around the world biased toward the conceptualizations of the dominant cultures in those countries and agencies? Do the functionalities of Geographic Information Systems (GIS) have similar biases? And if they exist, do such biases broaden the “digital divide” and further marginalize some groups and cultures within the global society? Can GIS facilitate retrieval of information by people from one culture that was collected and stored by another culture? Can culturally sensitive information be preserved in information systems without exposing it to outsiders? What changes in GIS, computer mapping systems, and so on are needed to allow indigenous groups to take full advantages of such technologies without having to step outside or abandon their traditional cultures? Since land is especially important to indigenous peoples and cultures, there has been considerable interest in making GIS technology available for use by indigenous communities and tribal governments. Rundstrom (1995) discussed a number of potential issues regarding the use of conventional GIS for the (mis)representation of traditional geographic knowledge. In the United States, the Bureau of Indian Affairs provided conventional GIS to the tribes in a topdown manner, which may have reinforced the colonial history of “Indian country” by depriving the tribes of control of their own GIS in the early period of GIS adoption (Palmer 2006). This topic has also been of interest to Australian



Landscape in language

researchers, especially in the context of native title land claims (Turk 1996, 2006; Turk & Mackaness 1995). Turk and Trees (1999) reviewed Rundstrom’s concerns in the context of ethical aspects of GIS projects with indigenous Australians. One response to the call for more culturally appropriate systems for information about the landscape was the International Forum on Indigenous Mapping held in Vancouver, British Columbia (Indigenous Mapping, 2004). This forum brought together about 200 people to discuss outcomes from indigenous mapping projects conducted in the United States and Canada and to make plans for future efforts. The technology needed to support such efforts is unclear, and the relative roles of commercial GIS software, mapping systems such as Google Maps and Google Earth, ontology software, or other solutions remain to be seen. The research on ethnophysiography is seen as potentially allowing indigenous concepts to form the basis of GIS data structures (Mark & Turk 2003b, 2004). In a project to implement more culturally appropriate GIS, Renée Sieber and Christopher Wellen of McGill University in Québec, Canada, have been working with a Cree community in northern Québec, to develop an environmental information system for community use that is based on an ontology rather than on commercial GIS software (Sieber 2004, 2006, 2007, in press; Sieber & Wellen 2006, 2007; also Sieber & Wellen this volume). These issues were discussed at the 2008 workshop, described below.

8.

Overview of the remaining chapters

In the second chapter of this book, entitled Ethnophysiography, Andrew Turk, David Mark and David Stea present this new branch of ethnoscience, first introduced by Mark and Turk (2003a). They describe case studies of two language communities, Yindjibarndi and Navajo, providing a cultural, linguistic and environmental background for each group and explaining the methods and results of the studies. The authors go on to sketch an initial ethnophysiographic descriptive model aimed at representing factors of landscape conceptualization as elucidated in these and other case studies. Such a model, it is argued, is a necessary precursor for any causal or predictive model in this undertheorized field. In a chapter entitled Exploring philosophy of place: Potential for synergy between phenomenology and ethnophysiography, Andrew Turk makes the case that an intrinsically transdisciplinary inquiry like ethnophysiography requires an overarching and unifying paradigm. He suggests that this paradigm is best represented by philosophy and, specifically, phenomenology. The chapter describes the relevance of phenomenology to landscape studies and, conversely, explains how landscape can contribute to an enhanced understanding of phenomenology.

13

14

David M. Mark et al.

Furthermore, the author underlines that such a synergy should involve a meaningful dialogue between European philosophy and indigenous cultural concepts of place. In his chapter, Embedded in place: ‘Mirror knowledge’ and ‘simultaneous landscapes’ among Māori, Brian Murton explores the understanding that Māori, the indigenous people of Aotearoa/New Zealand, have of landscape. Landscape is experienced by Māori as the ‘face of place,’ incorporating deep cultural and spiritual associations. The author contends that the approach to landscape in European thought which most closely approximates Māori understanding of the concept is through the use of phenomenology; a ‘dwelling approach’ resonates with Māori metaphysics. He emphasizes the role of traditional beliefs and narratives, linked to genealogies, as the basis for Māori place names, using detailed discussion of particular examples from the region where he himself has close cultural ties. The chapter by Bruce B. Janz, Philosophical issues in ethnophysiography: Landform terms, disciplinarity, and the question of method, discusses the goals of ethnophysiography (as a nascent discipline), its methods, data types and how its approaches to analysis relate to other ways of exploring the nature of ‘place.’ Such investigations must take account of cultural and spiritual aspects of peoples’ relationships with landscape, including: meaningful actions; social structures; and individual interests and desires. The chapter analyzes the potential for phenomenology to combine synergistically with empiricist/positivist approaches within ethnophysiography and the role of formal ontologies in applying the results of ‘place language’ research. In his ‘Land’ and life: Ethnoecology and ethnogeography as complementary approaches to the analyses of landscape perception, Chris Duvall explores the relationship between landscape categories and local knowledge and use of the biophysical environment among the Maninka, subsistence farmers in southwestern Mali, who speak a Niger-Congo language. In particular, he looks at the significance of land-cover categories in local classification of site arability, arguing that studies of local knowledge systems related to subsistence benefit from careful analysis of physiographic concepts. The account includes a detailed description of the elaborate Maninka taxonomy of landforms. Gary Holton’s chapter Landscape in Western Pantar, a Papuan outlier of southern Indonesia provides an in-depth account of the relationship between language and landscape in a little-known language spoken in an extraordinary environment. Western Pantar presents eye-opening examples of what a given community prefers to encode semantically in its landscape categories, notably an unusual strategy of categorizing hydrological features according to their chemistry rather than their size, shape or permanence. The language also raises



Landscape in language

some fundamental questions about the relationship between generic landscape categories and place names. Renee Louis speaks from an indigenous Hawaiian perspective in her chapter Hawaiian storied place names: Re-placing cultural meaning. She discusses sensuous and cultural aspects of traditional Hawaiian place names and their role in narratives and ceremony, using detailed examples. Place names and their cultural meanings are the result of constant interactions of people and their environment and provide a ‘placescape.’ This traditional cartographic perspective is contrasted with the practices of European colonists in terms of a ‘culture clash,’ which led, in part, to indigenous people adopting some Western surveying and mapping techniques. The author discusses recent attempts to develop a ‘third space’ integration of indigenous traditions and Western GIS approaches. In the chapter Between the trees and the tides: Inuit ways of discriminating space in a coastal and boreal landscape, Scott Heyes provides an account of how the Inuit of Kangiqsualujjuaq, a coastal community in Nunavik, Northern Quebec, conceptualize their world of land, sea, and ice, forest and tundra. Extreme tides and high latitude produce strong cyclic variations on daily and annual scales. The people have developed a system of terms and names that allows them to communicate this ‘landscape of variability.’ Narratives of hunters’ journeys reveal aspects of their conceptualization of the coast. Heyes also reports on the past and present roles of spiritual, mythological, and cosmological dimensions in these conceptualizations. In a chapter entitled Differing conceptualizations of the same landscape: The Athabaskan and Eskimo language boundary in Alaska, Gary Holton introduces a comparative language setting with potential to address some of the key questions in ethnophysiography. The same landscape – the Yukon Intermontane Plateau€– is shared by two communities representing unrelated language families and distinct environmental and subsistence histories. Are these differences reflected in how the two communities categorize their common landscape? Or has the nature of the landscape itself caused differing categorial systems to converge? Holton addresses these issues with a focus on terms referring to elevations and water features. James Kari’s chapter Ahtna geographic names: A case study in Athabascan geographic knowledge encapsulates the author’s long-term dedication to understanding and describing the principles of labeling, naming and navigating . Note: The term “Athabaskan” is spelled in various outlets with either a ‘b’ or a ‘p’, and also with either a ‘k’ or a ‘c’. The editors have decided to use “Athabaskan” in the sections that they write, but to respect the spelling choice of the chapter authors, since sometimes the choice of spelling has political implications.

15

16

David M. Mark et al.

landscape among the Northern Athabaskan-speaking communities of Alaska. Arguably one of the most thoroughly researched landscape systems to date, Ahtna geographic classification makes up a well-defined, purposeful and highly economic semantic engine, centered on the flow of water. The chapter explains how this system structures landscape terms, place names, directionals, and geographical narratives, providing a holistic showcase example of how language and landscape can interrelate. In Revitalizing place names through stories and songs, Susan Paskvan draws on her own experience with Koyukon Athabaskan place names. As with other Athabaskan languages, including Navajo as discussed by Jett in this book, Koyukon Athabaskan place names normally are descriptive. Paskvan tells an interesting and useful story of the roles of places and stories in the preservation of language and culture and the transmission of traditional environmental knowledge. She reports on how innovative teaching techniques and technologies can contribute to intergenerational transmission of this knowledge. Thomas Thornton’s chapter, Landscape and language among the Tlingit, tells how the Tlingit people of southeastern Alaska conceptualize their marine and coastal environment. Most of their traditional and modern activities take place in a narrow band along the shore, between the tides, offshore and inland. The land/sea dichotomy provides fundamental structure to their understanding of directions, animals, and spirits. Geographical names cluster along the land/sea interface, indicating the cultural significance of this zone. Thornton discusses the Tlingit conceptualization of landscape and environment in terms of J. J. Gibson’s idea of affordances – day-to-day activities expose affordances and structure their experiences. Leslie Main Johnson’s chapter, Language, landscape and ethnoecology, reflections from northwestern Canada, provides a fascinating description of landscape terms and their meanings in three different languages in adjacent or nearby areas of northern British Columbia. The Gitksan and Witsuwit’en people share similar landscapes but their languages are from different language families. The Witsuwit’en and Kaska people are members of the same language group (Athabaskan), but live in different environments. Comparisons among landscape terms and concepts across these languages provides a basis for untangling the factors that influence which kinds of landscape features have lexicalized terms, and which do not. Language, culture, ways of life, and the nature of the landscape itself all shape a group’s conceptualization of landscape and its components. Stephen C. Jett’s chapter continues his exploration of Landscape Embedded in Language: The Navajo of Canyon de Chelly, Arizona, and their Named Places. The chapter takes on special significance because the majority of the 2008 workshop upon which this book is based was conducted at the mouth of Canyon de Chelly,



Landscape in language

and Jett’s presentation of this material at the workshop set the stage for a field trip into the canyon. Jett places his Canyon de Chelly study in the context of Athabaskan toponyms in general, and points out a number of distinctive aspects, such as the predominance of descriptive names, and rarity of places named after people or holy beings. Currently, place names fill traditional origin narratives, grounding the stories in tangible reality and providing a mutual memory reinforcement between stories and places that served, and continue to serve, a key role in a society that until recently had no written language. In her essay Navajo landscape and its contexts, Carmelita Topaha discusses cultural aspects of landscape from a Navajo perspective. She relates this to her own experience as a young woman visiting significant places with family members and more recent experiences as an archaeologist working on the Navajo Reservation and in nearby areas. She also discusses her involvement in the Ethnophysiography research project with David Mark, David Stea and Andrew Turk, commenting on her role as a facilitator of collaborations with other Navajo participants. Elizabeth Lynch’s chapter, Navigating regional landscapes with Jicarilla personal narrative, investigates how Jicarilla hunter-gatherers may have navigated the landscape, based on analysis of historical narratives. The Jicarilla (an Athabaskan language group) have lived in the southwest of the USA since at least the 15th century. Jicarilla traditional stories were recorded in the early 20th century and the author uses selected stories to construct abstract maps of hunting journeys between named locations, some of which align with current place names. The author suggests that such narrative analysis can complement archaeological data to provide a richer understanding of how these people interacted with the€landscape. Ontology, especially formal ontology, has become popular in information science as a framework that enhances semantic interoperability. In his chapter Ontology of Landscape in Language, Werner Kuhn outlines an ontology-based approach to knowledge representation for research on landscape and language. Ontologies specify vocabularies and associated concepts. Kuhn describes a foundational ontology called DOLCE, which may be especially appropriate for landscape in language because it is has been designed to formalize concepts found in language. Kuhn gives examples of how a formal ontology can accommodate both realist views of physical reality and spiritual dimensions of landscape. Renée Sieber and Chris Wellen write about The role of geospatial technologies for integrating landscape in language, drawing examples from their work with a Cree community in northern Quebec. Rather than employing commercial GIS, Sieber and Wellen used ethnographic methods to elicit the ontology of Cree landscape terms and concepts and represented these concepts in ontology software. The digital ontology provided the backbone of an information system

17

18

David M. Mark et al.

for community management of their land and environment. They found that the Cree ontology of hydrologic features was based primarily on functional relationships between water flow and human travel potential by water. The Cree ontology was more “shallow” topologically than official government ontologies of the same hydrologic systems, that is, do not have as many superordinate categories. Ontology-based environmental information systems may provide a more appropriate basis for indigenous environmental management than other software currently in use. The chapter by Lars Brabyn and David M. Mark, Classifying landscape character, discusses approaches to classification of ‘landscape character’ and how GIS can be used to assist. The complexities of classifying landscape are discussed, since landscape has a number of meanings and can be conceptualized at a range of scales. The authors contend that language plays an important role, especially if the opinions of members of the public are utilized as well as the views of scientists. This leads to discussion of the important role that landscape classification plays in communication and in determining how geographic space itself is conceptualized. The chapter discusses a collaborative approach to improving the New Zealand Landscape Classification system. The book concludes with an edited transcript of a panel discussion held at the workshop, entitled Ethical issues in indigenous landscape in language research: A panel discussion. Statements by workshop participants provide multiple perspectives on the diversity and subtlety of issues involved in landscape and language research, especially in the case of indigenous languages and societies.

Acknowledgments Funding for the workshop was provided by the U.S. National Science Foundation under grant BCS-0753737; their support is greatly appreciated, as is support from the Max Planck Institute for Psycholinguistics in Nijmegen, the Netherlands, and the hospitality of the Navajo Nation.

References Basso, Keith H. 1996. Wisdom Sits in Places: Landscape and Language Among the Western Apache. Albuquerque NM: University of New Mexico Press. Beaucage, Pierre & Taller de Tradición Oral del CEPEC. 1997. Integrating innovation; traditional Nahua coffee-orchard (Sierra Norte de Puebla, Mexico). Journal of Ethnobiology 17(1): 45–67.



Landscape in language

Bell, Paul, Greene, Thomas, Fisher, Jeffrey & Baum, Andrew. 1996. Environmental Psychology. Fort Worth TX: Harcourt Brace. Bender, Barbara (ed.). 1993. Landscape: Politics and Perspectives. Providence RI: Berg. Bohnemeyer, Jürgen, Burenhult, Niclas, Enfield, Nick J. & Levinson, Stephen C. 2004. Landscape terms and place names elicitation guide. In Field Manual Volume 9, Asifa Majid (ed.), 75–79. Nijmegen: Max Planck Institute for Psycholinguistics. Booth, Michael, Rodgers, Stephen & AgInsight. 2000. Interdisciplinary Research Methodologies in Natural Resource Management [Report to LWRRDS Social and Institutional Research Program]. Institute for Sustainability and Technology Policy, Murdoch University, Western Australia . Brown, Penelope. 2008. Up, down, and across the land: Landscape terms, place names, and spatial language in Tzeltal. Language Sciences 30(2/3): 151–181. Burenhult, Niclas (ed.). 2008a. Language and Landscape: Geographical Ontology in Cross-linguistic Perspective [Language Sciences 30(2/3)]. Amsterdam: Elsevier. Burenhult, Niclas. 2008b. Streams of words: Hydrological lexicon in Jahai. Language Sciences 30(2/3): 182–199. Burenhult, Niclas & Levinson, Stephen C. 2008. Language and landscape: a cross-linguistic perspective. Language Sciences 30(2/3): 135–150. Burenhult, Niclas & Levinson, Stephen C. 2009. Semplates: A guide to identification and elicitation. In Field Manual Volume 12, Asifa Majid (ed.), 44–50. Nijmegen: Max Planck Institute for Psycholinguistics. Cablitz, Gabriele H. 2008. When “what” is “where”: A linguistic analysis of landscape terms, place names and body part terms in Marquesan (Oceanic, French Polynesia). Language Sciences 30(2/3): 200–226. Casey, Edward S. 1996. How to get from space to place in a fairly short stretch of time: Phenomenological prolegomena. In Senses of Place, Steven Feld & Keith H. Basso (eds), 13–52. Santa Fe NM: School of American Research Press. Cosgrove, Denis. 2004. Landscape and Landschaft. “Spatial Turn in History” Symposium, German Historical Institute, lecture delivered February 19, 2004. GHI Bulletin No. 35 (Fall 2004) 57–71. Downs, Roger M. & Stea, David. 1977. Maps in Minds: Reflections on Cognitive Mapping. New York NY: Harper and Row. Enfield, Nick J. 2008. Linguistic categories and their utilities: The case of Lao landscape terms. Language Sciences 30(2/3): 227–255. Feld, Steven. 1996. Waterfalls of song: an acoustemology of place resounding in Bosavi, Papua New Guinea. In Senses of Place, Steven Feld & Keith H. Basso (eds), 91–135. Santa Fe NM: School of American Research Press. Gibson, James J. 1979. The Ecological Approach to Visual Perception. Boston MA: Houghton Mifflin. Gibson, James J. 1987. The perception of visual surfaces. American Journal of Psychology 100(3/4): 646–664. Gibson, James J. & Bridgeman, Bruce. 1987. The visual perception of surface texture in photographs. Psychological Research 49(1): 1–5. Granö, Johannes Gabriel. 1997. Pure Geography. Baltimore MD: Johns Hopkins University Press. Hirsch, Eric & O’Hanlon, Michael (eds). 1995. The Anthropology of Landscape: Perspectives on Place and Space. Oxford: Clarendon Press.

19

20 David M. Mark et al.

Hunn, Eugene S., with Selam, James, and family. 1990. Neh’I-Wana “The Big River.” Mid-Columbia Indians and Their Land. Seattle WA: University of Washington Press. Indigenous Mapping. 2004. The International Forum on Indigenous Mapping: Mapping for Indigenous Advocacy and Empowerment, held in Vancouver, BC, Canada, March 11–14, 2004. Ingold, Tim. 2000. The Perception of the Environment: Essays on Livelihood, Dwelling and Skill. London: Routledge. Ingold, Tim. 2010. The shape of the land. Epilogue in Landscapes Beyond Land: Routes, Aesthetics, Narratives [European Association of Social Anthropologists (EASA) Series], J. Vergunst, A. Árnason, N. Ellison & A. Whitehouse (eds). Oxford: Berghahn. Jett, Stephen C. 1997. Place-naming, environment, and perception among the Canyon de Chelly Navajo of Arizona. Professional Geographer 49(4): 481–493. Jett, Stephen C. 2001. Navajo Placenames and Trails of the Canyon de Chelly System, Arizona. Frankfurt: Peter Lang. Jett, Stephen C. 2006. Reconstructing the itineraries of Navajo chantway stories: A trial at Canyon de Chelly, Arizona. In Southwestern Interludes: Papers in Honor of Charlotte J. and Theodore R. Frisbie, Regge N. Wiseman, Thomas C. O’Laughlin & Cordelia T. Snow (eds), 75–86. Albuquerque NM: The Archaeological Society of New Mexico. Johnson, Leslie M. 1999. Aboriginal burning for vegetation management in northwest British Columbia. In Indians, Fire and the Land in the Pacific Northwest, Robert Boyd (ed.), 238–254. Corvallis OR: Oregon State University Press. Johnson, Leslie M. 2000. “A place that’s good,” Gitksan landscape perception and ethnoecology. Human Ecology 28(2): 301–325. Johnson, Leslie M. 2002. Chapter 2: Indigenous knowledge as a basis for living in local environments. In Ethnographic Essays in Cultural Anthropology: A Problem Based Approach, R.€Bruce Morrison & C. Roderick Wilson (eds), 28–49. Itasca IL: F.E. Peacock. Johnson, Leslie M. & Hargus, Sharon. 2006. Witsuwit’en words for the land – a preliminary examination of Witsuwit’en ethnogeography. In ANLC Working Papers in Athabaskan Linguistics, Vol. 6, Siri Tuttle, Leslie Saxon, Suzanne Gessner & Andrea Berez (eds). Fairbanks AK: Alaska Native Language Center. Kofod, Frances. 2003. My relations, my country: language, identity and land in the East Kimberley of Western Australia. In Maintaining the Links: Language, Identity, and the Land [Proceedings of the Seventh Conference of the Foundation for Endangered Languages, Broome, Western Australia, 22–24 September 2004], Joe Blythe & R. McKenna Brown (eds), 41–47. Bath: Foundation for Endangered Languages. Kuhn, Werner. 2002. Modeling the semantics of geographic categories through conceptual integration. In Geographic Information Science – Second International Conference (GIScience 2002) [Lecture Notes in Computer Science 2478], Max J. Egenhofer & David M. Mark (eds), 108–118. Berlin: Springer. Kuhn, Werner. 2003. Semantic reference systems. International Journal of Geographical Information Science 17(5): 405–409. Kuhn, Werner. 2005. Geospatial semantics: Why, of what, and how? In Journal on Data Semantics III [Lecture Notes in Computer Science 3534], Stefano Spaccapietra & Esteban Zimányi (eds), 1–24. Berlin: Springer. Kuhn, Werner. 2007. An image-schematic account of spatial categories. In Proceedings of the 8th International Conference on Spatial Information Theory [Lecture Notes in Computer Science 4736], Stephan Winter, Matt Duckham, Lars Kulik & Benjamin Kuipers (eds), 152–168. Berlin: Springer.



Landscape in language

Levinson, Stephen C. 2003. Space in Language and Cognition: Explorations in Cognitive Diversity. Cambridge: CUP. Levinson, Stephen C. 2008. Landscape, seascape and the ontology of places on Rossel Island, Papua New Guinea. Language Sciences 30(2/3): 256–290. Levinson, Stephen C. & Burenhult, Niclas. 2009. Semplates: A new concept in lexical semantics? Language 85(1): 153–174. Levinson, Stephen & Wilkins, David (eds). 2006. Grammars of Space. Cambridge: CUP. Lyons, John. 1977. Semantics. Vols. I & II. Cambridge: CUP. Majid, Asifa, Bowerman, Melissa, Kita, Sotaro, Haun, Daniel & Levinson, Stephen. 2004. Can language restructure cognition? The case for space. Trends in Cognitive Sciences 8(3): 108– 114. Malpas, Jeff E. 1999. Place and Experience: A Philosophical Topography. Cambridge: CUP. Malpas, Jeff E. 2007. Heidegger’s Topology: Being, Place, World. Cambridge MA: The MIT Press. Malt, Barbara C. 1995. Category coherence in cross-cultural perspective. Cognitive Psychology 29(2): 85–148. Mark, David M. 1993. Toward a theoretical framework for geographic entity types. In Spatial Information Theory: A Theoretical Basis for GIS [Lecture Notes in Computer Sciences No. 716], Andrew U. Frank & Irene Campari (eds), 270–283. Berlin: Springer. Mark, David M., Kuhn, Werner, Smith, Barry & Turk, Andrew G. 2003. Ontology, natural language, and information systems: implications of cross-linguistic studies of geographic terms. In AGILE 2003 – 6th AGILE Conference on Geographic Information Science, Michael Gould, Robert Laurini & Stéphane Coulondre (eds), 45–50. Lyon: Presses Polytechniques et Universitaires Romandes. Mark, David M. & Smith, Barry. 2004. A science of topography: From qualitative ontology to digital representations. In Geographic Information Science and Mountain Geomorphology, Michael P. Bishop & John F. Shroder (eds), 75–100. Berlin: Springer-Praxis. Mark, David M., Smith, Barry & Tversky, Barbara. 1999. Ontology and geographic objects: An empirical study of cognitive categorization. In Spatial Information Theory: A Theoretical Basis for GIS [Lecture Notes in Computer Science 1661], Christian Freksa & David M. Mark (eds), 283–298. Berlin: Springer. Mark, David M. & Turk, Andrew G. 2003a. Ethnophysiography. Paper presented at the Workshop on Spatial and Geographic Ontologies, 23 September, 2003 (prior to COSIT03), in Ittingen, Switzerland. Mark, David M. & Turk, Andrew G. 2003b. Landscape categories in Yindjibarndi: ontology, environment, and language. In Spatial Information Theory: Foundations of Geographic Information Science [Lecture Notes in Computer Science 2825], Werner Kuhn, Michael Worboys & Sabine Timpf (eds), 28–45. Berlin: Springer. Mark, David M. & Turk, Andrew G. 2004. Ethnophysiography and the ontology of the landscape. In GIScience 2004 Extended Abstracts and Poster Summaries, Max Egenhofer, Christian Freksa & Harvey Miller (eds), 152–155. Santa Barbara CA: Regents of the University of California. Mark, David M., Turk, Andrew G. & Stea, David. 2007. Progress on Yindjibarndi Ethnophysiography. In Proceedings of the 8th International Conference on Spatial Information Theory [Lecture Notes in Computer Science No. 4736], Stephan Winter, Matt Duckham, Lars Kulik & Benjamin Kuipers (eds), 1–19. Berlin: Springer.

21

22

David M. Mark et al.

Mark, David M., Turk, Andrew G. & Stea, David. 2010. Ethnophysiography of arid lands: Categories for landscape features. In Landscape Ethnoecology, Concepts of Biotic Physical and Space, Leslie M. Johnson & Eugene S. Hunn (eds), 27–45. New York NY: Berghahn Books. Mohanty, Jitendranath N. 1997. Phenomenology: Between Essentialism and Transcendental Philosophy. Evanston IL: Northwestern University Press. Myers, Fred R. 1986. Pintupi Country, Pintupi Self: Sentiment, Place, and Politics among Western Desert Aborigines. Berkeley CA: University of California Press. O’Connor, Loretta & Kroefges, Peter K. 2008. The land remembers: Landscape terms and place names in Lowland Chontal of Oaxaca, Mexico. Language Sciences 30(2/3): 291–315. Olwig, Kenneth. 2008. Performing on landscape versus doing landscape: perambulatory practice, sight and the sense of belonging. In Ways of walking: Ethnography and practice on foot, Tim Ingold & J. Lee Vergunst (eds), 81–91. Aldershot: Ashgate. O’Meara, Carolyn & Bohnemeyer, Jürgen. 2008. Complex landscape terms in Seri. Language Sciences 30(2/3): 316–339. Palmer, Mark H. 2006. Creating Indigital Peripheries: The Bureau of Indian Affairs, Geographic Information Systems, and the Digitization of Indian Country. PhD dissertation, University of Oklahoma. Patocka, Jan. 1996. An Introduction to Husserl’s Phenomenology. Peru IL: Open Court Publishing. Pattison, William D. 1964. The four traditions of geography. Journal of Geography 63(5): 211–216. Pinxten, Rik, Van Dooren, Ingrid & Harvey, Frank. 1983. Anthropology of Space: Explorations into the Natural Philosophy and Semantics of the Navajo. Philadelphia PA: University of Pennsylvania Press. Reed, Edward & Jones, Rebecca. 1982. Reasons for Realism: Selected Essays of James J. Gibson. Hillsdale NJ: Lawrence Erlbaum Associates. Rundstrom, Robert A. 1995. GIS, indigenous peoples, and epistemological diversity. Cartography and Geographic Information Systems 22(1): 45–57. Schunn, Christian D., Paulus, Paul B., Cagan, Jonathan & Wood, Kristin. 2006. Final Report from the NSF Innovation and Discovery Workshop: The Scientific Basis of Individual and Team Innovation and Discovery. August 2006. (23 August 2007). Schutz, Alfred. 1977. Husserl and his influence on me. In Interdisciplinary Phenomenology, Don Ihde & Richard M. Zaner (eds), 124–129. The Hague: Martinus Nijhoff. Senft, Gunter. 2008. Landscape terms and place names in the Trobriand Islands – the Kaile’una subset. Language Sciences 30(2/3): 340–361. Sieber, Renée E. 2004. Rewiring for a GIS/2. Cartographica 39(1): 25–39. Sieber, Renée E. 2006. Public participation Geographic Information Systems: A literature review and framework. Annals of the American Association of Geographers 96(3): 491–507. Sieber, Renée E. 2007. Spatial data access by the grassroots. Cartography and Geographic Information Science 34(1): 47–62. Sieber, Renée E. In press. The nature of GIS: The social implications of using GIS for the environment. In Geographic Information Systems and Society Handbook, Timothy Nyerges, Helen Couclelis & Robert McMaster (eds). London: Sage.



Landscape in language

Sieber, Renée E. & Wellen, Christopher C. 2006. Participatory indigenous spatial ontology. Paper presented at the 4th GIScience Biennial Conference, September 22, 2006, Munster, Germany. Sieber, Renée E. & Wellen, Christoper C. 2007. Blending participatory GIS and geo-spatial ontologies for indigenous knowledge preservation. Keynote speech at SAGEO/CQFDGéo: The International French Geomatics Conference, June 20, 2007, Clermont-Ferrand, France. Smith, Barry & Mark, David M. 1998. Ontology and geographic kinds. In Proceedings of 8th International Symposium on Spatial Data Handling [SDH ‘98], K. Tom Poiker & Nicholas Chrisman (eds), 308–320. Vancouver BC: International Geographical Union. Smith, Barry & Mark, David M. 2001. Geographic categories: An ontological investigation. International Journal of Geographical Information Science 15(7): 591–612. Smith, Barry & Mark, David M. 2003. Do mountains exist? Towards an ontology of landforms. Environment and Planning B: Planning and Design 30(3): 411–427. Smith, David W. 2007. Phenomenology. Stanford Encyclopaedia of Philosophy, (20 April 2007). Smith, Neil. 1989. Geography as museum: Private history and conservative idealism in The Nature of Geography. In Reflections on Richard Hartshorne’s The Nature of Geography, J.€Nicholas Entrikin & Stanley D. Brunn (eds), 91–120. Washington DC: Occasional Publication of the Association of American Geographers. Taller de Tradición Oral del CEPEC & Beaucage, Pierre. 1996. La bonne montagne et l’eau malfaisante: Toponymie et practiques environmentales chez les Nahuas de basse montagne (Sierra Norte de Puebla, Mexique). Anthropologie et Sociétés 20(3): 33–54. Tilley, Christopher. 1994. A Phenomenology of Landscape: Places, Paths and Monuments. Oxford: Berg. Tuan, Yi-Fu. 1974. Topophilia: A Study of Environmental Perception, Attitudes, and Values. Englewood Cliffs NJ: Prentice Hall. Turk, Andrew G. 1996. Presenting aboriginal knowledge: Using technology to progress native title claims. Alternative Law Journal 21(1): 6–9. Turk, Andrew G. 2006. Representations of tribal boundaries of Australian Indigenous peoples and the implications for Geographic Information Systems. In Information Technology and Indigenous People, Laurel E. Dyson, Max Hendricks & Stephen Grant (eds), 232–244. Hershey PA: Information Science Publishing. Turk, Andrew G. 2007. A phenomenology basis for trans-disciplinary research in ethnophysiography. Proceedings of the 26th International Human Science Research Conference, Rovereto, Italy. CD-ROM. Turk, Andrew G. & Mackaness, William A. 1995. Design considerations for spatial information systems and maps to support native title negotiation and arbitration. Cartography 24(2): 17–28. Turk, Andrew, Mark, David, O’Meara, Carolyn & Stea, David. In press. Geography – documenting terms for landscape features. In Oxford Handbook of Linguistic Fieldwork, Nicholas Thieberger (ed.). Oxford: OUP. Turk, Andrew G. & Trees, Kathryn A. 1999. Ethical issues concerning the use of Geographic Information Systems technology with indigenous communities. Proceedings of the Australian Institute of Computer Ethics Conference (AICEC99), Melbourne, Australia, 385–398.

23

24

David M. Mark et al.

Voegelin, Carl F. & Voegelin, Florence M. 1957. Hopi Domains: A Lexical Approach to the Problem of Selection [Memoir 14 of the International Journal of American Linguistics]. Baltimore MD: Waverly Press. Widlok, Thomas. 2008. Landscape unbounded: Space, place, and orientation in ≠Akhoe Hai//om and beyond. Language Sciences 30(2/3): 362–380. Yu, Sarah. 2002. Ngapa Kunangkul (Living Water): An indigenous view of groundwater. In Country: Visions of Land and People in Western Australia, Andrea Gaynor, Mathew Trinca & Anna Haebich (eds), 33–55. Perth: Western Australian Museum. Zelinsky, Wilbur. 1955. Some problems in the distribution of generic terms in the place-names of the Northeastern United States. Annals of the Association of American Geographers 45: 319–349.

chapter 2

Ethnophysiography Andrew G. Turk, David M. Mark and David Stea

This chapter provides an overview of recent progress in the research field of ethnophysiography. It provides a summary of two case studies, one with the Yindjibarndi people from northwestern Australia, and the other with the Diné from southwestern United States of America. The main findings to date from these studies are that most basic terms for landscape features in each of these languages do not have a one-to-one equivalence with any landscape term in English. The findings point to key research issues in the way in which landscape is referred to in different languages. To facilitate this transdisciplinary research, an initial descriptive model is presented, which includes key observed differences in the way languages treat landscape and also a set of factors which might be considered as possible reasons for such differences.

1.

Introduction

This chapter briefly describes developments in ethnophysiography, defined as the study of different human conceptualizations of landscape, especially as indicated by differences in the way languages use generic terms and proper names (toponyms) for landscape features. The field of ethnophysiography was initiated in 2002 by David Mark and Andrew Turk (Mark & Turk 2003a); David Stea joined the research group two years later. More complete descriptions of the motivations, methodology and findings of this research project are provided in: Mark and Turk (2003a, b); Mark, Turk and Stea (2007, 2010); Turk, Mark, O’Meara and Stea (in press); and in Chapters 1 and 3 of this book. The first ethnophysiography case study, with the Yindjibarndi people from northwestern Australia, has been underway since 2002. Land is absolutely central to Yindjibarndi culture. Hence, terms for parts of landscape play an important role in their language. Terms for landscape features relate to practical activities of Yindjibarndi life and also to deep cultural/spiritual associations with ngurra

26 Andrew G. Turk, David M. Mark and David Stea

(country) from the earliest time, Ngurra Nyujunggamu ‘when the world was soft’ (Ieramugadu Group Inc. 1995). The second ethnophysiography case study conducted by the authors involves the Diné. The majority of the Diné, known as Navajo from the time of Spanish settlement, currently live on a reservation half the size of England, situated in the southwestern United States of America. The main reason for choosing the Navajo language for the second case study of the Ethnophysiography Project was the superficial resemblance of Navajo country, semi-arid with many exposed rock formations, to Yindjibarndi landscapes. This holds promise for comparing sets of terms for similar landscape features between languages of vastly different origins. In addition to the ethnophysiography case studies described in this chapter, the authors have also reviewed a number of similar studies, especially those carried out by the Language and Cognition Group at the Max Planck Institute for Psycholinguistics, Nijmegen (Burenhult 2008a). This work, which is currently seen as fitting predominately within the disciplines of linguistics and geography, could also fall under such other rubrics as philosophical or environmental anthropology. A way to improve understanding of such diverse contributions is to construct a descriptive model of factors relevant to ethnophysiography. Attributes of places and peoples differ in various ways. The case studies reviewed indicate a number of common ways that languages also seem to vary in how they refer to landscape. The former factors could be possible reasons for the latter linguistic effects, although they are unlikely to be related via a deterministic causal model. Indeed, the reasons any specific language has particular concepts and terms may be lost in the mists of time. An initial version of the Ethnophysiography Descriptive Model is developed in the fourth section of this chapter, which was presented by Turk as a paper at the Landscape in Language Workshop.

2.

Yindjibarndi case study

2.1

Description of language community

Language and location Yindjibarndi belongs to the Coastal Ngayarda language group, within the southwest group of the Pama-Nyungan languages (SIL 2000). According to Thieberger (1996), about 1,000 Yindjibarndi speakers remain, of whom about 500 live in and near Roebourne in the Pilbara region of northwestern Australia. The population of the Roebourne community is mostly indigenous and people use both



Ethnophysiography

their own languages and English to differing degrees, depending on the context, sometimes mixing words from different languages in a single utterance. Several linguists have studied the Yindjibarndi language and partial dictionaries have been produced: Wordick (1982); Anderson (1986); and a digital version by Anderson and Thieberger (n.d.). The Yindjibarndi language, like other Pilbara languages, relies mainly on nouns and verbs (Wordick 1982). Almost 70 percent of the words listed in the aforementioned dictionaries are nouns and 20 percent verbs. Yindjibarndi has no word class equivalent to adjectives per se. However, nouns are used to modify other nouns in ways somewhat equivalent to adjectives in English. Word order is relatively flexible, and the roles of words in utterances often are indicated by an elaborate set of suffixes, including two indicating location (-ngga, â•‚la) and another indicating a particular instance of a generic term or name (-na). Yindjibarndi does not make any formal distinction between mass and count nouns.

Brief description of landscape and ecology Traditional Yindjibarndi country starts at an escarpment about 50 km from the sea and extends on tablelands for about 200 km to the east, through the broad valley of what is called the Fortescue River in English, and for 50 km south up to the Hamersley Ranges. The relief (elevation differences) is mostly relatively low, with rolling hills, extensive flats, and some cliffs and rock outcrops. The climate of this region is quite hot, sometimes reaching over 50°C (122°F) in summer, and relatively dry, with average annual rainfall of about 300 mm (12 inches). There are no permanent or even seasonal rivers or creeks in Yindjibarndi country. Larger watercourses have running water only after major rainfall events, usually associated with cyclones (hurricanes). Permanent pools occur where the lie of the land and the geology cause the water table to break the surface of the ground, e.g. along the channels of the Fortescue River. There are also some small permanent springs and soaks (where water can be obtained by digging) but no significant intermittent or seasonal lakes in Yindjibarndi country. The vegetation is mostly sparse grass or spinifex (Triodia), low scrub and scattered eucalypt trees, except in the floodplain of the Fortescue River, where vegetation is more dense. Brief description of lifestyle and history of language community Until the 1860s the Yindjibarndi people lived mostly along the middle part of the valley of what Europeans named the Fortescue River and on adjacent uplands, pursuing a semi-nomadic hunter/gatherer lifestyle, living in small groups (Tindale 1974). However, they have been progressively displaced from their ‘country’ over the last 150 years as part of the colonizing process, having been obliged

27

28

Andrew G. Turk, David M. Mark and David Stea

to make way for, initially, pastoral activities and, in more recent times, the mining industry (Ieramugadu Group Inc. 1995; Rijavec, Harrison & Soloman 1995). Almost all Yindjibarndi people now live outside their traditional territory, with about half living in the nearby small town of Roebourne (Ireamugadu), near the coast, together with Ngaluma and Banjima peoples. Many Yindjibarndi retain strong cultural links to the land, visiting their traditional country for hunting, gathering and ceremonies.

2.2

Case study description

Participants This language group was chosen for study partly because of a long history of research and community development projects undertaken in Roebourne by Turk and his partner Kathryn Trees (e.g. Trees & Turk 1998; Turk & Trees 2000). The language participants (listed in the acknowledgments section) are all members of the Roebourne community. Some of these people have, sadly, passed away but the researchers have been requested to continue to list their contributions, despite a tradition of not using the names of deceased persons. The Yindjibarndi cultural organization Juluwarlu and the Pilbara Aboriginal Language Centre (Wangka Maya) also provided invaluable assistance. Brief description of case study methodology The methodology used in this case study consisted of the following five major stages: 1. Dictionary work and photo collection; scoping the domain and preparing fieldwork materials; 2. Photo interpretation sessions with language consultants; transcription; clarifying landscape terms identified from dictionaries and collecting new terms; 3. Field interviews; travelling through Yindjibarndi-related country and identifying the set of landscape terms and distinctions among them, cultural associations and usage; 4. Semi-structured follow-up sessions with selected photos; clarifying confusions, probing for extra meanings and evaluating the quality of interpretations; 5. Reporting back; presentation of the initial results to community members (as a draft pictorial landscape dictionary) and obtaining feedback.



2.3

Ethnophysiography

Main findings

Summary of main landscape terms A draft photo-illustrated dictionary including about 100 Yindjibarndi landscape terms has been compiled (Turk & Mark 2008). This version was discussed with participants in 2009 and has been revised, in collaboration with Juluwarlu. Table€1 presents the meanings of the most interesting and important of the terms collected. Table 1.╇ Selected Yindjibarndi landscape terms and explanations Yindjibarndi term

English words with similar meaning

bargarra birlin

Convexities / eminences mound, heap, pile small rough hill or steep, rocky hillside small, low, smoothly-rounded hill summit, peak (of hill/mountain) bottom (foot) (of hill/mountain) flat-topped (e.g. hill) hill, mountain, ridge, range, rock, stone Horizontal areas (plains) flat (area) or plain flat rock outcrop

ganyjiya yirra

Edges edge of hilltop; edge of water body exact edge of hilltop or water body

barlu burnda garga wundu yalimbirr yarndirr

Longitudinal depressions bank (of river, creek or pool) depression, ditch, hollow gully riverbed steep, high riverbank gorge

bawa jinbi mankurdu thanardi thardarr thurla yijirdi yinda yurrama

Water and water features water spring strong flow, flood (of water) sea, ocean waterfall place small, temporary pool (of water) slow, shallow flow (of water) permanent pool (of water) soak

bantha bargu burbaa gankarni jinangga marlirri marnda

29

30

Andrew G. Turk, David M. Mark and David Stea

Table 1.╇ (continued) Yindjibarndi term

English words with similar meaning

jardungu muji ngathiny wirga

Openings and gaps rock-hole cave, hole space or gap between rocks gap (in ridge line or between hills)

biyiila bura ngurra thalu thunggawaya wana wankurri wardiwardi

Areas and locations in the open “out bush” (i.e., far away from settlements) earth, ground, place, camp, land, country sacred site sandy area ground in the middle distance corner (in cliff) slope

barurru garrarnmarra

Land cover (vegetation) spinifex, area of thick vegetation, area of

mardiya (resp.) yirdiya

Tracks road, track, trail (respect language) road, track, trail

Key aspects of interest The way that Yindjibarndi terms subdivide and represent landscape features differs considerably from terms used in the English language. Key aspects of these differences are: – There are no terms for types of landscape features not found in traditional Yindjibarndi country, except for thanardi (ocean, sea). – One term marnda covers a wide range of features of different sizes: things that would be called ‘hill,’ ‘mountain,’ ‘ridge,’ ‘range,’ ‘rock,’ ‘stone,’ etc. in English. – Terms for convex landscape features do not match those in English: marnda, bargu, burbaa do not equate to mountain, hill, rise; the last two are smaller but mainly differentiated by shape. – Yindjibarndi uses some compound phrases for landscape features denoted by single words in English; e.g. marnda marlirri (flat-topped hill, mesa). – Yindjibarndi hydrology terms separate the water and the magnitude of its flow (bawa, mankurdu, yijirdi) from longitudinal depressions (garga, wundu)



–

–

– –

Ethnophysiography

along which the water sometimes flows, whereas English incorporates both the water and the channel in the terms ‘river’ and ‘creek,’ with the water and the bed probably being considered to be parts of one entity. Similarly, thardarr is the term for the place where water sometimes falls down a cliff; if the falling water is present it is referred to by the flow magnitude words above. Some Yindjibarndi terms effectively refer to shape rather than topographic objects; e.g. burbaa, which can refer to a small marnda that is low, smooth and rounded, or to a rise in a yirdiya (road), or to low rounded areas of higher ground, for instance between garga (gullies). There are alternative words for some landscape features in a special ‘respect’ language (used in major ceremonies involving visitors from other tribes); e.g. mardiya (track, trail). Spiritual aspects of features are part of the meaning of terms; e.g. yinda (permanent pool with warlu spirit). Place names tend not to include generic terms, are unanalyzable, rather than descriptive, and are sometimes grouped, especially when they are related by ‘the dreaming’ (creation beliefs in the form of narratives); e.g. Jindawurru.

3.

Navajo case study

3.1

Description of language community

The Diné (Navajo) people In their own language, the people refer to themselves as Diné, but they normally use the term Navajo in English-language contexts. In the 2000 U.S. census, 298,197 people reported that they were of Navajo ancestry. With about 240,000 officially registered members, Navajos are second in numbers only to the Cherokees among U.S. Indian groups. The 2000 U.S. census also reported a total of 178,014 people living in Navajo-speaking households. The Navajo language is the most frequently spoken indigenous language in the United States; 47 percent of those who speak an Indian language at home in the United States speak Navajo. The Navajo Reservation consists of 67,339 square kilometers and is located in northeastern Arizona, northwestern New Mexico, and southeastern Utah. The 2000 U.S. census reported that 58% of Navajos (173,987 people) lived on the reservation. Other Navajos live in such cities as Phoenix, Arizona, and Los Angeles, California, and the U.S. census reported Navajo speakers living in 47 of the 50 U.S. states.

31

32

Andrew G. Turk, David M. Mark and David Stea

The Navajo language The Navajo language, which the people call Diné Bizaad, is a member of the Athabaskan language family (Young & Morgan 1992). Most of the Athabaskan languages are spoken today in Alaska and northwestern Canada, but those spoken in the western United States include Navajo, various Apachean languages, and several others along the Pacific Coast. Like other Athabaskan languages, Navajo has an elaborate prefixing system with words formed by fitting elements into templates, with many words combining several components (Kari 1989). The Navajo language has a large number of verbs, relatively few nouns, and no word class that corresponds to adjectives in English. Language community history and lifeways, past and present The Navajo origin stories place their beginnings in the area where they now live, bounded by their four Sacred Mountains. Research in non-Navajo archaeology and anthropology, however, suggests that people with a material culture associated with Athabaskan arrived relatively recently in the American Southwest. While sources vary regarding the estimated time of arrival of these peoples in New Mexico, they almost certainly moved into the area between 1100 and 1400€A.D. (see Iverson with Roessel 2002:â•›16). When they arrived in northern New Mexico, the people now known as Navajos apparently led a hunter-gatherer lifestyle, perhaps complemented by raiding. However, they soon adopted corngrowing from the Puebloan peoples living in the area, and began cultivating fruit trees. When Spaniards introduced horses and sheep, various tribes adopted them. Both horses and sheep became central to Navajo culture and were incorporated into origin stories. The majority of the Navajo people were forcibly removed from their land by the United States government in 1864, but after four and a half years of exile on the plains of northeastern New Mexico, they were allowed to return to their traditional lands, some of which were designated as a reservation straddling what is now the Arizona-New Mexico border. They promptly expanded into neighboring lands and have continued to gain considerable land up to the present. Navajos are traditionally matrilineal and matrilocal, i.e., people’s primary clan affiliation is with their mother’s clan, and daughters traditionally settle near their mothers’ homes. Today, many Navajos live on the reservation in dispersed settlements, and some herd sheep, while many others, wage-earners living in towns on and off the reservation, return home on weekends. For a more detailed history of the Navajo people, see Iverson with Roessel (2002).



Ethnophysiography

The physical landscape and ecology Navajo land, ranging from semi-arid to arid, is forested at high elevations. The physiography is basin-and-range, and much of the Navajo Reservation is on the Colorado plateau. In many areas, flat-lying rocks and volcanic necks form features that in English would be called buttes, mesas, and plateaus. While vegetation in lower areas consists of sagebrush (Artemisia tridentata) and other shrubs and grasses, there are junipers (Juniperus monosperma and J. osteosperma), piñon pines (Pinus edulis and P. monophylla) and ponderosa pines (Pinus ponderosa) at intermediate elevations. A few permanent rivers and streams cross through or bound Navajo country after originating outside the area, but most watercourses are dry for most of the year. 3.2

Case study description

Brief methodological description Following preliminary examination of published dictionaries and other literature, the first exploration of Navajo landscape terminology involved having Navajo speakers sort landscape photographs into groups; participants were then asked to describe why they grouped the photos as they did. Groupings differed widely from speaker to speaker and were difficult to interpret. However, the results of the grouping task did help the researchers in their selection of images for the photograph response protocol (see below). This exploratory phase was followed by a second phase using two main research protocols. During field interviews, individual participants or small groups rode in vehicles with the researchers along routes chosen by the participants in areas familiar to them, and talked about what they saw in the landscape. Some participants spoke mainly English to the researchers and simply included Navajo landscape terms. Other participants talked about the landscape in complete Navajo sentences. The conversations were recorded on digital audio media; numerous photographs were taken and GPS coordinates were collected along most of these routes. Field interviews were conducted at sites distributed across the Navajo Reservation. A third protocol involved showing landscape photographs to small groups of participants (with a mean of about 3 people), recruited by local Navajo consultant Carmelita Topaha. Participants talked about what they saw in the photographs, and the sessions were again recorded on digital audio media. For some sessions, photographs were projected using a digital projector; elsewhere, page-sized color prints of the same photographs were passed around among the participants as

33

34

Andrew G. Turk, David M. Mark and David Stea

they sat at a table. The photographs were selected by the researchers to represent various “types of landforms” (etic categories) and different regions of the reservation and were always shown in the same order. The set of photographs was adjusted in later fieldwork to better elicit terms about which the researchers were less certain. Fieldwork was conducted between July 2003 and April 2009, and, as of early 2010, is in the analysis stage. Participants were paid for their participation in accord with locally appropriate practice and the research ethics approvals, including a research permit issued by the Navajo Nation.

3.3

Main findings

As of early 2010, analysis of data had begun but only preliminary results had been obtained. Thus, presentation of lists of landscape terms and definitions here would be premature. There are, however, some more general findings.

Key aspects of landscape terms The Navajo language has a large number of terms that refer to features or parts of the landscape. A draft list of terms for an illustrated landscape dictionary for community use, prepared in 2009, includes about 175 terms, which can be placed under 47 main entries. In contrast, the draft illustrated landscape dictionary for Yindjibarndi (Turk & Mark 2008) has about 100 terms under 50 main entries. With a similar number of “main entries” (somewhat abstracted terms), the greater number of low-level or compound terms in Navajo may relate to the synthetic nature of the Navajo language. Another interesting aspect of Navajo landscape terminology is that many terms are compounds that explicitly indicate the materials of which the feature consists. Thus, there are many terms beginning with tsé (rock) and to (water), as well as lesser numbers beginning with words for other earth materials such as sand and clay. The stem verb of a term often indicates what the material is “doing” or how it is arranged. This superficially resembles the structure of complex landscape terms in the Seri language of Sonora, Mexico (O’Meara & Bohnemeyer 2008), a language isolate unrelated to the Athabaskan languages. The Navajo language also has a number of simple landscape terms with obscure etymologies. An example of the relative importance of materials is provided by terms for longitudinal depressions in the landscape. English and many European languages have two or more terms for longitudinal depressions, depending on the size of the feature. English-language terms such as ‘valley,’ ‘canyon,’ ‘gorge,’ ‘gully,’ ‘ravine,’ and ‘rill’ are distinguished from each other mainly by size. If material is relevant, the material is included as an adjective (e.g., “rock canyon”). For Navajo, there are



Ethnophysiography

different terms depending on the material into which the feature cuts: bikooh for a large canyon or small gully with sides of dirt or soil, and tsékooh for small or large longitudinal depressions cut into rock. If the size of a feature must be communicated in Navajo, a general-purpose size term is added.

Names, terms, and descriptions (links to toponymy) An interesting characteristic of landscape reference in the Navajo language is that the differences among toponyms (place names), landscape descriptions, and landscape terms are subtle. This differentiates Navajo from both Yindjibarndi and English, but also complicates analysis and interpretation. As noted by Jett (1997, 2001), almost all Navajo toponyms are translatable into English, and the majority translate into descriptive English phrases. The Navajo language has a term, hoolyé, translated as “a-place-called…” (Kelley & Francis 1994:â•›85; Kelley, personal communication, 2006). However, hoolyé is often omitted, so while its presence removes ambiguity and clearly indicates a toponym, a word not followed by hoolyé might still be a toponym. The Navajo language also has a particularizing enclitic, -í, which is similar in function to the English definite article the in that it denotes a particular instance, e.g. “the rock” or “that rock”. The presence of -í transforms a generic term or description into a reference to a single instance, but once again, this enclitic is sometimes omitted. Generic landscape terms, such as “hill,” commonly occur as elements of toponyms in many languages, indicating the category to which the feature belongs. Jett (2001, this volume) has coined the term ‘semi-generic’ to refer to descriptive feature names in Navajo that occur frequently within a region, yet are richer and more complex than ordinary generics. An example is tsé ’íí’áhí, literally ‘the rock that is standing’. Is tsé ’íí’áhí a type of landscape feature to be included in a Navajo landscape dictionary, or a feature name that occurs frequently, or a description? Perhaps it is all three. Relationship of landscape terms to culture and spirituality As described in published and internet sources, traditional Navajo beliefs do not make a distinction between animate and inanimate objects. Also, every kind of thing and, sometimes, individual landscape features (such as specific mountains and rivers) are considered to be male or female. Mountains, rock formations, and rivers have “inner forms”. Sometimes the land on one side of a certain river is male and on the other side it is female. In Navajo “most plants have at least three names, the ‘real’ name, the way-in-which-it-is-used name, and a descriptive name” (Vestal 1952:â•›57). Similarly, many landscape features have two or more names: a name used for secular contexts such as hunting or wayfinding, and another name in the context of sacred stories (Williams & Blackhorse 2006).

35

36

Andrew G. Turk, David M. Mark and David Stea

Also, within traditional beliefs, many angular rock formations are considered to be the turned-to-stone bodies of monsters killed by the Hero Twins, as described in origin stories. The origin stories often contain lists of places visited by Holy Beings, enhancing both memory for the events in the stories and knowledge of the geography of Navajo country.

4.

Ethnophysiography Descriptive Model

4.1

Beyond descriptive ethnography

This section discusses the formation of an initial descriptive model explaining the ethnophysiographic factors emerging from analyzing existing case studies. Of course, as factors are likely to be interdependent, separating them is only for the purpose of trying to understand and explain the differences found between the ways in which various languages refer to landscape. It is hoped that this initial version of the model will stimulate discussion and lead to more effective analysis of generic issues underlying the results of the individual case studies. The basic ethnophysiography hypothesis is: people from different language groups/cultures have different ways of conceptualizing landscape, as evidenced by different terminology and ways of talking about, and naming, landscape features. To investigate this proposition it is hence necessary to have some way of developing constructs for its key elements: ‘different language groups/cultures’; ‘different ways of conceptualizing landscape features’; ‘different terminology and ways of talking about landscape features’; ‘different ways of naming landscape features’. Development of each of these formal constructs is only beginning. Some of the issues relevant to each are as follows: Different language groups/cultures: – Grouping in terms of languages vs. dialects; – Use of geographically and/or historically coherent groups; – Interaction between ‘language’ and other constructs of culture, such as ‘religion’ and ‘ethnicity.’ Different ways of conceptualizing landscape features: – Ways of cutting up landscape into features; – Conceptualization of features as object vs. field; – Spiritual significance of features.



Ethnophysiography

Different terminology and ways of talking about landscape features: – Use of generic landscape terms; – Grammatical issues: e.g. use of single words vs. phrases for features; – Talking about landscape differently in different contexts of discourse. Different ways of naming landscape features: – What types of features are given proper names; – Relationship of proper names to landscape terms; – Cultural significance of naming practices. Development of such constructs is being advanced by the case studies reported in this and other chapters of this book. If ethnophysiography is to be merely ethnographically descriptive it may be sufficient to refine these formal constructs in such a way as to enable them to adequately express all data obtained in cases studies. However, there is a deeper analytical objective in most, if not all, studies of the representation of landscape in language: the desire to understand which language feature terms are universal and which are culturally determined (and therefore different); and the causal apparatus involved in observed cultural differences. It is, however, important to recognize that some people who contribute data to this general field of investigation are primarily motivated by the desire for language documentation and preservation. However, even in these cases, the linguists have an underlying interest in at least some types of inter-language comparisons, such as differences between or among grammatical systems.

4.2

Towards a descriptive model of ethnophysiographic factors

As a field of theoretical development ethnophysiography must ‘walk before it runs.’ Theory (“grounded theory”) should be developed in a logical manner from the ground up. It is extremely easy to start developing causal theories that apply only to part of the domain, ignoring yet undiscovered confounding variables. Hence, it is important to start with a sequence of ever more complete and coherent descriptive models before attempting a causal or predictive model. Theory describes what is observed and then predicts what might be observed under specified conditions: the latter involves relations and even, in many cases, causation. By explicitly laying out the details of a descriptive model (together with examples of each element drawn from published case studies) it is possible to uncover weaknesses in the current taxonomy of factors and to test whether any new data fits the existing descriptive model or requires its revision.

37

38

Andrew G. Turk, David M. Mark and David Stea

Each iteration of a descriptive model can also assist researchers to seek out new case study data which might conflict with the existing model and perhaps reveal new factors. For instance, European scientists may well have felt justified in making the leap from correlation (“all swans so far observed are white”) to causation (“all swans are necessarily white”) prior to the discovery of black swans in Australia. Indeed, part of the drive for exploration of parts of the world poorly known to Europeans in the ‘enlightenment project’ was seeking out examples of flora and fauna new to the West, and documenting the environmental conditions of their location. The activities of naturalist Charles Darwin during the voyage of HMS Beagle (from 27 December 1831 to 2 October 1836) is a classic example of this (Darwin 1845, as cited in Quammen 2008). Darwin’s deep consideration of the variations he discovered and the mechanisms of differentiation he hypothesized resulted in perhaps the most celebrated scientific theory of modern times: the theory of the origin of species by means of natural selection.

4.3

Preliminary version of Ethnophysiography Descriptive Model

Understanding of the differences in the way landscape is treated in language is still emerging. However, based on the ethnophysiography studies to date and informed by treatment of these issues by MPI researchers (Burenhult & Levinson 2008), those differences currently identified may be assembled into groups. These include the following (with examples drawn from the Yindjibarndi case study of the Ethnophysiography Research Group and some studies included in Burenhult 2008a): 1. Completeness of the set of landscape terms used in any particular language, within the range of all possible landscape types [e.g. in Yindjibarndi (NW Australia), there are no terms for a cinder cone or a coral reef]; 2. Granularity of classification of landscape types used in sets of terms [e.g. Yindjibarndi uses the same term marnda for entities termed mountains, ranges, ridges, and (some) hills in English. Similarly, in the Lowland Chontal language (in Oaxaca, Pacific coast of Mexico) ijwala labels any elevated entity, whether mountain, hill, or highlands, in general; a single word, ane’, refers to any path, from a dirt track to a paved highway (O’Connor & Kroefges 2008)]; 3. Use of compound words and phrases, instead of simple generic landscape terms [e.g. in Yindjibarndi, a “mesa” is called a marnda marlirri (flat-topped hill); in Seri (Mexico) most landscape terms are complex and based on four substances plus posture semantics (O’Meara & Bohnemeyer 2008)];



Ethnophysiography

4. Role of nouns vs. verbs vs. prepositions in landscape terms [e.g. in Kilivila (Kaile’una Island, PNG) landscape terms are mostly (common) nouns, but some are prepositional phrases (locatives and directionals) (Senft 2008)]; 5. Role of metaphors and/or frames for structuring landscape terms (referred to as ‘semplates’ in Burenhult & Levinson 2010) [e.g. in Jahai (Malay Peninsula) terms for hydrological features, there is a strong use of metaphor (from animate domains, mainly human) of two sorts: body parts (creating partonomy) and kinship terms (creating size taxonomy) (Burenhult 2008b)]; 6. Distinctions between referring to parts of landscape as objects or places [e.g. in Marquesan (spoken on steep rocky volcanic islands in the Pacific Ocean) there are two different prepositions for “at,” one for ‘at a (landscape) object’ and the other for ‘at a place’ – “Both islands and islets and small lava rocks sticking out of the sea are all motu, but small ones take the object-marking preposition ‘io and large islands the place preposition ‘i” (Cablitz 2008)]; 7. Role of place names (toponyms) vs. generic landscape terms: e.g., where each mountain is given a proper name, it may not be necessary to have a generic term for ‘mountain.’ Alternatively, types of landscape features may have both generic terms and toponyms [e.g. in Yélî Dnye (Rossel Island, PNG) toponymy seems to follow (almost exactly) terms for landscape features (rather than working as a complementary system) (Levinson 2008)]; 8. Structure of place names: whether they are descriptive; if they include generic landscape terms; and how they arise and are constructed [e.g. in Marquesan most place names are semantically transparent and have been derived from definite descriptions (Cablitz 2008); in Kilivila toponyms are almost all in the form of simple nouns (although their meaning may be complex) and are very culturally specific (Senft 2008)]; 9. Role of religious beliefs relating to landscape, implicitly embedded in landscape terms [e.g. in Yindjibarndi, each yinda (permanent pool) has a warlu (spirit); in Jahai waterfalls and rocky outcrops are linked to mythology (Burenhult 2008b)]. The factors which might cause a language to refer to landscape in a particular way include the following (again with examples drawn from the Yindjibarndi case study of the Ethnophysiography Research Group and some studies included in Burenhult 2008a): A. Physical environment factors: A1. Topography of the region occupied by the language group: whether mountainous, hilly or flat; the presence or absence of particular landscape features, such as volcanic cinder cones, sand dunes, coral reefs, etc. [for

39

40 Andrew G. Turk, David M. Mark and David Stea

example, since there are no cinder cones in Yindjibarndi country or nearby, the language would not be expected to include terms for them. Yindjibarndi does have a term for ocean/sea (thanardi), which occurs in the territory of the adjacent language group but not in Yindjibarndi ‘country.’ However, Yindjibarndi does not appear to have a term for sand ridges, which occur a similar distance inland. Perhaps there was more traditional trade with the coastal people, who are also considered less fierce than the Yindjibarndi’s inland neighbors. Of course the sea is a much more important feature than sand ridges near Yindjibarndi country. However, sand ridges are very significant features for the Ngaanyatjarra people (from country 1,200 km inland from Yindjibarndi country), who have a similar multi-meaning mountain/hill/rock term like Yindjibarndi marnda (i.e. purli) but also a separate term for ‘sand ridge’ (tali) (Glass & Hackett 2003)]; A2. Climate of the region: the “strength” of seasons (e.g. does it snow in winter?); seasonal variability (e.g. does rainfall come only from seasonal cyclones/hurricanes?); etc. [e.g. there are no permanent rivers or creeks in Yindjibarndi country and larger watercourses have running water only after major precipitation events, usually associated with cyclones (hurricanes). Hence it is not unreasonable for the term wundu (river) to refer only to the longitudinal depression through which water passes, and to use other terms for different sorts of water flow: yijirdi (shallow and slow) and mankurdu (deep and fast)]; A3. Vegetation of the region: its variability in space and time; its density; its uses (e.g. for shelter, food and medicine); its affordances regarding travel, blocking long views, etc. [e.g. in ≠Akhoe Hai//om (a Khoisan language from the Oshikoto region of Northern Namibia, in SW Africa) people use cardinal direction terms to describe walking tracks, perhaps because “in the central part the thick bush and woodland covers permanent dunes which are generally too low (and bush too high) to allow wide vistas” (Widlok 2008)]; B. Social environment factors: B1. Lifestyle and traditional economy of the people: whether they are hunter/ gatherers, cultivators, etc. [e.g. in Kilivila, where people cultivate food gardens, there are numerous words for different soil types (Senft 2008)]; B2. Settlement patterns: whether settlements are permanent or nomadic camps; and the way in which buildings and/or gardens/fields interact with the natural landscape to produce settled places [e.g. the ≠Akhoe Hai//om language generally uses female word endings for roundish objects and male word endings for long and tall things, distinguishing between areas



Ethnophysiography

of land that are round (e.g. a settlement) and those which are long (e.g. a riverine region) (Widlok 2008)]; B3. Religious beliefs or spiritual concerns linked to landscape, such as creation beliefs, presence of spirits in landscape features, and cultural practices (e.g. ceremonies, taboos) which accompany such beliefs [e.g. among Lowland Chontal speakers, springs are commonly venerated as sacred places and the locations of supernatural forces; there are sacred hills and sacred rivers, dwelling places of either dwarf-like creatures or the Devil (these places receive a plural prefix in their toponyms to mark them grammatically) (O’Connor & Kroefges 2008)]; B4. Historical factors, such as: movement of the people into their current region; colonization of the people and their land by outsiders with a significantly different culture/language; major changes in lifestyle and economy; etc. [e.g. as Lowland Chontal uses a Spanish loan word for a territorial boundary, perhaps boundaries were not considered at all (or thought of differently) prior to colonization (O’Connor & Kroefges 2008). There has been a similar effect observed among the Kelabit people of central Borneo (Bala 1999)]; C. Linguistic factors: C1. Structure/grammar of the language, such as the roles played by nouns, verbs, adjectives, compound words, noun phrases, etc. [e.g. in Jahai a small number of general landscape terms are of simplex form (e.g. hGp = ‘forest’) but most terms are complex, always consisting of a simplex term combined with a metaphorical noun (e.g. kuy tfm = ‘water-head’, for English ‘headwaters’) (Burenhult 2008b)]; C2. Influence of other languages, through cross-fertilization by neighboring languages, impact of colonial language(s), etc. [e.g. in Jahai, “loanwords” from Malay (about 20% of landscape terms collected) often do not have indigenous synonyms (Burenhult 2008b); in Lowland Chontal some binomial toponyms mix a Lowland Chontal and a Spanish word, in either order (O’Connor & Kroefges 2008)]; C3. Use of a non-traditional language for a particular purpose. One example would be a ‘pidgin’ or ‘lingua franca,’ especially for dealings with colonizers or visitors. Another would be the special version of a traditional language used for ceremonies [e.g. in Yindjibarndi the term for a road/ trail/track is yirdiya; however, mardiya is used in the ‘respect language’ (used in major ceremonies involving visitors from other tribes) and is sometimes used to refer to a track followed through country, assisted by landmarks].

41

42

Andrew G. Turk, David M. Mark and David Stea

5.

Conclusions

This chapter has provided some details of two ethnophysiography case studies. The fieldwork for the Yindjibarndi case study has been completed and all that remains is finalization of the Illustrated Dictionary of Yindjibarndi Landscape Terms and preparation of final summary articles. This first case study assisted the researchers greatly in developing an understanding of ethnophysiographic methodology and theory. It established prima facie support for the basic ethnophysiography hypothesis. The Navajo case study will complement the one undertaken with Yindjibarndi speakers. Once all Navajo study field recordings have been transcribed, detailed analysis of landscape terms and toponyms can be undertaken. This will allow more detailed comparison with the findings of the Yindjibarndi case study and the results of other ethnophysiography-related research. The main findings to date from these case studies were discussed in the context of how they point to key research issues regarding the way in which landscape is referred to in different languages. These results, together with those from similar studies undertaken by the Language and Cognition Group at the Max Planck Institute of Psycholinguistics, Nijmegen, formed the basis for developing the transdisciplinary “Landscape in Language Workshop” and, thus, this book. An initial descriptive model of ethnophysiographic factors is presented here, including key observed differences in the way languages treat landscape and also a set of factors which might be considered as possibly explaining such differences. Much more work is required to render the Ethnophysiography Descriptive Model more complete and effective. However, the model indicates some of the key issues involved in ethnophysiography. It can also be useful to compare the results of particular case studies and to assist researchers to decide which future studies may best fill in gaps in existing knowledge, and can help to explain the significance of particular factors in the model.

Acknowledgments The authors acknowledge the assistance of the Yindjibarndi consultants, including: Trevor Soloman, Allery Sandy, Dora Soloman, Nita Fishook, Nelly Wally, Pansy Manda, Aileen Sandy, Marleen Harold, Patricia Pat, Donna Willis, Wendy Darby, Marion Cheedy, Jane Cheedy, Cherry Cheedy, Ned Cheedy, Rosie Cheedy, Sylvia Allen, Pansy Sambo, Lorraine Coppin and Michael Woodley. The cultural organization Juluwarlu and the Pilbara Aboriginal Language Centre (Wangka Maya) provided invaluable assistance regarding the Yindjibarndi language. Thanks also



Ethnophysiography

to the Australian Institute of Aboriginal and Torres Strait Islander Studies for providing material from the Aboriginal Studies Electronic Data Archive (ASEDA). For assistance with the Navajo case study, the authors thank Carmelita Topaha, Larry King, Jay Williams, and all the Navajo consultants, too numerous to list here. The Navajo portion of the project has been conducted under permit C0513-E from the Historic Preservation Office of the Navajo Nation. Researchers at the Max Planck Institute for Psycholinguistics have also been very helpful. Funding by the US National Science Foundation (grants BCS-0423075 and BCS-0423023) and from Murdoch University, supporting the ethnophysiography projects, is gratefully acknowledged.

References Anderson, Bruce. 1986. Yindjibarndi dictionary. Photocopy. Anderson, Bruce & Thieberger, Nicholas. n.d. Yindjibarndi Dictionary [Document 0297 of the Aboriginal Studies Electronic Data Archive (ASEDA)]. Canberra: Australian Institute of Aboriginal and Torres Strait Islander Studies. Bala, Poline. 1999. Permanent Boundary Lines in the Kelabit Highlands of Central Borneo: A Colonial Legacy. MA thesis, Cornell University. Burenhult, Niclas (ed.). 2008a. Language and Landscape: Geographical Ontology in Cross-linguistic Perspective [Language Sciences 30(2/3)]. Amsterdam: Elsevier. Burenhult, Niclas. 2008b. Streams of words: Hydrological lexicon in Jahai. Language Sciences 30(2/3): 182–199. Burenhult, Niclas & Levinson, Stephen C. 2008. Language and landscape: A cross-linguistic perspective. Language Sciences 30(2/3): 135–150. Burenhult, Niclas & Levinson, Stephen C. 2010. Semplates: A guide to identification and elicitation. In Field Manual Volume 13, E. Norcliffe & N. J. Enfield (eds), 17–23. Nijmegen: Max Planck Institute for Psycholinguistics. Cablitz, Gabriele H. 2008. When ‘what’ is ‘where’: A linguistic analysis of landscape terms, place names and body part terms in Marquesan (Oceanic, French Polynesia). Language Sciences 30(2/3): 200–226. Darwin, Charles. 1845. Journal of Researches into the Natural History and Geology of the Countries Visited during the Voyage of H. M. S. Beagle round the World, under the Command of Capt. Fitz Roy, R. N., 2nd edn. London: John Murray. Glass, Amee & Hackett, Dorothy. 2003. Ngaanyatjarra and Ngaatjatjarra to English Dictionary. Alice Springs: IAD Press. Ieramugadu Group Inc. 1995. Know the Song, Know the Country: The Ngarda-Ngali Story of Culture and History in the Roebourne District. Roebourne, Western Australia: Ieramugadu Group. Iverson, Peter, with Roessel, Monty. 2002. Diné: A History of the Navajos. Albuquerque NM: University of New Mexico Press. Jett, Stephen C. 1997. Place-naming, environment, and perception among the Canyon de Chelly Navajo of Arizona. Professional Geographer 49(4): 481–493.

43

44 Andrew G. Turk, David M. Mark and David Stea

Jett, Stephen C. 2001. Navajo Placenames and Trails of the Canyon de Chelly System, Arizona. Frankfurt: Peter Lang. Kari, James. 1989. Affix positions and zones in the Athapaskan verb complex: Ahtna and Navajo. International Journal of American Linguistics 55(4): 424–454. Kelley, Klara B. & Francis, Harris. 1994. Navajo Sacred Places. Bloomington IN: Indiana University Press. Levinson, Stephen C. 2008. Landscape, seascape and the ontology of places on Rossel Island, Papua New Guinea. Language Sciences 30(2/3): 256–290. Mark, David M. & Turk, Andrew G. 2003a. Ethnophysiography. Paper presented at the Workshop on Spatial and Geographic Ontologies, 23 September, 2003 (prior to COSIT03), in Ittingen, Switzerland. Mark, David M. & Turk, Andrew G. 2003b. Landscape categories in Yindjibarndi: Ontology, environment, and language. In Spatial Information Theory: Foundations of Geographic Information Science [Lecture Notes in Computer Science 2825], Werner Kuhn, Michael F. Worboys & Sabine Timpf (eds), 28–45. Berlin: Springer. Mark, David M., Turk, Andrew G. & Stea, David. 2007. Progress on Yindjibarndi ethnophysiography. In Spatial Information Theory [Lecture Notes in Computer Science 4736], Stephan Winter, Matt Duckham, Lars Kulik & Ben Kuipers (eds), 1–19. Berlin: Springer. Mark, David M., Turk, Andrew G. & Stea, David. 2010. Ethnophysiography of arid lands: categories for landscape features. In Landscape Ethnoecology, Concepts of Biotic Physical and Space, Leslie M. Johnson & Eugene S. Hunn (eds), 27–45. New York NY: Berghahn Books. O’Connor, Loretta & Kroefges, Peter K. 2008. The land remembers: landscape terms and place names in Lowland Chontal of Oaxaca, Mexico. Language Sciences 30(2/3): 291–315. O’Meara, Carolyn & Bohnemeyer, Jürgen. 2008. Complex landscape terms in Seri. Language Sciences 30(2/3): 316–339. Quammen, David (ed.). 2008. Charles Darwin, On the Origin of Species: The Illustrated Edition. New York NY: Sterling. Rijavec, Frank, Harrison, Noelene & Soloman, Roger. 1995. Exile and the Kingdom [documentary film]. Roebourne, Western Australia: Ieramugadu Group & Film Australia. Senft, Gunter. 2008. Landscape terms and place names in the Trobriand Islands – the Kaile’una subset. Language Sciences 30(2/3): 340–361. SIL. 2000. Yindjibarndi: a language of Australia. Ethnologue: Languages of the World, 14th ed., (December 2001). Thieberger, Nicholas (compiler). 1996. Handbook of Western Australian languages south of the Kimberley region. (August 2003). Tindale, Norman B. 1974. Aboriginal Tribes of Australia. Their Terrain, Environmental Controls, Distribution, Limits, and Proper Names. Canberra: Australian National University Press. Trees, Kathryn A. & Turk, Andrew G. 1998. Culture, collaboration and communication: Participative development of the Ieramugadu Cultural Heritage Information System (ICIS). Critical Arts 12(1–2): 78–91. Turk, Andrew G. & Mark, David M. 2008. Illustrated Dictionary of Yindjibarndi Landscape Terms. Informal publication, Murdoch University (Australia). Turk, Andrew, Mark, David, O’Meara, Carolyn & Stea, David. In press. Geography – documenting terms for landscape features. In Oxford Handbook of Linguistic Fieldwork, Nicholas Thieberger (ed.). Oxford: OUP.



Ethnophysiography

Turk, Andrew G. & Trees, Kathryn A. 2000. Facilitating community processes through culturally appropriate informatics: An Australian Indigenous Community Information System case study. In Community Informatics: Enabling Communities with Information and Communication Technologies, Michael Gurstein (ed.), 339–358. Hershey PA: Idea Group. Vestal, Paul A. 1952. Ethnobotany of the Ramah Navaho [Papers of the Peabody Museum of American Archaeology and Ethnology 40(4)]. Boston MA: Harvard University. Widlok, Thomas. 2008. Landscape unbounded: Space, place, and orientation in ≠Akhoe Hai//om and beyond. Language Sciences 30(2/3): 362–380. Williams, Jay S. & Blackhorse, Taft. 2006. Haash holyé? “What’s this place called?”: Metaphor, metonymy, and iconicity in Navajo. Ms, Albuquerque NM. Wordick, Frank J. F. 1982. The Yindjibarndi Language [Pacific Linguistics C-71]. Canberra: Australian University. Young, Robert W. & Morgan, William. 1992. Analytical Lexicon of Navajo. Albuquerque NM: University of New Mexico Press.

45

chapter 3

Exploring philosophy of place Potential for synergy between phenomenology  and ethnophysiography Andrew G. Turk

This chapter discusses aspects of place from the perspective of phenomenology. It draws upon the author’s collaborative research on ethnophysiography discussed in earlier chapters of this volume. Ethnophysiography research requires an integrated approach by researchers from many disciplines; i.e. it should be transdisciplinary and utilize an overarching paradigm. Phenomenology is a candidate to provide this overarching paradigm. This paper discusses this issue and also the potential for research in ethnophysiography to contribute to an enhanced understanding of phenomenology. These twin objectives must be carried out in a manner which respects Indigenous knowledges. Hence, the chapter also discusses Indigenous Australian concepts of ‘The Dreaming’ and  ‘Tjukurrpa’ and their role as organizing structures for concepts of place.

1.

Introduction

As discussed in earlier chapters (Turk et al. this volume), ethnophysiography is a recently defined field of study that seeks to understand cultural differences in conceptualizations of landscape, via comparisons between the meanings of terms that people from different cultures use to refer to the landscape and its components (Mark & Turk 2003a). Landscape is an interesting topic of study because relationships with land are central to many cultures and landscape features pose problems for classification. Ethnophysiography also includes study of the nature of place names (toponyms) and their relationship to generic landscape terms. In addition, a central concern for ethnophysiography is the study of the knowledge systems, beliefs and customs of peoples concerning landforms and landscapes. Thus, ethnophysiography involves the study of ‘place’ and ‘place attachment’ and examines how these are related to traditional beliefs and people’s

48 Andrew G. Turk

everyday activities, including cultural practices. The goal is to describe the meanings of landscape held by particular cultural/linguistic groups. Hence, there is a considerable overlap with topics of primary interest to phenomenology, and initial research by the author supports the potential for a synergistic relationship between ethnophysiography and the theoretical and methodological framework provided by phenomenology. A prima facie case for use of phenomenology as an overarching paradigm for transdisciplinary research in ethnophysiography was established in Turk (2007). This investigation also provided an initial examination of the research trajectories of the two fields and a brief introduction to how the concept of ‘essences’ might be usefully applied within ethnophysiography research. This chapter also seeks to demonstrate the potential for the reverse relationship, i.e. the proposition that ethnophysiography research might assist in explicating aspects of place in phenomenology. In his chapter in Feld and Basso’s collection “Senses of Place” (1996), Casey suggests a revitalized phenomenological examination of place: To reinstate place in the wake of its demise in modern Western thought – where space and time have held such triumphant and exclusive sway – one can equally well go to the premodern moments described in ethnographic accounts of traditional societies or to the postmodern moment of the increasingly non-traditional present, where place has been returning as a reinvigorated revenant in the writings of ecologists and landscape theorists, geographers and historians, sociologists and political thinkers – and now, in this volume, anthropologists.  (p. 20)

In the present chapter, the author suggests a possible way to integrate these two related, but different, approaches (premodern and postmodern) via ethnophysiography.

2.

Complexities of ethnophysiography

A summary of the ethnophysiography case studies with Yindjibarndi and Navajo peoples is provided by Turk et al. (this volume). The work to date indicates that terms for landscape features from different languages differ significantly. The following list of types of landscape terms (using examples from Yindjibarndi unless otherwise stated) indicates that such differences become increasingly complex as understandings of languages in the cultural context are developed: a. Terms for fairly equivalent concepts between languages – e.g. muji ‘cave’: Muji is the word for a deep hole in marnda (a rock face), with an overhanging roof (a cave), large enough for a person or animal to shelter in.



Exploring philosophy of place

b. Terms for sets of landscape features that do not match up – e.g. marnda, bargu, burbaa ‘mountain, hill’: Marnda refers to an area of ngurra (ground) that stands higher than the country around it – a hill, mountain, ridge or range. A bargu is a gubija (small) marnda made of rock or sand. Burbaa can refer to a gubija (small) marnda that is low, smooth and rounded – not as steep as a bargu. c. Different whole-part concepts – e.g. wundu, mankurdu, yijirdi ‘river’: A wundu is a relatively broad, level channel where water flows or lies after heavy rain. In Yindjibarndi country, there are no wundu that flow with water all the time. But the wundu is always there, and water may flow in the wundu after rainstorms. If there is water in a wundu, it is referred to as mankurdu if flowing deep and fast, yijirdi if flowing gently, and bawa if it is just lying there temporarily. d. Terms effectively referring to shape rather than topographic objects – e.g. burbaa: Burbaa can refer to a gubija (small) marnda that is low, smooth and rounded€– not as steep as a bargu. Burbaa also refers to a rise in a yirdiya (road, track or trail), especially at the crest. Burbaa can also be used to refer to low rounded areas of higher ground, for instance between garga (gullies). e. Physical point of view – e.g. (in Navajo) initial results suggest that tséyi’ could be used for a rock canyon seen from the canyon floor, whereas tsékooh could be used when the same feature is seen from the rim above. f. Spatial locations rather than landscape features – e.g. wana, wanangga: Wana refers to a hillside (or perhaps a flat area) in the middle distance; where you can still see something (like a kangaroo) but it is much too far away to throw a stone at it (or shoot the kangaroo). Wanangga could refer to the location of something in the middle distance. g. Spiritual aspects of place – e.g. yinda: Every yinda (permanent pool) has a warlu (spirit) that formed and protects the yinda. h. Sets of places related by ‘The Dreaming’ – e.g. Jindawurru: During an audio recording session at Juluwarlu (the Yindjibarndi culture maintenance group) where photos of landscape features were examined, with the researchers (AT and DM) asking for landscape terms, the informants (CC, ED and FV) spontaneously referred to spiritual aspects of the features. The following (edited) example from the transcripts illustrates this: Photo 31 (very large pool at Millstream) CC: This is a yinda. This is Deep Reach. ……

49

50

Andrew G. Turk

CC: I think we will have bin explained to you what this is (referring to previous discussions with AT) ED: It means a lot to us. AT: Yes CC: That’s where the two men been taken, you know? ED: And buried – very frightening for us to go and stop there, you know … CC: You see the hole on top of the bank in Millstream (addressing DM)? DM: I have not seen it. ED: You ought to see it. Maybe more bigger than this (showing size with her hands). Snake got up, got the two men, and come back – this wasn’t like this, it was a dry riverbed – ‘til that. Camping ground for the old people as well, long ago. CC: In the wundu, yes. ED: In the Dreamtime, yes. ……. CC: They bin crying, crying, and that warlu got wild, sent the big flood come down. FV: Made them permanent pools.



Here the informants were talking about how the warlu (mythic snake) had come up out of the ocean near the location of the town of Onslow and traveled up the route of what is now called the Fortescue River, chasing two boys who broke the law, until he got to Jindawurru (Millstream) (Ieramugadu Group 1995). This is part of an explanation of what Yindjibarndi call the ‘learning times’ or ‘when the world was soft.’ This is also known as ‘The Dreaming,’ a translation of Indigenous explanations of the formation of the world into its current landscape and ongoing spiritual aspects of place.

The sequence above indicates some of the complexities of the role of landscape in language. The preliminary Ethnophysiography Descriptive Model (presented Turk et al. this volume) seeks to explicate the range of factors involved in the different ways that languages treat landscape and the possible range of reasons for this. Thus it can be seen that the development of theories for ethnophysiography is a very complex undertaking, requiring contributions from many disciplines.

3.

Collaborative research

There are many complex questions associated with the study of the role of landscape in language. This requires input from a range of disciplines, including: linguistics,



Exploring philosophy of place

geography, cognitive and experimental psychology, information systems, anthropology, philosophy and cultural studies. Hence, ethnophysiography research needs to be carried out in a way that integrates theory and analysis in an effective manner. However, Gallagher (2008:â•›18), in discussing how cognition is embedded in context, analyzes how language operates within scientific disciplines and suggests that this “brings us to the possibility that theories of situated cognition are themselves differently situated, within different disciplines or discourses, shaped by specific debates and specialized vocabularies.” Thus it is necessary to contextualize and explain our use of terms within disciplines and to try to express concepts in ways that build bridges between disciplines, rather than walls to hide behind. Increased levels of integration of fields of knowledge and research paradigms are necessary in order to address complex “messy” research problems. Alternative approaches, at increasing levels of research integration, are summarized in Table 1. The last of the levels of research integration listed in Table 1 (postdisciplinary) is difficult to achieve other than in individual or small group projects. Hence the highest level of integration practically possible for ethnophysiography is a transdisciplinary approach. This requires an overarching paradigm (spanning the various disciplines involved) that is operationalized by researchers in a manner which permits theoretic discussions from within the individual disciplines to be integrated in a coherent way. It must also provide a mechanism for the integrated analysis of empirical data from case studies. The next section introduces phenomenology, a potential approach to such integration. Bruce Janz (this volume) suggests that ethnophysiography can be considered as a nascent discipline (although the researchers have never claimed this). He contends that the strength of the discipline approach is the synergistic integration of a variety of methodological approaches that it potentially affords, although some disciplines display deep and acrimonious fissures along philosophical and methodological fault lines. He goes on to suggest that a transdisciplinary paradigm might stifle methodological plurality. Although this is possible, it is not necessary if the overarching paradigm is suitable and appropriately applied. A great strength of the ethnophysiography research project to date has been the synergistic integration of phenomenological and positivist methodologies and interpretative approaches. In addition, there has been an iterative approach to shifts from emic to etic and back to emic representations (Janz this volume). Emic data was collected, clarified and analyzed, leading to initial ontological (etic) interpretations. These were then re-evaluated by further emic data gathering and, as the project progressed, (etic) draft pictorial landscape dictionaries were reviewed by local participants (Turk, Mark, O’Meara & Stea in press).

51

52

Andrew G. Turk

Table 1.╇ Differing levels of research integration (after Booth et al. 2000) Level of integration

Characteristics of this type of research approach

Disciplinary

• Research using only particular bodies of knowledge, paradigms, theories, methods; • Disciplines are a relatively recent phenomenon (perhaps only 200 years old); • Some are very recent; e.g. Information Science. • Researchers from more than one discipline working on a research project; • Each discipline uses its own paradigm and methods to do part of the research; • Results from different parts of the research are combined. • Researchers from more than one discipline working on a research project; • The disciplines try to integrate aspects of the research by using multiple paradigms and methods; • Results from different paradigms and methods of research are combined. • Researchers from more than one discipline working on a research project; • The disciplines integrate their research paradigms and methods via some overarching paradigm and meta-methodology; • Integration of methods and results is more effective than other approaches, provided one discipline is not privileged over the others. • A holistic approach is used, which seeks to avoid disciplinary divides; • Highly reflective practice.

Multidisciplinary

Interdisciplinary

Transdisciplinary1

Postdisciplinary

4.

Phenomenology

Phenomenology is a branch of European philosophy started in the early 20th century by Edmund Husserl (1859–1938), although the term is now used more broadly in many disciplines, such as: psychology; ecology; architecture and archaeology (Fuchs 1976; Heidegger 1962; Ihde & Zaner 1975; Moran & Embree 2004; Patocka 1996). It seeks to understand how a person’s consciousness interacts with things around them, including landscape, as part of their mode of ‘being in the world.’ It provides an alternative approach to objectivism and positivism.

. Although this is the dominant meaning of “transdisciplinary,” some authors adopt other meanings (Nicolescu 2008; Nowotny 2003; Schunn et al. 2006).



Exploring philosophy of place

Mohanty (1997) provides an analysis of the field of phenomenology in terms of a trajectory of concerns, including the following: a. As an ontological endeavour, phenomenology describes the nature of the core phenomena of domains of human experience: “Phenomenology started with the program of describing essences and essential structures of various regions of phenomena” (p. 1). b. This search for ‘essences’ and ‘structures,’ within particular domains, led phenomenology to be concerned with meanings: “Whereas the concept of essence was an ontological concept, an essence being an entity of a certain sort, an ideal entity to be sure, nevertheless an entity belonging to the world, the concept of meaning is a semantic concept to begin with but then, in the extended sense given to it by phenomenology, becomes ambiguously poised in between the mind and the world, being the way the world is presented to experience” (p. 2). c. Thus, consciousness is a central concern of phenomenology: “… consciousness constitutes the world, confers sense on all things, not only provides access to the world, but is the very presenting of the world, making it evident, the source of its being and vitality” (p. 2). d. Phenomenology considers consciousness in the context of people’s activities in the world: “… the world as lived by us in our everyday life, the world of perception and interest, valuations and actions” (p. 2). e. Phenomenology contends that people experience and represent their world in an integrated manner: “Husserlian phenomenology starts with an insistence upon the integral unity of language, thought, and experience” (p. xi). f. It is not enough for phenomenology to merely bring to the surface and illuminate particular aspects of ‘lifeworld’ (Lebenswelt); rather the project is more analytical, seeking explanation of relationships between ontologies, consciousness, meaning and human activities: “… if a science of the lifeworld is not to be empirical ethnology, if it is still to be philosophy, it has to uncover the essential structure of lifeworld, and not merely its contingent, variable features, features which vary from community to community” (pp. 2, 3). g. Concern with experiences and activities in the world was further advanced in the version of phenomenology developed by Heidegger (e.g. Being and Time, 1962): “For Heidegger, we and our activities are always ‘in the world,’ our being is being-in-the-world, so we do not study our activities by bracketing the world, rather we interpret our activities and the meanings things have for us by looking to our contextual relations to things in the world” (D. Smith 2007:â•›7).

53

54

Andrew G. Turk

As phenomenology has developed, it has more adequately addressed the meaning of the world for groups, not just individuals. Kockelmans (1994:â•›25) asserts that “I experience the world as being not my own private world but an intersubjective world in which others exist both as others as well as for others,” hence a community “is constituted in the sphere of my ownness, which in its communalized intentionality constitutes the one identical world for everybody.” He goes on to explain the process of moving from individual to group concepts: “my transcendental subjectivity is gradually expanded into a transcendental intersubjectivity or community, which in turn is the transcendental ground for the intersubjective value of nature and the world in general” (25/26). For instance, the common conceptualizations of a speech community extend beyond the merely physical aspects of topographic features to encompass also their cultural and spiritual dimensions: Whether we happen to act alone or, cooperating with others, engage in common pursuits, things and objects with which we are confronted, our own plans and designs, finally the world as a whole appears to us in the light of beliefs, opinions, conceptions, certainties, etc., that prevail in the community to which we belong. … Besides being objects in the Lebenswelt, we are at the same time subjects with respect to the Lebenswelt, insofar as it derives its meaning and the sense of its existence from our collective mental life, from our acts (concatenated and interlocked with those of our fellow men) of perceiving, experiencing, reasoning, purposefully acting, etc.  (Gurwitsch 1957:â•›372)

Thus, phenomenology has the potential to assist in understanding the way that landscape terms may incorporate cultural and spiritual aspects of the worldview of the speech community.

5.

Phenomenology as overarching paradigm for ethnophysiography

Because of the nature of the field of ethnophysiography, phenomenology is seen as a candidate to provide an overarching paradigm for the transdisciplinary study of ethnophysiography. However, there have been many versions of phenomenology, some of which will be more effective than others in the context of ethnophysiography. For instance, Husserl’s universalism could perhaps be seen as contradictory to the fundamental premise of ethnophysiography. There is a long history of using phenomenology in the social sciences, e.g.: the work of Alfred Schütz, starting from the late 1920s (Schütz 1977; Schütz & Luckmann 1973); ethnomethodology developed by Harold Garfinkel in the 1960s (Psathas 1977); and Max Scheler’s theory of knowledge: “In his works he



Exploring philosophy of place

has also shown, in an exemplary manner, how phenomenology and sciences (in his case, biology, anthropology, and sociology) can be blended in one’s thinking” (Mohanty 1997:â•›23). Ihde (1986:â•›26) contends that “in its essence, phenomenology is neither obscure nor esoteric and that it holds important implications for a whole range of disciplines.” Malpas (2006) makes a strong case that an understanding of the relationship of individuals and communities to the places where they dwell is central to phenomenology. Basso (1996:â•›54) suggests that “As formulated by Martin Heidegger … the concept of dwelling assigns importance to the forms of consciousness with which individuals perceive and apprehend geographical space.” In the Introduction to their edited collection of essays on place and environment, Seamon and Mugerauer (1985) discuss the emerging engagement with phenomenology by scholars of various disciplines interested in ‘place’ – the relationship between persons and their environment. This includes consideration of themes such as: “environmental ethics, sacred space, environmental behaviour, sense of place, and a phenomenology of architectural design” (p. 2). They emphasize the significance of the work of Martin Heidegger in providing a theoretical basis for phenomenological approaches, including recent research in geography. D.€Smith (2007:â•›2) asserts that the study of consciousness via phenomenology can explain how it is directed towards things in the world: Basically, phenomenology studies the structure of various types of experience ranging from perception, thought, memory, imagination, emotion, desire, and volition to bodily awareness, embodied action, and social activity, including linguistic activity. The structure of these forms of experience typically involves what Husserl called ‘intentionality,’ that is, the directedness of experience towards things in the world, the property of consciousness that is a consciousness of or about something.

An initial comparison of key aspects of ethnophysiography and phenomenology has indicated that examination of their interaction is useful and feasible (Turk 2007). Hence, phenomenology may well be able to provide an overarching paradigm for transdisciplinary research in ethnophysiography; or, inverting the spatial metaphor, phenomenology can “stand under” (Fuchs 1976:â•›27) the disciplines involved in ethnophysiography, providing a firm foundation and scaffolding to support their activities. Some possible methods for using phenomenology in ethnophysiography are discussed in Section 10 below. The author also contends that research in ethnophysiography can perhaps be useful within phenomenology, by providing interesting examples of cultural differences in understanding the nature of place. Some possible contributions that ethnophysiography may make to phenomenology are discussed in Section€11

55

56

Andrew G. Turk

below. However, before pursuing the topics in Sections 10 and 11, it is necessary to ask whether adopting any Western philosophical position is appropriate, given that ethnophysiography examines the worldview of Indigenous, as well as non-indigenous, peoples.

6.

Indigenous knowledges

If ethnophysiography is to genuinely seek to understand the way that landscape is treated in languages across the world it must do so from a perspective that is not biased towards Western (European) concepts of knowledge. In particular, if key case studies of differences between languages are carried out with Indigenous peoples, then their worldviews must be reflected in theories that result. Such an approach is justified on a scientific and an ethical basis, but is also pragmatically necessary to ensure collaboration by Indigenous people in ethnophysiography research. Whiteley (1998) discusses what he sees as a crisis in ethnography as it has been practiced with Hopi people and explains why tribal officials have adopted severe restrictions on research activities. These have applied to the ethnophysiography researchers who have so far failed to convince tribal officials to approve a study with the Hopi. Whiteley (1998) suggests that The reasons for indigenous resistance to cultural commodification by academic ethnography are several … but at base they are the result of the social and political estrangement of anthropology as a research-university discipline from the perspectives and situated interests of its subjects.  (p. 6)

Researchers must not only respect Indigenous knowledges but also carry out their research using ethical methods (Turk et al. in press; Turk & Mark this volume). These concerns regarding research with colonized Indigenous peoples are also highlighted by L. Smith (1999) – “The globalization of knowledge and Western culture constantly reaffirms the West’s view of itself as the centre of legitimate knowledge, the arbiter of what counts as knowledge and the source of ‘civilized’ knowledge” (p. 63) and “Most of the ‘traditional’ disciplines are grounded in cultural world views which are either antagonistic to other belief systems or have no methodology for dealing with other knowledge systems” (p. 65). Of course individual researchers working in particular contexts are not necessarily guilty of such ethnocentrism. Watson-Verran and Turnbull (1995) suggest that cross-cultural comparisons of knowledge systems are receiving renewed interest from researchers in the field of science and technology studies, “… as fresh insights are gained from the intersections of the social study of science with anthropology, postmodernism,



Exploring philosophy of place

feminism, postcolonialism, literary theory, geography, and environmentalism” (p.€115). Unfortunately, Indigenous knowledges have often previously been thought to be merely of interest within a particular anthropological context. However, such a restricted view is giving way to one which encourages a respectful examination of local knowledges within the global context. Studying representations is the key way that outsiders have access to understanding of Indigenous knowledges. Jovchelovitch (2007:â•›2) suggests that representation “is at the basis of all knowledge systems and understanding its genesis, development and realisation in social life provides the key to understanding the relationship that ties knowledge to persons, communities and lifeworlds.” Hence, language (as well as pictorial representations) provides the basis for understanding alternative worldviews, including cultural aspects of place. Indigenous worldviews are determined by physical and social environments and the day-to-day activities of members of a speech community. Gallagher (2008) discusses Wittgenstein’s notions of language, for instance: “Language is grounded on acting in particular contexts, and in the immediate reactions we have to others. The meanings of words are not the products of the linguistic system nor derived from a one-to-one correspondence to items in the world; rather they are generated in the activities in which they are used” (p. 15). Further: “The meaning of a concept is not fixed or universal, as it is dependent on its use in specific contexts, which are subject to temporal and historical change” (p. 16). Thus the Yindjibarndi word marnda can quite conveniently mean both ‘a mountain range’ and ‘a small stone’ (and indeed in colonial times ‘a coin’), with the particular meaning made clear to the listener via the context of utterance. The Yindjibarndi term bargarra refers to a large flat area, prototypically an extensive plain, usually grassy, although it can be stony or sandy, and could even have bushes. On the other hand, biyiila is a term meaning ‘in the open’ or ‘outdoors’ or ‘outside the dwelling’ or ‘a clearing in a forest’. Older speakers pointed out to the researchers (Mark & Turk 2003b) that, in the speech of younger people, bargarra is gradually replacing biyiila as a term for smaller open spaces. They said that young people felt less need to differentiate between the two terms, as they now live in town. Languages are living and mutable. It is important that Indigenous languages be studied in terms of their cultural/spiritual content as well as in the context of everyday social and practical activities. In discussing a chapter by Layton in their book, James et al. (1997:â•›7) mention that he … bids us address the distinction between representations that carry a direct reference to an external, locally situated material reality – a hole in a rock, a track, a river valley – and representations that are self-referential, which carry

57

58

Andrew G. Turk

meaning only to the extent that they make sense within the framework of a culturally specific knowledge-base. This might be a sacred site which marks the passage of a totemic creature.

Thus it is important to discuss landscape terms in the context of Indigenous peoples’ attachment to place.

7.

Indigenous attachment to ‘country’

Explorations of ‘sense of place’ span many disciplines and traditions. Tuan (1974) discussed environmental perception, attitudes and values about place and coined the term ‘Topophilia’ – “the affective bond between people and place or setting” (p.€4). He explains that “Natural environment and worldview are closely related: world view, unless it is derived from an alien culture, is necessarily constructed out of the salient elements of a people’s social and physical setting” (p. 79). The affective bond is strong if it is one’s own place. A phenomenological perspective of a dwelling place is explored by Casey (1996): There is no knowing or sensing a place except by being in that place, and to be in a place is to be in a position to perceive it. Knowledge of place is not, then, subsequent to perception – as Kant dogmatically assumed – but is ingredient in perception itself. … Local knowledge is at one with lived experience if it is indeed true that this knowledge is of the localities in which the knowing subject lives.  (p. 18)

To love a place is to know it deeply, through dwelling in that landscape. The author and his wife have owned for the last seven years about 25 ha (60 acres) of land at Porongurup, near the southwest tip of the Australian continent, and they have spent most of the last four years living there. The land is mostly covered by native forest and the author is gradually revegetating the remaining cleared areas. Even after this short association, he knows intimately all parts of the property, having walked it many times chipping weeds, planting seedlings or just enjoying the flora and fauna. He knows the variations in soil, where water runs after rain and which birds feed from which species of trees in which seasons. He is developing topophilia for this place. Of course this cannot be compared with the strength of sentiment felt by some non-indigenous Australian property owners whose family have farmed a piece of land for perhaps six generations. Some such people refuse to leave their land even after repeated droughts and financial ruin. Consider, then, Indigenous Australians, some of whom have occupied the same territories for more than 50,000 years; in excess



Exploring philosophy of place

of 2,000 generations. Their culture is imbued with associations from this immemorial past and their knowledge of, and attachment to, their own place is of an immeasurably greater magnitude. For Indigenous Australians, the term ‘country’ has a localized meaning; i.e. the area of land traditionally occupied by a particular tribe or language group (Benterrak et al. 1984; Sutton 1995). Thus ‘country’ comes with a complex set of shared rights and obligations, arising from that group’s unique attachment to that place. Davis and Prescott (1992:â•›132) summarize this connection as follows: Aboriginal people generally believe that each sociopolitical group variously described in the literature as a clan or tribe was made corporate by its birth from ancestral beings upon an area which became identifiable as a territory at the moment of birth of the group with whom the territory is identified. The acts of ancestral beings in conferring territories and giving birth to each group established for all time the identity of those groups and the boundaries of the territories with which they are identified.

Further, the specific associations with ‘country’ are unique for each individual. Within a tribe individuals will have different sorts of rights and responsibilities for different places (within the general tribal area) because of factors such as: the special country of their father and that of their mother; the place where they were born; and places where they have spent large periods of their life (Sutton 1995). Tuan (1974:â•›99–100) discusses the depth of Indigenous peoples’ association with their ‘country,’ citing Strehlow (1947:â•›30–31): Strehlow, an ethnologist who knows the Australian aborigines intimately, says of the Aranda that he “clings to his native soil with every fibre of his being …” Mountains and creeks and springs and water holes are to the Aranda not merely interesting or beautiful scenic features; they are the handiwork of ancestors from whom he himself has descended.

The people’s relationship with their country is at such a fundamental level that it constitutes the religious and philosophical basis of their being. Such a ‘situated’ view of conceptions of landscape is also fundamental to phenomenology. Heidegger (1962:â•›100) asserts this phenomenological view of nature in terms of an engaged (‘ready-to-hand’) relationship to place, in which deep meanings of landscape arise from the activities that people undertake, the uses they make of their place: The wood is a forest of timber, the mountain a quarry of rock, the river is waterpower, the wind is wind ‘in the sails.’ As the ‘environment’ is discovered, the ‘Nature’ thus discovered is encountered too. If this kind of Being as ready-tohand is disregarded, this ‘Nature’ itself can be discovered and defined simply in

59

60 Andrew G. Turk

its pure presence-at-hand. But when this happens, the Nature which ‘stirs and strives,’ which assails us and enthrals us as landscape, remains hidden.

Casey (1996:â•›28) suggests that interaction with place determines how it is designated in the local language: We designate particular places by the place terms of the culture to which we as place designators and place dwellers belong, but the places we designate are not bare sub-strates to which these terms are attached as if to an unadorned bedrock. They are named or nameable parts of the landscape of a region, its condensed and lived physiognomy.

This is akin to the basic thesis of ethnophysiography. Mohanty (1997:â•›60; emphasis in the original) contends that connection to country is not just about practical aspects of living: The lifeworld is a world of practice (of action, making and doing) and praxis (of social action, of production of goods, and distribution of goods). It would, however, be mistaken to say that these modes of acting exhaust the lifeworld in all its dimensions. For example, there are religious, aesthetic, and cultural dimensions. By virtue of these, the world as well as things in the world are presented to subjects inhabiting that world with different sorts of values – as useful, as sacred, as beautiful – all of which can be brought under the general heading “cultural.”

Malpas (1999:â•›187) explains how a person’s worldview and social environment influences their concepts of landscape: “Embedded in the physical landscape is a landscape of personal and cultural history, of social ordering and symbolism … the narratives of the land as enculturated and humanised cannot be prised away from its physical structure.” Thus it is not just practical activities which turn space into place, but also its social, cultural and religious significance. The Proposition is a feature film set in the Australian outback in the late nineteenth century. In one scene the English colonial police captain is interrogating a group of captured tribal Aboriginal men. He requests his Aboriginal police tracker to ask the men (in their language) “how long they have been hiding up in the ranges,” where a band of white outlaws is believed to be camped. The tracker translates an elder’s reply: “Captain sir, he said they don’t hide in the ranges, they live in the ranges.” The captain then requests the tracker to ask them “how long they have lived in the ranges.” The tracker says, “Them always been living in the ranges, captain sir” (Cave 2005). To the colonial policeman this landscape is a hiding place for outlaws; to the tribal Aboriginals it is their home, since time immemorial.



8.

Exploring philosophy of place

The dreaming, Tjukurrpa and gathered places

For Yindjibarndi people, their law is not just an age-old myth (from ‘when the world was soft’), but a vital part of their current existence: Biidarra Law is an infinitely rich, veiled, subtly mutable, mercurial, carnival, intimate skein of elements and influences that derive from, and speak of, the Yindjibarndi conception of creation, their country and the life within it, including living landforms and water, and which embodies and gives carriage to the relationships between the people, creatures, spirits and things that share (Rijavec 2008 – emphasis in the original) existence in the creation. 

This version of ‘The Dreaming’ is the philosophy of life for Yindjibarndi, incorporating detailed instructions for maintaining a cohesive society. For instance, it stipulates Galtharra relationships, including which of four ‘skin’ (subsection) groups should marry which other ‘skin’ group and the complex set of rights and responsibilities that each person has to specific members of their extended family. In addition, it provides the basis for interpretation of landscape as part of the law and culture (Ieramugadu Group 1995; Rijavec, Harrison & Soloman 1995). Myers (1986:â•›47, 48) discusses the meaning of ‘The Dreaming’ for a Western Desert Aboriginal group: “the constitution of the World by The Dreaming must be treated phenomenologically as a given condition of ‘what there is,’ an endowment of being and potential that defines for Pintupi the framework of human action.” He explores these issues from the phenomenological perspective of the ‘lifeworld’ of the Pintupi. He explains how the term ngurra (camp; country; place)€– designating “socialized space,” “the place where one belongs” (p. 55) – integrates land and culture with concepts of time, via The Dreaming. The relationship between the terms ‘The Dreaming’ and Tjukurrpa is complex; however, in this context, it is perhaps sufficient to think of the latter as primarily referring here to the stories/songs/dances associated with ‘dreaming tracks’ crossing the landscape, linking cultural sites. Academic understanding of these matters is (necessarily) limited: It is important to note that the core of Aboriginal philosophy and religious practice is subject to secrecy and knowledge on a “need-to-know” basis. Within Aboriginal society people are chosen as the eventual repositories of such knowledge, often over many years of proving their worth. There is much that is not known to the broader Australian society and non-Indigenous researchers such as those whose published work appears on this site, often have incomplete and sometimes puzzling information. (Grieves 2007)

61

62

Andrew G. Turk

Myers (1986:â•›12) explains how his examination of the meaning of ‘The Dreaming’ and ngurra (country), with the assistance of Pintupi informants, enabled him to “understand that the organization of people in space is itself a manifestation of what is called by some a ‘deep structure’ or an ‘inner logic’ and by others a ‘total system.’” Thus, it is impossible to adequately interpret ethnophysiography data without striving to comprehend (to the extent that an outsider ever can) the cultural structure which provides the logic of these peoples’ ‘lifeworld’ and the way that landscape is place; “As the invisible framework of this world, The Dreaming is its cosmic prototype” (p. 51). The Dreaming is for all time: It represents all that exists as deriving from a single, unchanging, timeless source. … Indeed, not only do the Dreaming narratives tell how the world came to be, but the raw material of the stories, the symbols themselves in the form of the landscape, signify the same concern on another level. Human life and being, they imply, are as permanent, enduring, and unchanging as the land itself.  (pp. 52, 53)

Casey (1996:â•›24, 25 – emphasis in the original) discusses how place gathers together physical and abstract aspects: Places gather: this I take to be a second essential trait (i.e., beyond the role of the lived body) revealed by phenomenological topoanalysis. Minimally, places gather things in their midst – where “things” connote various animate and inanimate entities. Places also gather experiences and histories, even languages and thoughts. … Third, the holding at issue in the gathering of a place reflects the layout of the local landscape, its continuous contour, even as the outlines and inlines of the things held in that place are respected. The result is not confusion of container with contained but a literal configuration in which the form of the place – for example, “mountain,” “mesa,” “gulley” – joins up with the shape of the things in it. Being in a place is being in a configurative complex of things.

Malpas (2006:â•›127 – emphasis in the original) links this ‘gathering’ to Heidegger’s notion of the relationships between places and activities: “the way in which our activity, and our orientation to things and places within that activity, is not merely determined by the end to which we are directed, but also by the structure of the spatiality in which that activity is situated.” This applies to the interconnectedness of cultural associations with topography expressed by Tjukurrpa: “the Ancestors do not move from one place to another but rather they link sites by common intentionality of place” (Swain 1993:â•›33 – emphasis in the original). Hence, each Tjukurrpa is a ‘gathering’ of particular locations into an extended place, via its narrative.



Exploring philosophy of place

The Spinifex People, from desert country several hundred kilometers south of the Pintupi, have a similar conception of The Dreaming (Cane 2002). An exploration of this can assist an understanding of the role of Tjukurrpa: “The combination of the social, political, environmental and spiritual dimensions of Tjukurrpa provides the Spinifex People with the philosophical and religious basis for understanding and defining their place within their country and the place of their country within the larger Western Desert socio-cultural bloc” (p. 85). Thus the Tjukurrpa of a particular group provides the basis for their mode of being in their particular place. However, much work is necessary to reach an understanding of how this meshes with phenomenology, as expressed within European philosophy. As Cane (2002:â•›86) points out: “No non-Aboriginal person has done enough systematic field work with the Spinifex People to understand the geographic intricacy, sanctity and philosophical depth of Spinifex Tjukurrpa other than to appreciate that there is a lot of it, that is extremely well known by them and that it is of the utmost importance to them.” Hence, ethnophysiography studies of assemblages of landscape elements can contribute to our understanding of social and linguistic aspects of place. This relates to generic landscape terms and also place names, which sometimes refer to groups of features (Hercus, Hodges & Simpson 2002; Sutton 1995). The Indigenous senses of place discussed in this chapter are impressive, deriving in some cases from tens of thousands of years occupation of the same territory and sometimes not much diluted by other cultural influences. However, it is important to note that places are of great significance to people of all cultures and these strong cases should serve to enliven our search for understanding of the phenomenon in general. Basso (1996:â•›87) asserts the urgency of ethnographic studies of cultural constructions of sense of place: We should begin to explore them with all deliberate speed, and not, I would emphasize, solely for the purpose of enlarging our knowledge of particular social groups. For as surely as place is an elemental existential fact, sense of place is a universal experiential genre, and therefore, as more and more work gets done, it may be found to exhibit transcultural qualities.

Ethnophysiography research serves this purpose.

9.

Towards philosophical reconciliation

In Australia there is a possibility for a reconciliation between versions of philosophy – Indigenous and European – as part of a broader movement of respectful cohabitation of the continent by original inhabitants, descendents of colonizers

63

64 Andrew G. Turk

and recent immigrants. This could provide an intellectual framework for a more informed dialogue about how to address the many problems inherent in practical aspects of current Indigenous Australian life. Muecke (2004) explores what the term ‘Aboriginal Philosophy’ might mean, especially its relationship to ‘Aboriginal Religion,’ and whether such terms have real meaning within aboriginal culture itself. He refutes Stanner’s (1965) claim that because there is no tradition of intellectual detachment there can be no ‘Aboriginal Philosophy’: “Stanner’s argument about why there is an Aboriginal religion, but not a philosophy, is based on the need for stated ‘axioms’ to be non-ritual and non-mythic, that is, to be ‘scientific.’ … The problem is how to recognise the ‘potencies’ Aboriginal people understand, since Western categories flip between the theistic and the scientific” (p. 115). An Aboriginal philosophy will of necessity take a different form from Western philosophy: “Dreaming stories, ritual acts and experiences are the forms of expression of indigenous Australian philosophies. By ‘expression’ I do not mean the representation or construction of some pregiven content or essence, I mean the expressions are the dreaming” (p. 21). An Aboriginal philosophy will also be based on the physical landscape, experienced as a dwelling place. Brian Murton (this volume) discusses how Maori narratives link specific people to specific places. In discussing an example narrative he explains: The first line identifies a feature, a mountain or hill, the second line a river, and this unvarying form emphasizes the significance of these features. The third and fourth lines identify the people. More can be added to such sayings, such as a canoe, an ancestor, a marae (a complex of buildings, including a meeting house, belonging to a related group of people). They are identity axioms linking land and people into a whole in such a way as to make them inseparable.

Thus, in a sense, the Indigenous philosophy does have axioms, but they are about associations between people and place, the core of their world view. An Aboriginal philosophy is ‘in place’ and based on imperatives of the particular physical landscape in which it was formed and has its meaning. It is this author’s intention to develop a set of ideas (a framework perhaps) based on phenomenology which will permit effective discussion of ethnophysiography in the context of Indigenous worldviews. This is not to say that this line of reasoning can in any sense completely contain or subsume, imitate or appropriate those worldviews. Rather the desire is that it might be sufficiently commensurate to enable discussion of Indigenous concepts of landscape (within the Western philosophical framework) without doing serious injustice to those worldviews. This involves engagement with Indigenous communities in a manner based on mutual respect and a desire to work together to reach common outcomes, which



Exploring philosophy of place

have been discussed and agreed to be important and valuable for both the academic and local communities. An effective approach to philosophical reconciliation can only be accomplished in terms of landscape as place. Clearly there is much work needed to develop an effective approach to this daunting enterprise; however, since topography is primary, ethnophysiography is likely to be of assistance.

10.

Methods for using phenomenology in ethnophysiography

The discussion above, and in Turk (2007), has established that there is a strong overlap between the concerns of phenomenology and ethnophysiography in the context of understanding the concept of place. Aspects of place have been explored indicating how philosophers, geographers, anthropologists, and writers from other disciplines have linked them to phenomenology. However, the author needs to carry out much more detailed research on the fundamentals of phenomenology before he can determine exactly how it can operate as an overarching transdisciplinary paradigm for ethnophysiography. This is being undertaken via examination of the writings of key philosophers who have contributed to phenomenology, including: Edmund Husserl, Martin Heidegger, Hans-Georg Gadamer, Jürgen Habermas, Maurice Merleau-Ponty, Emmanuel Lévinas, Pierre Bourdieu, Michel Foucault and Jan Patocka. This includes study of both primary texts by these philosophers and also secondary texts interpreting their work in the context of place (e.g. by Edward Casey and Jeff Malpas). Additionally, works by writers from other disciplines utilizing phenomenology are being examined (e.g. texts by Alfred Schütz). Investigation of the potential role of phenomenology in ethnophysiography is, however, unlikely to be advanced without engaging directly with the methods of inquiry utilized in phenomenology: Without doing phenomenology, it may be practically impossible to understand phenomenology … Phenomenology, in the first instance, is like an investigative science, an essential component of which is experiment. Phenomenology is experimental and its experiments are conducted according to a carefully worked out set of controls and methods.  (Ihde 1986:â•›14 – emphasis in the original)

The set of classical phenomenological methods suggested by D. Smith (2007) seems directly applicable to research in ethnophysiography:

. See also Maso (2001).

65

66 Andrew G. Turk

(1) We describe a type of experience just as we find it in our own (past) experience. Thus, Husserl and Merleau-Ponty spoke of pure description of lived experience. (2) We interpret a type of experience by relating it to relevant features of context. In this vein, Heidegger and his followers spoke of hermeneutics, the art of interpretation in context, especially social and linguistic context. (3) We analyze the form of a type of experience. In the end, all the classical phenomenologists practiced analysis of experience, factoring out notable features for further elaboration.

Analyzing ethnophysiography case studies in this manner could be considered a Grounded Theory Methodology approach (Charmaz 2006; Charmaz & Mitchell 2001; Glaser 1994; Glaser & Strauss 1967), seeking to develop ethnophysiography theory from carefully analyzed data. The initial work on the Ethnophysiography Descriptive Model (discussed in Turk et al. this volume) represents an initial informal application of this approach. Orleans (2008) discusses the role of phenomenology in investigations of the relationship between individual consciousness and social life: The central task in social phenomenology is to demonstrate the reciprocal interactions among the processes of human action, situational structuring, and reality construction. Rather than contending that any aspect is a causal factor, phenomenology views all dimensions as constitutive of all others. Phenomenologists use the term reflexivity to characterize the way in which constituent dimensions serve as both foundation and consequence of all human projects. The task of phenomenology, then, is to make manifest the incessant tangle or reflexivity of action, situation, and reality in the various modes of being in the world.

Considerably more research is needed to verify the suggested approaches and to achieve an appropriate synergistic interplay between phenomenological and empiricist approaches to ethnophysiography theory and practice.

11.

Potential for ethnophysiography research to inform phenomenology

The project discussed in this chapter sits squarely within the tradition of philosophical concepts informing research in other disciplines. The author is also investigating the possibility that ethnophysiography research may enhance phenomenology. As it is early days in this endeavour, the following examples of how this may occur are merely exploratory.



Exploring philosophy of place

From the author’s own experience of many interdisciplinary projects, and from review of relevant literature, it is clear that transdisciplinarity is very difficult to achieve. If this research project can demonstrate that phenomenology can operate effectively as a transdisciplinary paradigm for ethnophysiography, this will be evidence of the power and utility of this philosophy. It would also be coherent with one of Husserl’s original objectives: to establish phenomenology as the integrative basis for all science. Kockelmans (1994) notes that phenomenology seeks to provide the “absolute” foundation for other sciences and “pertain to every science and every form of knowledge” (p. 11) by attempting “to stay away from all free-floating constructions, accidental findings, unexamined preconceptions, and arbitrary prejudices” (p. 12). Patocka (1996:â•›1) suggests that, through phenomenology, the rigor of science does not need to be divorced from spiritual and cultural aspects of being: What makes Husserl’s approach distinctive is that (1) it seeks to be a rigorous science and (2) it singles out such rigor not only as one instance but as the central, most important, and profound access to meaning; as such science can claim a fundamental, crucial significance for human existence. Science ought to and can provide human lives with a “spiritual meaning,” the content and aim of life we need in order to be truly at home, at one with ourselves, with our life, and with our world.

Heidegger (1962:â•›76) suggests engagement with Indigenous lifeworlds, although from what seems a rather Eurocentric perspective: To orient the analysis of Dasein towards the ‘life of primitive peoples’ can have positive significance [Bedeutung] as a method because ‘primitive phenomena’ are often less concealed and less complicated by extensive self-interpretation on the part of the Dasein in question. Primitive Dasein often speaks to us more directly in terms of a primoidal absorption in ‘phenomena’ (taken in a pre-phenomenological sense). A way of conceiving things which seems, perhaps, rather clumsy and crude from our standpoint, can be positively helpful in bringing out the ontological structures of phenomena in a genuine way.

He goes on to caution that ethnology “operates with definite preliminary conceptions and interpretations of human Dasein in general, even in first ‘receiving’ its material, and in sifting it and working it up” (Heidegger 1962:â•›76). Hence, ethnographers should base their studies on phenomenology “recapitulating what has already been ontically discovered, and by purifying it in a way which is ontologically more transparent.” Apparently Husserl (and presumably, through him, Heidegger) was influenced to adopt an inappropriate view of non-European ‘primitive’ societies by

67

68 Andrew G. Turk

the second-hand anthropology of French philosopher Lucien Levy-Bruhl (1857– 1939). Levy-Bruhl did not himself collect ethnographic data, but collated it for the purpose of developing his thesis that archaic thought is prelogical and fundamentally different from modern thought. It could be contended that these foundation views of phenomenology can be more adequately critiqued via ethnophysiography explicating Indigenous peoples’ relationships to places and the meanings of the terminology that they employ for aspects of landscape. As discussed above, dwelling place is a central theme of Indigenous philosophy. Exploring key aspects of that philosophy, via ethnophysiography, could lead to a deeper understanding of ‘intentionality,’ one of the main tenets of phenomenology. In phenomenology ‘intentionality’ is used in a specific sense to indicate that which our consciousness is directed at; a physical or mental object, which may not necessarily exist (e.g. fear of a ‘bogey-man’) (Mathews 2006). Intentionality is not capable of being explained by a series of psychological facts, since the totality of its content, structure and variables cannot be effectively made explicit. However, by examining the way landscape is treated in language for a wide variety of cultures, the range of examples of intentionality can be extended, especially regarding spiritual aspects of Indigenous sense of place. Ethnophysiography research can also help explicate issues of universalism central to debates within phenomenology. The purpose of the Ethnophysiography Descriptive Model (discussed by Turk et al. this volume) is to investigate the possibility that there are universal factors behind the individual cultural differences observed in the ways landscape is represented in language, although it is not suggested that deterministic causation is at play. Van Manen (1990:â•›10) suggests: The essence of a phenomenon is a universal which can be described through a study of the structure that governs the instances of particular manifestations of the essence of that phenomenon. In other words, phenomenology is the systematic attempt to uncover and describe the structures, the internal meaning structures, of lived experience. A universal or essence may only be intuited or grasped through a study of the particular or instances as they are encountered in lived experience.

Whether these particular objectives can be achieved only time will tell. Further research may also reveal other potential benefits (and problems) of utilizing phenomenology with ethnophysiography.

. See, for instance: http://www.bookrags.com/biography/lucien-levy-bruhl/.



12.

Exploring philosophy of place

Conclusions

In this chapter it has only been possible to scratch the surface of the potential interactions between phenomenology and ethnophysiography. It has, however, provided an initial indication that a fruitful relationship between the two is possible. For Indigenous Australians, such as the Yindjibarndi, Aranda, Pintupi and Spinifex People, land and culture are integrated with concepts of time via The Dreaming, the ever present flux of ideas connecting people with their own place and informing both practical and cultural aspects of their ‘lifeworld.’ Examination of Indigenous concepts of place in Australia – and elsewhere – can provide a rich vein of phenomenological understanding. Since such concepts are intricately tied with landscape, the methods of ethnophysiography can assist in unravelling these ideas. However, to accomplish this synergy, in the context of Australian Indigenous cultures, it is necessary to both understand and respect concepts such as The Dreaming and Tjukurrpa. Hence, development of a synergistic relationship between phenomenology and ethnophysiography entails a meaningful dialogue between European philosophy and Indigenous cultural concepts of place.

Acknowledgments The author acknowledges the contribution of his collaborators David Mark, David Stea and Carmelita Topaha in the ethnophysiography research projects, as well as assistance from philosopher Bruce Janz and linguist Niclas Burenhult in the preparation of this chapter. The role of Yindjibarndi and Navajo informants is also acknowledged. Many members of the Roebourne community, the cultural organization Juluwarlu and the Pilbara Aboriginal Language Centre (Wangka Maya) provided invaluable assistance regarding the Yindjibarndi language. Funding support by the US National Science Foundation (grants BCS-0423075 and BCS-0423023) and from Murdoch University for the ethnophysiography projects is gratefully acknowledged.

References Basso, Keith H. 1996. Wisdom sits in places: Notes on a Western Apache landscape. In Senses of Place, Steven Feld & Keith H. Basso (eds), 53–90. Santa Fe NM: School of American Research Press.

69

70 Andrew G. Turk

Benterrak, Krim, Muecke, Stephen & Roe, Paddy. 1984. Reading the Country. Fremantle, Western Australia: Fremantle Arts Centre Press. Booth, Michael, Rodgers, Stephen & AgInsight. 2000. Interdisciplinary Research Methodologies in Natural Resource Management [Report to LWRRDS Social and Institutional Research Program]. Institute for Sustainability and Technology Policy, Murdoch University, Western Australia . Cane, Scott. 2002. Pila Nguru: The Spinifex People. Fremantle, Western Australia: Fremantle Arts Centre Press. Casey, Edward S. 1996. How to get from space to place in a fairly short stretch of time: phenomenological prolegomena. In Senses of Place, Steven Feld & Keith S. Basso (eds), 13–52. Santa Fe NM: School of American Research Press. Cave, Nick (screenwriter). 2005. The Proposition. Feature film, directed by John Hillcoat. UK Film Council. Charmaz, Kathy. 2006. Constructing Grounded Theory: A Practical Guide through Qualitative Analysis. London: Sage. Charmaz, Kathy & Mitchell, Richard G. 2001. Grounded theory in ethnography. In Handbook of Ethnography, Paul Atkinson, Amanda Coffey, Sara Delamont, John Lofland & Lyn Lofland (eds), 160–176. London: Sage. Davis, Stephen & Prescott, J. R. V. 1992. Aboriginal Frontiers and Boundaries in Australia. Melbourne: Melbourne University Press. Feld, Steven & Basso, Keith H. (eds). 1996. Senses of Place. Santa Fe NM: School of American Research Press. Fuchs, Wolfgang W. 1976. Phenomenology and the Metaphysics of Presence: An Essay in the Philosophy of Edmund Husserl. The Hague: Martinus Nijhoff. Gallagher, Shaun. 2008. Philosophical antecedents to situated cognition. In Cambridge Handbook of Situated Cognition, Philip Robbins & Murat Aydede (eds), 35–52. Cambridge: CUP. Glaser, Barney G. (ed.). 1994. More Grounded Theory Methodology: A Reader. Mill Valley CA: Sociology Press. Glaser, Barney G. & Strauss, Anselm L. 1967. The Discovery of Grounded Theory. Hawthorne NY: Aldine de Gruyter. Grieves, Vicki. 2007. Aboriginal Philosophy. (22 November 2007). Gurwitsch, Aron. 1957. The last work of Edmund Husserl. Philosophy and Phenomenological Research 17(3): 370–398. Heidegger, Martin. 1962. Being and Time. New York NY: Harper and Row. Hercus, Luise, Hodges, Flavia & Simpson, Jane (eds). 2002. The Land is a Map: Placenames of Indigenous Origin in Australia. Canberra: Pandanus Books. Ieramugadu Group. 1995. Know the Song, Know the Country: The Ngarda-Ngali Story of Culture and History in the Roebourne District. Roebourne, Western Australia: Ieramugadu Group. Ihde, Don. 1986. Experimental Phenomenology: An Introduction. Albany NY: State University of New York Press. Ihde, Don & Zaner, Richard M. (eds). 1975. Dialogues in Phenomenology. The Hague: Martinus Nijhoff. James, Allison, Hockey, Jenny & Dawson, Andrew (eds). 1997. After Writing Culture: Epistemology and Praxis in Contemporary Anthropology [ASA Monograph 34]. London: Routledge. Jovchelovitch, Sandra. 2007. Knowledge in Context: Representations, Community and Culture. London: Routledge.



Exploring philosophy of place

Kockelmans, Joseph J. 1994. Edmund Husserl’s Phenomenology. West Lafayette IN: Purdue University Press. Malpas, Jeff E. 1999. Place and Experience: A Philosophical Topography. Cambridge: CUP. Malpas, Jeff E. 2006. Heidegger’s Topology: Being, Place, World. Cambridge MA: The MIT Press. Mark, David M. & Turk, Andrew G. 2003a. Ethnophysiography. Paper presented at the Workshop on Spatial and Geographic Ontologies, 23 September, 2003 (prior to COSIT03), in Ittingen, Switzerland. Mark, David M. & Turk, Andrew G. 2003b. Landscape categories in Yindjibarndi: ontology, environment, and language. In Spatial Information Theory: Foundations of Geographic Information Science [Lecture Notes in Computer Science 2825], Werner Kuhn, Michael F. Worboys & Sabine Timpf (eds), 28–45. Berlin: Springer. Maso, Ilja. 2001. Phenomenology and Ethnography. In Handbook of Ethnography, Paul Atkinson, Amanda Coffey, Sara Delamont, John Lofland & Lyn Lofland (eds), 136–144. London: Sage. Mathews, Eric. 2006. Merleau-Ponty: A Guide for the Perplexed. London: Continuum. Mohanty, Jitendranath N. 1997. Phenomenology: Between Essentialism and Transcendental Philosophy. Evanston IL: Northwestern University Press. Moran, Dermot & Embree, Lester E. (eds). 2004. Phenomenology: Critical Concepts in Philosophy. London: Routledge. Muecke, Stephen. 2004. Ancient and Modern: Time, Culture and Indigenous Philosophy. Sydney: UNSW Press. Myers, Fred R. 1986. Pintupi Country, Pintupi Self: Sentiment, Place, and Politics among Western Desert Aborigines. Berkeley CA: University of California Press. Nicolescu, Basarab. 2008. Transdisciplinarity as methodological framework for going beyond the science-religion debate. The Global Spiral, 24 May 2007, (December 2008). Nowotny, Helga. 2003. The potential of transdisciplinarity. Rethinking Interdisciplinarity, 1 May 2003 <www.interdiscplines.org/interdisciplinarity/papers/5>. Orleans, Myron. 2008. Phenomenology. (accessed June 2008). Patocka, Jan. 1996. An Introduction to Husserl’s Phenomenology. Peru IL: Open Court Publishing. Psathas, George. 1977. Ethnomethodology as a phenomenological approach in the social sciences. In Interdisciplinary Phenomenology, Don Ihde & Richard M. Zaner (eds), 73–98. The Hague: Martinus Nijhoff. Rijavec, Frank. 2008. Personal communication. Draft PhD dissertation, Murdoch University, Western Australia. Rijavec, Frank, Harrison, Noelene & Soloman, Roger. 1995. Exile and the Kingdom [documentary film]. Roebourne, Western Australia: Ieramugadu Group & Film Australia. Schunn, Christian D., Paulus, Paul B., Cagan, Jonathan & Wood, Kristin. 2006. Final Report from the NSF Innovation and Discovery Workshop: The Scientific Basis of Individual and Team Innovation and Discovery. August 2006. (23 August 2007). Schütz, Alfred. 1977. Husserl and his influence on me. In Interdisciplinary Phenomenology, Don Ihde & Richard M. Zaner (eds), 124–129. The Hague: Martinus Nijhoff.

71

72

Andrew G. Turk

Schütz, Alfred & Luckmann, Thomas. 1973. The Structures of the Life-World. Evanston IL: Northwestern University Press. Seamon, David & Mugerauer, Robert. 1985. Dwelling, Place and Environment: Towards a Phenomenology of Person and World. Dordrecht: Martinus Nijhoff. Smith, David W. 2007. Phenomenology. Stanford Encyclopaedia of Philosophy. (20 April 2007). Smith, Linda T. 1999. Decolonizing Methodologies: Research and Indigenous Peoples. Dunedin NZ: University of Otago Press. Stanner, William E. H. 1965. Religion, Tokenism and Symbolism. Reprinted in Religion in Aboriginal Australia: An Anthology, Max Charlesworth, Howard Murphy, Diane Bell & Kenneth Maddock (eds, 1984), 137–172. Queensland: University of Queensland Press. Strehlow, Theodor G. H. 1947. Aranda Traditions. Melbourne: Melbourne University Press. Sutton, Peter. 1995. Country: Aboriginal Boundaries and Land Ownership in Australia [Aboriginal History Monograph 3]. Canberra: Australian National University. Swain, Tony. 1993. A Place for Strangers: Towards a history of Australian Aboriginal Being. Cambridge: CUP. Tuan, Yi-Fu. 1974. Topophilia: A Study of Environmental Perception, Attitudes and Values. Englewood Cliffs NJ: Prentice Hall. Turk, Andrew G. 2007. A phenomenology basis for trans-disciplinary research in ethnophysiography. Proceedings of the 26th International Human Science Research Conference, Rovereto, Italy. CD-ROM. Turk, Andrew, Mark, David, O’Meara, Carolyn & Stea, David. In press. Geography – documenting terms for landscape features. In Oxford Handbook of Linguistic Fieldwork, Nicholas Thieberger (ed.). Oxford: OUP. Van Manen, Max. 1990. Researching Lived Experience: Human Science for an Action Sensitive Pedagogy. London ON: Althouse Press. Watson-Verran, Helen & Turnbull, David. 1995. Science and other indigenous knowledge systems. In Handbook of Science and Technology Studies, Sheila Jasanoff, Gerald E. Markle, James C. Petersen & Trevor Pinch (eds), 115–139. London: Sage. Whiteley, Peter M. 1998. Rethinking Hopi Ethnography. Washington DC: Smithsonian Institute Press.

chapter 4

Embedded in place ‘Mirror knowledge’ and ‘simultaneous landscapes’ among Māori Brian Murton

This paper explores the understanding that Māori, the indigenous people of Aotearoa/New Zealand, have of “landscape”. I argue that although Māori in the past did “gaze” at their surroundings, and did have visual forms of representation, the concept of “landscape” as it is commonly used in Western scholarly and popular representation is inappropriate. What replaces the profound visuality of the West for Māori is language, especially sound and speech. Māori represent the world primarily through the act of naming, in which naming places becomes an integral way of actively engaging perceptually with the animate and inanimate world. Māori “imagine” named places as “simultaneous landscapes” reflecting cosmology, ancestors, history and everyday life.

1.

Introduction

In this paper I explore the understanding that Māori, the indigenous people of Aotearoa/New Zealand, have of “landscape”. The word “explore” is used purposely because like any other exploration it will provide a map of the territory, but different explorers will notice different aspects of a territory and will produce different maps. It will seek to demonstrate that the approach to landscape in European thought which most closely approximates Māori understanding of the landscape emerges from phenomenology. Such an approach, I suggest, makes available aspects of Māori understanding of the landscape concept that other European approaches (for example, those deriving from positivism, structuralism, and post-modernism) do not provide. The people now collectively called Māori are the indigenous Polynesian inhabitants of Aotearoa, the Māori name for the islands of New Zealand (the European name) as a whole. The word Māori means “normal,” “natural,” or “ordinary,”

74

Brian Murton

and in oral tradition the word distinguishes ordinary mortal human beings from deities and spirits. It has only been used to refer to Māori collectively since the early nineteenth century. The ancestors of the Māori came from Eastern Polynesia around 700–800 years ago, and linguistically, the various dialects of the Māori language (te reo māori or simple te reo) belong to the Tahitic subgroup (which also includes Tahitian, Tuamotuan, Rarotongan) of the Eastern Polynesian subgroup within the Eastern Oceanic branch of the Austronesian language family (Cablitz 2008:â•›204). I have italicized Māori concepts and place names, primarily to demonstrate, as did Wiri (2001), that the Māori and English languages have different epistemological foundations. I am of part Māori descent (Te Popoto and Te Honihoni of Ngā Puhi, Te Ihutai of Te Rarawa, with whānaunga (relatives, family) among Te Aitanga a Mahāki) and was involved in research during the 1990s and early 2000s to support a number of claims to the Waitangi Tribunal (the body established by legislation in 1975 to provide Māori with judicial means to address breaches by the Crown of the Treaty of Waitangi of 1840; see Ward 1999; Byrnes 2004; Belgrave et al. 2005), as well as a collaborative project dealing with landscape change in the Bay of Islands in northern New Zealand, 1769–1840. An enormous difficulty encountered in this work was the very different understandings of key concepts and how they should be represented, both among researchers, who consist of university academics of both Māori and non-Māori ancestry, and members of the tribal groups involved. This was especially the case with the term “landscape,” which, although a quintessentially European term, is widely used throughout the world today, even among peoples whose homelands have been settled by Europeans over the past several hundred years, especially where English is the dominant language. In spending many hours listening to and talking with people, especially elders, it became very apparent that they continue to have a very different understanding of what is now referred to as “landscape” from the understanding held by academic disciplines and people of European descent generally. Fundamentally, in this paper I argue that although Māori “saw,” and still “see,” things in the world (“landscape”), things became “visible” primarily through language, and that a European visual approach is subversive of non-Western modes of knowledge, its acquisition, revelation, and articulation. Section 2 of the paper is an exploration of the “landscape problematic,” including a review of the salient aspects of the landscape concept as it emerged in Europe, a discussion of whether it can be considered a universal concept, an examination of whether Māori “gazed” at their surroundings, and the presentation of an alternate phenomenological perspective on landscape. In Section 3 I present information on how Māori know the world about them. I first outline an approach which, while based in a European epistemological framework, enables



Embedded in place

me to weave into the Māori world. I then identify and discuss three dimensions of Māori knowledge critical to a phenomenological approach to “landscape” as the “face of place”: understandings of time, genealogy as a way of knowing things, and the importance of the creative word. In Section 4 I examine how Māori places can be constituted and how people and land are inseparable. The last major section, Section 5, focuses on named places. I briefly overview past studies of Māori place names, and then give examples of terms for “landscape” features. I then outline how Māori “imagine” named places, and finally present a limited number of selected examples from one area of Aotearoa/New Zealand to illustrate how named places are the landscape.

2.

The landscape problematic

2.1

Landscape: An ambiguous concept

Today landscape is a very loosely used term. Once used primarily in art, art history, landscape architecture, and geography, it now has resonance well beyond academia, and is part of our everyday vocabulary. The word has of late become a well-worn metaphor, and there are a confusing array of uses of the term, ranging from, for example, the “legal landscape,” the “religious landscape,” “landscape ecology,” the “leisure landscape,” the “cultural landscape,” and even the “changing landscape of bladder augmentation” (Creswell 2003:â•›269). Because “landscape” is used by so many different people for such a variety of purposes, it is invariably an ambiguous term. There are problems of translation between fields and often uncertainties of exact meaning within any one, as well as between languages, and it is not surprising that there is not a clean and clear definition that satisfies all. A number of scholars have reviewed how the term “landscape” has been used in Europe since it became a word in common usage from the fifteenth and sixteenth centuries, especially art historians and historians (for example, Clark 1956; Appleton 1975; Schama 1995), philosophers (Casey 2002), but also geographers, notably Cosgrove (1998) and Olwig (2002). These writers have emphasized that the idea of landscape is a cultural concept that emerged in Europe and involves a particular way of seeing; a way in which some Europeans have represented to themselves and to others the world about them and their relationships with it (Cosgrove 1998:â•›1). Most of the commentators on the landscape idea point out that only since the late nineteenth century has there been any technical or academic discussion of the meaning of landscape (Relph 1981:â•›23). Our current notions of landscape derive from a complicated array of ideas, linked to many of the key concepts of Western thought. These include what

75

76

Brian Murton

Martin Jay (1992:â•›178–195) calls a “resolute ocularcentrism” in which the eye is the primary way of sensing the world, the basis for the development of the methods of perspective in art, which helped define and delimit relations between observer and the world, and helped create an “optic” for framing the “objective science” emerging in the seventeenth century and the separation of humanity from nature. It also involves the subject-object and mind-body separation emerging from the philosophy of René Descartes, in which the mind contains two elements, representations of the external world and the scanning activity of what Macnaughten and Urry (1998:â•›110) call the “Inner Eye”. In this abstract and disembodied Western approach the starting point is the self-centered subject, who confronts a domain of isolable objects, entities, that a perceiver first has to make sense of, rendering them intelligible by categorizing them and assigning to them meanings or functions, before they can be made available for use (Ingold 2000:â•›168–169). What is more, Martin Heidegger (1977:â•›115–154), in his essay “The age of the world picture,” contends that the “metaphysical ground” of modern science is the idea of the world conceived and grasped as a picture. Within this modern optic, the “certainty of truth” is made to turn on the need to establish a distance between observer and observed. Only then may order be discovered and re-presented. The key binary of mind/matter (the former irreducible, impenetrable, and disembodied and the latter entirely knowable and manipulable), which underlies this perspective, is one of the many separations (other ontological binaries include nature/culture, individual/society, space/place) which are much less emphasized within Indigenous knowledge systems. Relph (1981:â•›28–29) notes that just as mind had been separated from matter, so the viewer had been separated from landscape.

2.2

Landscape: A universal concept?

A number of commentators have urged caution when using the landscape idea in relationship to non-European societies. Corner (1999:â•›5), for example, observes: … the landscape idea is neither universally shared nor manifested in the same way across cultures and time; its meaning and value, together with its physical and formal characteristics, are not fixed. To assume that every society shares an American, English, or French view of landscape, or that other societies possess any version of landscape at all, is to wrongly impose on other cultures one’s own image. Indeed, there have been societies and times wherein the notion of landscape simply did not exist.



Embedded in place

Corner (1999:â•›6) also states that while every society has historically been aware of an environment, most did not develop the idea of a unified scene, a visual landscape in the sense that emerged among educated Europeans. Burenhult and Levinson (2008:â•›135–150) raise the issue of whether landscape is a cross-linguistically distinct and easily identifiable semantic domain, like kinship, color, or folk biology. They note that European cultures identify a category “landscape,” but the results of collaborative work on the relationships between language and landscape carried out by the Language and Cognition Group at the Max Planck Institute for Psycholinguistics suggest that caution is required when applying it to other cultures. In this work, the term “landscape” is taken to refer to features and entities of the environment. Burenhult and Levinson (2008:â•›148) conclude that there is a great deal of plasticity in how language models the earth and in what is considered to be the essence of its features, and they warn that there are no unique beginner terms for landscape as a whole, and that notions of landscape are often intertwined with notions of settlement, migration, resources, and ethnicity (Burenhult & Levinson 2008:â•›143). The study of these issues is what Mark and Turk (2003, 2004) have labeled “ethnophysiography”.

2.3

Did Māori conceptualize “landscape”?

There is evidence that Māori did have a notion of “landscape” in a visual sense, although not as a dominant mode of representation. This emerges from interpretations of early nineteenth-century European paintings of landscape containing Māori, from observations by European travelers and early ethnographers, from other types of Māori visual material, and from language.

Interpreting landscape painting W. T. J. Mitchell (2002:â•›22–27) uses a painting (ca. 1827) by Augustus Earle, “Distant View of the Bay of Islands”, which has a carved Māori figure in the right foreground, to argue that the figure is intended to reflect the Māori experience and representation of landscape. On this basis, Mitchell argues that Māori did place importance on looking at the land, and that Earle uses the figure to represent the Māori gaze as a presence in the landscape. Mitchell (2002:â•›26) concludes that Earle’s painting suggests that Māori did look at the land: The Maori statue indicates at a minimum that they erect a statue to keep surveillance over a place. Nor is this surveillance confined only to the carved figure. The Maori bearer on the left seems to be hesitating as he walks, turning to the side to scan the tabooed territory, while raising his war club slightly to ward off a potential threat.

77

78

Brian Murton

Ethnographic commentary By the late nineteenth century it was well known by European commentators that Māori obtained overviews of the land from high points called taumata (“gaze on, let the eyes rest on, resting place on a hill, brow of a hill”). Elsdon Best (1942 reprinted 1977:â•›28), who began to carry out ethnographic research among Tūhoe in the late nineteenth and early twentieth century, described the role of taumata thusly: The old type of Māori much appreciated a commanding taumata from which a fine view of his tribal lands might be obtained, and when resting at such places I have frequently heard them crooning old songs that referred to long past occurrences at the places they looked upon.

Of significance is that narratives about people and past events were remembered when places were viewed: when a place was named, hearers thought about narratives attached to the name. Many place names, incidentally, incorporate the word taumata.

Māori visual representation Although Māori do not appear to have forms of representational painting before the nineteenth century, they did have a number of visual methods of representing the world around them. Roger Neich (1994:â•›16) notes that at the time of early European contact oral forms of representation were the most important, but visual forms, such as dance, weaving, tattoo, woodcarving and sculpture in stone and bone, were significant. There were also several types of painting, including face and body painting, the painting of carvings, the painting of canoe hulls and architectural timbers, parietal painting in caves and rock shelters, and most significantly the abstract curvilinear patterns painted on house rafters, on monuments and mausoleums, on paddles and on the underside of canoe prows, known as kōwhaiwhai (Neich 1994:â•›17). A number of these visual forms represented ancestors and genealogically related entities, and could elicit oral descriptions of places and events. Linguistic evidence The Māori language does contain several words which provide connotations of “landscape”. These include tiro or titiro, which as a verb can mean “look,” “survey,” “view,” “examine,” and in the noun form, tirohanga, a “view,” “sight” or “aspect,” all of which seem to be of some antiquity (Williams 1844 reprinted 1992:â•›424). A phrase tirohanga paruhi (panorama), on the other hand, seems to be of more recent origin. A number of phrases (for example, takoto o whenua, literally “the lie of



Embedded in place

the land,” or tohu o whenua, literally “marks on the land”), when used in the appropriate context, are used to refer to landmarks. There are several publications which use these phrases in their titles, notably F. L. Phillips’s (1989) Nga Tohu a Tainui, the New Zealand Historical Atlas: Ko Papatuanuku e Takoto Nei (McKinnon 1997), and Taimoana Tūroa’s Te Takoto o Te Whenua o Hauraki (2000). There is no question that Māori in the past did view their environment, gaze upon it, and they may have even had a number of terms for landscape as a whole. But the question remains how they primarily represented it. Edward Casey (2002:â•›xvi) notes that what is problematic about representation is not found in the fact that it happens, but in how it happens. He points out that representing landscape occurs in line and in paint, in gesture and in photograph, in image and in word (Casey 2002:â•›xiv), and that it is integral to the perception of landscape itself, indeed, part of its being and essential to its manifestation (Casey 2002:â•›xv).

2.4

An alternate perspective on landscape

More than a decade ago Casey (1996:â•›13–52) noted that landscape, which humans could only grace by being bodily there, could be construed as the “face of place, its expressive facies or sensuous surface”. This phenomenological expression of landscape stands in contrast with the dominantly visual notions of landscape in Western understandings, whether it be in the popular mind or in the writings of a range of scholars in different disciplines, where the “gaze” is at the center, where the landscape, however it is interpreted, is an assemblage of objects marked by their existence at a distance. It is the contention of this paper that, as Gregory Cajete (2000:â•›23, 25) notes, phenomenology parallels the approach of Indigenous understandings of the world. He argues that Indigenous knowledge is rooted in the soil of direct physical and perceptual experience of the earth. This process of intersubjectivity is based on the notion that there is a primal affinity between the human body and other bodies of the natural world. In this perspective, lived experience of this “lifeworld” is the ultimate source of human knowledge and meaning. Central to this participation with the world is the body, as the source of thinking, sensing, acting (Cajete 2000:â•›185). A number of other Native American scholars, including Vine Deloria and Daniel Wildcat (2001), Leroy Little Bear (2000:â•›77–85), and N. Scott Momaday (1974) have also identified phenomenological approaches as being aligned with Indigenous thinking. The phenomenological work carried out in Indigenous societies by non-Indigenous scholars (for example that of Myers 1986; Weiner 1991, 2001; Basso 1988, 1996a, 1996b; Thornton 2008) has drawn on the concept of dwelling as articulated by Martin Heidegger (1971:â•›141–160), a concept that assigns importance

79

80 Brian Murton

to the forms of consciousness with which individuals perceive and apprehend geographical space (Basso 1996a:â•›106). It also uses ideas from Maurice MerleauPonty (1962), especially that of the “body-subject.” There are several dimensions of a dwelling approach which have been identified by recent commentators, including the philosophers Edward Casey (1993, 1996, 1997, 2001, 2002) and Jeff Malpas (1999, 2006), as important to understanding place. First, the human physical constitution as embodied is seen as the ultimate arbiter of human interaction with the environment (Thornton 2008:â•›24). As Edward Casey (1993:â•›48) puts it: My body continually takes me into place. It is at once an agent and vehicle, articulator and witness of being-in-place … Without … our bodies, not only would we be lost in place – acutely disoriented and confused – we would have no coherent sense of place itself. Nor could there be any such thing as lived places, i.e. places in which we live and move and have our being. Our living-moving bodies serve to structure and configurate entire scenarios of place.

Second, the question of “self ” is linked to the body in the literature on place, and closely related to the issue of identity. Such identity entails a body in place, an implaced self, a “body/self ” (Casey 2001:â•›689). Third, because it is through the body that humans make a living in an environment, the experience of doing this is central to what it means to inhabit a place, to dwell in it. Heidegger (1962:â•›100) used the term “work-world” to emphasize that place and self are intimately locked in the world of concrete work. Fourth, both Ingold (2000:â•›193) and Casey (2001:â•›689, 2002) identify the concept of landscape as a complementary term to that of body. Both concepts place an emphasis on form. Fifth, in a dwelling approach, in everyday life humans are thought of as getting along unreflectively using words, implements, clothing and other pieces of equipment, engaging in casual social encounters without consciously posing the problem of our relationship to such entities. Most familiar things around us are thus already integrated into a set of practices for “coping” or getting by, what Pierre Bourdieu (1977) called habitus. Sixth, a dwelling approach insists that humans first of all find themselves enmeshed in a world and in a set of relationships, and that it is only subsequent to this that they begin to separate out a sense of themselves and a sense of things as they are apart from them: according to Heidegger human existence cannot be construed as coming before (either temporally or ontologically) the encounter with other things or persons (Malpas 2006:â•›51–52). Seventh, what creates the world, bringing forth the earth as a collection of human places, is speech. As David White (1978:â•›25) observes:



Embedded in place

… names bestow what Heidegger calls a measured command over entities. The measure is the extent to which any given name is a locus wherein may be experienced relations between the being of the entity named and being as totality.

Names are considered to be part of the things (entities) labeled and language is laid out as a property of the world and not as a coding of the world, as in a Cartesian approach (Weiner 2001:â•›89–90).

3.

How Māori know places

3.1

Introduction

Most of the scholarship using the concept of landscape has been carried out in Western settings, or, if located in the non-Western world, has employed approaches reflecting the ontological and epistemological legacies of European colonialism, producing a “hall of mirrors” in which theory reflects its own views rather than engaging with alternative ontologies (Howitt & Suchet-Pearson 2003:â•›557). To begin to address this issue, this paper uses a modified form of kaupapa Māori, an approach to research which is connected to Māori philosophy, principles, language, and customs (Irwin 1994:â•›25–43; L. Smith 1999, 2000:â•›225–247; G. Smith 2000:â•›209–224; Pihama 2005:â•›191–209). According to Linda Smith (1999:â•›188) kaupapa Māori does not mean the same as Māori knowledge and epistemology, as the concept of kaupapa implies a way of framing and structuring how Māori think about those ideas and practices. Wiri (2001:â•›57–58) notes that kaupapa Māori is bicultural in that it may weave into both Māori and European worlds, and that its theoretical underpinnings may be partially interpreted within a European epistemological framework. A kaupapa Māori approach is based on Māori ways of knowing, or mātauÂ� ranga Māori, which according to Tau (1999:â•›10–24) is the epistemology of Māori, to Salmond (1985:â•›240–262) a complex and open system of knowing the world, and to Royal (1998:â•›80) it is Māori knowledge created to understand the world. It is generally agreed that Māori knowledge is divided into two domains, that of the “upper jawbone” (Te Kauwae Runga) and that of the “lower jawbone” (Te€Kauwae Raro). The former refers to matters that are associated with the origins of the universe and cosmology, the latter with the world of humans. Mason Durie (1994:â•›27) notes that these realms embrace long-standing philosophies in which all power and authority came from higher forces, gods or atua, and mana wairua or mana atua, denoted activities associated with those various deities.

81

82

Brian Murton

While humans might be the agents of the gods or the medium through which their power is expressed, supreme control rests not with individuals, but with deities. In contrast, mana tangata is a reflection of human expertise, assisted by the gods but essentially an acknowledgement of the skills and knowledge which allows certain people to direct day-to-day activities within their communities. Mana whenua is another form of control and authority which enables a group to claim a particular area of land. All three domains of mana are components of a holistic Māori worldview. Thus, Māori see knowledge as involving an inseparable relationship between the world of matter and the world of spirit. In Māori understandings of the world, the Cartesian dichotomy between an observing thinking self and the outside world cannot exist. There are at least three dimensions of Māori knowledge critical to a phenomenological approach to “landscape” as the “face of place”: understandings of time, of genealogy as a way of knowing things, and the importance of the creative word.

3.2

Understandings of time

Michael Shirres (1997:â•›119) tells us that the deepest level of Māori understanding of being and time has it that there is a “singularity” located at the core of the cosmos, where there is neither space nor time. The latter exist only at the level of the material world, following the separation of the Earth Mother from the Sky Father, where they are firmly intertwined. Even in the material world the past, present, and the future are indissolubly interconnected, and for Māori, the ancient symbol of the double spiral (koru) represents the interrelationships of the past, present, and future, of time and space, of spirit and matter (Stewart-Harawira 2005:â•›34). Each circumbulation of the spiral incorporates the past into both the present and the future and, in doing so, reconstitutes both. Although different tribal traditions assign slightly different meanings to the spiral (the energy of precreation, the potential of being, the coming into being, a progression from potential being), in all it symbolizes the integration of the world of the spirit with the worlds of potentiality and creation (Shirres 1997). In Māori semantics, physical phenomena, including people, are held to proceed from this common primal source, and all things unfold their nature in a common dynamic process (Salmond 1985:â•›246–247). This cosmic generative power, this common dynamic process in which all things unfold, already contains the form of every possible being. Consequently, all things in the phenomenal world unfold their nature (tipu), live (ora), and have form (āhua) and so come to possess a body



Embedded in place

(tinana), an immaterial self (wairua), an essence of divine power, a unique living force (mauri), and a characteristic vitality (hau) (Roberts & Wills 1998:â•›49). A dwelling approach to place resonates with Māori understanding of the world. Both Ingold (2000:â•›193) and Casey (2001:â•›689) identify body and “landscape” as complementary terms, each implying the other, each represented by concrete forms, in an analogic/narrative sense. For Ingold landscapes must be understood as “taskscapes” in their embodied form: a pattern of activities collapsed into an array of features. Casey notes that body and landscape are the concretization and exfoliation, respectively, of the initially indefinite dyad of self and place, which need to be fleshed out downward into the body and outward into landscape. To Māori all things have an inherent nature (tipu) and a template (ahua) and these materialize as concrete bodies (tinana) (of people, trees, animals, birds, fish, rocks, and so on). It is the human body that carries the Māori self into place as an implaced self, what Casey (2001:â•›689) calls a “body/self ”. Landscape is complementary to the body. Casey (2001:â•›689) calls it a “cusp concept”. This is why it is necessary to know which particular body from which particular family was involved in making places.

3.3

Genealogy and knowing

For Māori, “to know” something is to locate it in space and time through genealogy (whakapapa). To “know” oneself is to know one’s genealogy. To “know” about a tree, a rock, the wind, or the fish in the sea is to know their genealogy. In its literal translation whakapapa means ‘to place layers, one upon another’. In a genealogical sense, it provides a framework for an understanding of historical descent, pattern, and linkages, whereby everything, animate and inanimate, is connected together into a single family tree or “taxonomy of the universe” (Roberts 1998; Roberts & Wills 1998; Roberts et al. 2004). In some traditions genealogy begins with the primal parents, Papatūānuku (Earth Mother) and Ranginui (Sky Father). These primal parents then produced many offspring, deified as gods (atua), who in turn act as the progenitors and personification of all known phenomena, both living and non-living. Genealogy thus acts as a cognitive template, a paradigm of reality, for the ordering and understanding of the visible and invisible worlds and is the means by which Māori “know” the phenomenal world. Moreover, by transcending this world and connecting all things on earth to the gods, to the universe, and ultimately to the Creator, it provides the framework for an all-encompassing, universal knowledge system. But because of Māori non-linear conceptions of time as represented in the double spiral, and the understanding that all things exist

83

84

Brian Murton

simultaneously, the conception of genealogy is spatial, or perhaps holographic, rather than involving descending vertical lines (Salmond 1985:â•›247). Knowledge associated with the individuals, be they deity, demigod, human, plant, animal, or rock, provides “flesh” for the epistemological “bones” provided by genealogy. This knowledge is primarily in the form of narratives of many kinds: songs, chants, proverbs, and prophetic sayings. All of these, as in oral cultures elsewhere (see Cajete 2000:â•›27–31; Basso 1996a:â•›68–74), make extensive use of personification and metaphor (Roberts & Wills 1998:â•›53).

3.4

Thought and the spoken word

In Māori understanding, creation is viewed as a process of continuous action or coming into being, in which particular forms of sound and thought play an essential role (Salmond 1985:â•›240–263; Stewart-Harawira 2005:â•›37–38). StewartHarawira (2005:â•›38) puts it this way: The cadences of ancient songs, of ritual calls, of sacred chants, through which the world is sung into existence, the flesh is sung onto the bones, and the relationships are sung which bind all together within the cosmos … [breathe] life into the network of subtle interconnections between human beings and the entire natural world.

Thought and spoken words exist together for Māori, with sound being the original foundation for thought to be conceptualized and expressed in word. According to Salmond (1985:â•›246) the primal energy surge produces thought, memory, the mind, then desire. From desire, ancestral knowledge generates darkness and then Te Kore, the primal power of the cosmos, the void or negation, yet containing the potentiality of all things afterwards to come. From Te Kore, space emerges, and then light, land, the gods and men. One northern Māori tradition, notes the tohunga, scholar, writer, healer, Anglican minister, and philosopher, Māori Marsden (Royal 2003:â•›16–23), has it that it is through recitation by a supreme entity, Io, that all things came into being. Marsden in his recounting of the Io tradition, which is appropriate for me to follow, states that it was through the mana of the creative word that everything has come into existence, including the phenomenal world, after the children of Rangi (the Sky Father) and Papatūānuku (the Earth Mother) separate their parents. Just as with the cosmogonic and with non-human things, the spoken word connects the breath of people to the breath of the world and animates, brings life to place. Māori understandings of how things or entities come to be known accord well with a dwelling approach to place. To quote Vincent Vycinas (1969:â•›270), in his



Embedded in place

work interpreting Martin Heidegger, “To say something is to bring it to life”. It is speech which enables Māori to create their world as a collection of human places. But sound, as well as being the vehicle for thought and language, has deep metaphysical and creative connotations that go beyond its use as the practical instrument of ordinary communication. It is names that expose things in their being. In Māori worldview, naming places, in my understanding, belongs to the realm of humans, who descend, depending on tribal tradition, either from Tāne (who is also the progenitor of forest trees, birds, insects, rocks and stones) or Tūmatauenga (who is primarily thought of as the deity of war). But naming places must be seen as part of a broader process of naming and ordering the world. This idea is supported by linguistic evidence. According to the linguist and Māori language scholar Richard Benton (personal communication, 13 April 2007), there is no Māori term equivalent to “place name”. He notes that he has seen the phrase “ingoa wahi,” but that it is a modern artificial construction, a recently invented technical term to convey an idea that did not need to be expressed explicitly in matauranga Māori. A similar phrase “ingoa takiwā, wāhi rānei” has also recently been used to mean ‘place names’ or more specifically region and particular place names. Benton goes on to state that the term “ingoa” (“name, namesake, acquire distinction”) covers the names for people, places, and taonga (property, anything highly prized), and as identity and status are bound up in a name, a place is its name, and the name is the place. He also notes that when a name changes, a place may well become a different one in Māori conceptualization. There are also a number of words and phrases which relate to naming in order to possess, or to affect a place in some way (Benton, pers. comm., 13 April 2007). Important among these is taunaha, to bespeak or reserve a place, object, or person for some future use or purpose. This could include a chief naming a portion of land after a part of his body in order to reserve it for his use or disposition (taunaha whenua). Further, the word taunahanahatanga was frequently used to describe the naming process by ancestors, such as in “Ngā taunahanahatanga o ngā roto o te waka ō aoraki” (“Naming great lakes of the canoe of Aoraki”) (Davis et al. 1990b:â•›86–91). Today the Māori name for the New Zealand Geographic Board, the body responsible for place names in New Zealand, is Ngā Pou Taunaha o Aotearoa (Davis et al. 1990b:â•›viii). Another important phrase is tapatapa whenua, literally ‘bespoken land.’ According to Benton (pers. comm., 13 April 2007) the term tapatapa is derived from tapa “to give a name to, recite, give a command” and denotes formal recitations (for example, an incantation or karakia) and the procedure of formally bestowing someone’s name on an object, either to enhance its value, give the person concerned some claim to it, or to insult the person whose name and therefore mana is so attached. Tapatapa whenua is the application of this custom to formally claim a piece of land, usually by someone

85

86 Brian Murton

of note discovering or settling on it. Another word often used in the context of bestowing names to both places and people was hua (“name, call by name, think, think of, know”).

4.

Māori places

4.1

Whenua: Place

The word whenua means both “earth” and “placenta,” and as such metaphorically represents the connection of people to their origins, material, historical, and spiritual. Māori bury both the placenta and the umbilical cord (pito), whereby symbollically recognizing the connection between place and people. It is also explanatory of the encoding of philosophical and metaphysical knowledge, which, along with genealogy and the rituals of sacred encounter, is exemplified in the geographical landscape (Stewart-Harawira 2005:â•›41). As Rose Pere of Waikaremoana expresses, I whakatangatahia te whenua (The land is imbued with human characteristics) (cited in Wiri 2001:â•›34), a sentiment echoed by Indigenous and many other scholars, who acknowledge that the earth is an entity very much like a person; as such the earth can speak, and one of the ways it does this is through the human act of place naming (Fair 1997:â•›466–480). This idea is similar to the phenomenological view that the body has a relationship to the world through two dimensions, “outgoing” where the lived body participates and is active in the world, and “incoming” where places come into the body, “placializing”, through “tenacity” or the lasting impression that places make on people and “subjection,” the idea that people are subject to places, that places constitute people as subjects. It is this notion that pepeha (a type of proverb or identity axiom) conveys for Māori. Rose Pere’s phrase quoted above is a metaphorical statement, associating the word whenua with both land and the placental membrane. It emphasizes the importance of the place where the placenta is buried, and as such reflects the individual and kin group’s place in the material world. The relationship between the body and land is clearly analogic/narrative. The burial of the placenta, which is regarded as part of the mother, the child, and the earth, also inserts the individual into the landscape of origin. Whenua, then, is the critical element of being and existing in the physical world and the critical element in understanding the holistic nature of Māori consciousness. It is a mirror where the land and the people reflect each other, and are each other. It is this holistic conception of whenua that has led scholars to identify it with the Western concept of place (Roberts & Wills 1998:â•›54–55).



4.2

Embedded in place

“Mirror knowledge”

Te Maire Tau (2001:â•›111–152) calls Māori knowledge “mirror knowledge”: for Māori the world is a mirror of their lives. All things reference back through the tribe to the family and ultimately to “self ”. The circle is confirmed and expressed in the proverb that asks the question, He tangata, he tipua, he atua rānei? (Am I human, demigod, or God?). The question, of course, is rhetorical, as the answer is all three. As the philosophers Perrett (2003:â•›258–259) and Patterson (1992:â•›109– 111) note, the traditional Māori view of self is markedly non-individualistic. A person’s identity is determined predominantly by his or her inherited status and relationship to a larger social group, membership of which is genealogically determined. This idea is carried to the point where what Europeans would think of as an individual is identified with the kinship group; hence the use in Māori of the personal pronoun au (“I”) to refer to either the individual or the tribe. As Tau (2001:â•›139) aptly observes, kinship identity reaches both backwards and forwards in time, and involves not only humans but deities. What gives unity to a life in this radically different non-individualistic conception of selfhood is tribal knowledge. This provides historical and social identity, which define the context within which an individual’s whole life takes place. Tribal knowledge is primarily narrativebased, and these narratives depict humans and other beings and their places in the cosmos, both in more universal forms (as, for example, in major cosmological and discovery narratives) and in fine detail (as, for example, in genealogy, family stories, place-naming stories, and so on). Basically, the whole Māori cosmos is conceived of as a gigantic “kin”, and humans are powerfully linked thereby to both the natural and supernatural world. This projection of “self” implies that when Māori claim descent from the ancestors and gods they did so horizontally. Among Māori, as we have noted, the ancient symbol of the double spiral (koru) represents the interrelationships of the past, present, and future, as well as of time and space, and spirit and matter (Stewart-Harawira 2005:â•›34). Within Māori ontological and cosmological paradigms it is impossible to conceive of the present and future as separate and distinct from the past, and that the past is constitutive of the present and is inherently reconstituted within the future. Ritual and naming practices impress ancestors and deities into the landscape, and they are thus immediate and available. After all, Papatūānuku, the Earth Mother, lies directly in front of the people. Ancestors, in claiming territory, named land marks, rocks, trees, waterfalls, and so on, and together tribal understanding with the entities themselves so that a place and its knowledge cannot be separated. Named places not only record and express important mythological and historical aspects of tribes, but by the

87

88

Brian Murton

virtue of the knowledge contained in the names, they – and the places they represent – are the physical manifestation of the tribe, just as the tribe is the manifestation of the land. The past can be read because it was before people. With the understanding that actions are incorporated into the earth, and into the minds of people, for Māori the phenomenal world, the world of things and objects, was a “mirror landscape”, a reflection of themselves, a metaphoric extension of the body, just as they are a reflection of their landscape.

5.

Named places as the face of place

5.1

Introduction: The study of Māori named places

Māori place names have been the subject of considerable scrutiny in the past. Most studies in the late nineteenth and through most of the twentieth century were the work of scholars of European ancestry who focused on collecting, classifying and seeking the origins of the name, although a few do recount the narratives in which the names appear. Notable examples of this work include the studies of Andersen (1942), Adkin (1948), Beattie (1945, 1948), Insull (1952), Reed (1974), David Simmons’ (1987) editing of George Graham’s manuscript “Māori Place Names of Auckland” (1927), and Stafford (1994, 1996) on Te Arawa. A number of tribal histories compiled by Europeans also contain many place names, including those by Best (1925 reprinted 1977) on Tūhoe, J. Mitchell (1944) on Ngāti Kahungunu, L. Kelly (1949) on Tainui, and Stafford (1982) on Te Arawa. Only recently, published studies, mostly now by Māori, have begun to integrate place names more comprehensively into tribal experience; these include Halbert (1999), Jones and Biggs (1995), and Tūroa (2000). But few of these (Wiri 2001 is an exception) have attempted to specifically locate Māori place names within the Māori knowledge system: their role is taken for granted, by knowledgeable Māori at least. Perhaps the best example of a publication which integrates names, both of places and ancestors, in the context of narratives is He Kōrero Pūrākau mo Ngā Taunahanahatanga a Ngā Tūpuna. Place Names of the Ancestors. A Māori Oral History Atlas (Davis et al. 1990a). Another recent approach has investigated how names (of people and places) were used in “oral mapping” (or “performance cartography”, Woodward & Lewis 1998:â•›1–10, 537–541), the use of spoken text to describe routes, sometimes accompanied by the drawing of lines on the ground with a stick (J. Kelly 1999:â•›1–30; Barton 1998:â•›493–536).



5.2

Embedded in place

Named places: Meaning and context

It should be noted that polysemy is a characteristic feature of the Māori language, and that many place names, when broken down into their separate parts, can be “translated” in several different ways. Davis et al. (1990b:â•›16) point out that it is only with knowledge of the appropriate narratives that the meaning of the name can be established and comment that “only the unwise attempt to translate a name lacking the evidence of tradition”. Further, Māori liked to name places after an event that had occurred to a person there (ibid.:â•›22). Very often in these cases “o,” meaning “the place of …” or “the place where,” is found as a prefix in a place name to which is added either the name of the person or what they did there. Some place names were, therefore, quite long. Today many names in common use are only a small part of a longer name, especially ones commemorating events and activities of ancestors.

5.3

Terms for landscape features

A very large number of named places include generic terms for parts of the environment. Gabriele Cablitz (2006, 2008), in her work on the Marquesas, makes an important distinction between what she terms “geographical entity/landscape feature” or “landscape term”, and the term “place name”. The former refers to generic terms for parts of the environment, and the latter to the class of proper nouns naming locations, which can include generic terms in them. Māori terms for landscape features include maunga (mountain), pae maunga (mountain range), puke (hill), hiwi (ridge of a hill), papa (flat ground, or in the context of a place name, ‘a place of ’), pari (cliff, precipice, flow of the tide), kōhatu (rock), toka (rock, often in a river or the sea), riu (valley), mānia (plain), ana (cave), wai (water generally), awa (river), manga (river, stream, stretch of water, watercourse), rere (flow, often a waterfall), puna (spring of water), rae, kūrae, matarae (promontory, headland, peninsula), whanga (bight, bay, large rive mouth), ngutuawa (river mouth), motu (small island), moutere (small island), kokoru (bay), one (beach), repo (swamp), and there are more (Davis et al. 1990b:â•›28–39; Tūroa 2000). Most of the terms are simplex forms, and some derive from body parts, e.g. mata, which can mean “face” but also “headland,” as well as having several other meanings. The terms for landscape features frequently occur as parts of place names, as they do elsewhere in the Pacific (Cablitz 2006, 2008). The importance to Māori of landscape features, notably mountains and hills, as well as rivers and lakes, is demonstrated in pepeha, proverbs which take a set form, and which define the relationships between such features and the tribe of

89

90 Brian Murton

the area. Let me illustrate with an example drawn from the Hokiānga area (the ancestral area of my father’s people).

Ko Whakarongorua te maunga Ko Utakura te awa Ko Te Honihoni te hapu Ko Te Ngā Puhi te iwi



The mountain is Whakarongorua The river is Utakura The sub-tribe is Te Honihoni The people are Ngā Puhi

The first line identifies a feature, a mountain or hill, the second line a river, and this unvarying form emphasizes the significance of these features. The third and fourth lines identify the people. More can be added to such sayings, such as a canoe, an ancestor, a marae (a complex of buildings, including a meeting house, belonging to a related group of people). They are identity axioms linking land and people into a whole in such a way as to make them inseparable.

5.4

Imagining named places

Like many other peoples (see Weiner 2001:â•›16–30; Thornton 2008:â•›27, 70), Māori imagined sequences of place names as the “footsteps of the ancestors” (Ngā tapuwae o ngā matua tupuna), paths or trails followed for a variety of purposes. Many Māori names can only be understood through their connection to other names and other places, and whole series of names belong together in groups, commemorating journeys of exploration by an ancestor, the understanding of how the land was made by gods and demigods, or a series of relationships between events and/or people (Davis et al. 1990a:â•›xiii). Some groups of names, as well as individual names, were of such significance that when a tribe migrated elsewhere it “transplanted” its history in its new home by using names from the place of origin. Because of the role of place names as a device for recording and remembering tribal history, the historical events themselves sometimes became relocated in the new setting. It is little wonder that the biggest single group of Māori place names consists of the names brought from homelands (either in the tropical Pacific, usually referred to as Hawaiki, or elsewhere in Aotearoa). Another way of imagining the naming process is in terms of the ways in which Māori recognized how a group, such as a tribe or sub-tribe, could base claims to stretches of country in pre-European times; these ways became enshrined in the



Embedded in place

Native Land Court (now the Māori Land Court) records in the nineteenth century. The Native Land Court, which was established in 1865, used and over time modified traditional ways (take) of claiming stretches of country. Take included the right of discovery (whenua kite hou), the right to occupy based on bloodlines (take tipuna), right of conquest (take raupatu), ability to hold off invasion (ringa kā), ability to keep the home fires burning (ahi kā), and gifting (tuku whenua) (Law Commission 2001:â•›47–49, 61–66; E. T. Durie 1994:â•›61–81; Wiri 2001:â•›101– 102, 164). These rights are complementary and all involve the naming of places associated with the activities of ancestors.

5.5

The names are the places

Places bearing names were the outcomes of activities of the ancestors (He tangata, he tipua, he atua rānei). These involve several levels of generalization. All tribal groups have slightly different narratives about how the islands that comprise Aotearoa came into being, and about their discovery. But most named places are only known and understood at a local level, within the territories of tribes and sub-tribes, even if they involve “transported” names. Tribes have a sense of the boundaries of their territory (rohe), usually high points (maunga), and usually the area included within such points overlaps with the rohe of adjacent tribes. All tribal groups also have stories about the formation of striking natural features (for example, mountains, lakes, harbors) and their associated names within their territories. But the vast majority of names are those of settlement sites, both fortified (pā) and undefended (kāinga), various kinds of sacred sites (wāhi tapu), food and other resource sites of various kinds, including gardens, birding sites, fishing sites, fern grounds and collectively referred to as mahinga kai, sites associated with initial exploration by an ancestor, and places where ancestors fought off intruders. These latter named features were the outcome of immersion in place through the Māori self/body’s involvement with the environment as part of their normal business of life (Ingold 2000:â•›194–195 uses the term “taskscape” and Casey 2001:â•›689 “habitation” to describe such activity). The examples of names which follow are a mere glimpse of places that received names. Most names at the local level are not on modern maps, and many are unknown to all but a few today and in a sense no longer exist as places. Attention is restricted to the lands surrounding the Hokiānga Harbor, an area which has names associated with an early discoverer, Kupe, and with mythical creatures, taniwha, which created landforms and with which my family has long associations.

91

92

Brian Murton

Naming by a distant ancestor There are many variations in the traditions of Kupe and many place names drawn from the stories of his explorations and adventures (Davis et al. 1990a:â•›8–17). There may well have been several people called Kupe, but most traditions say he was the first of the early Polynesian ancestors to arrive from Hawaiki. One of the areas where a Kupe tradition is very strong is around the Hokiānga harbor on the west coast of the northern part of Te Hiku o Te Ika a Maui (the tail of the fish of Maui, the pan-Polynesian demigod). The full name for the harbor is Te Hoki Ānga a Kupe, named because this was the place from where Kupe set out on his return voyage to his homeland. There are also a number of names around the harbor ascribed to Kupe, including Kohukohu (the place where Kupe cursed those responsible for food being cold when ovens were opened), Ngā Kuri a Kupe (“the dog of Kupe”), and a number of other rocks at the harbor’s entrance. He also left one of his pet taniwha (a powerful spirit placed in guardianship over a certain area, often living in a water body), Ārai Te Uru (a name well known in other traditions as well), as a guardian on the southern side of the harbor’s entrance. The formation of the Hokiānga Harbor Throughout Aotearoa, traditions attribute landscape change and especially the creation of landforms through sudden natural events to taniwha. According to one tradition (Hokianga Harbour Wairere Boulders, n.d.) the Hokiānga Harbor was formed by the children of Ārai Te Uru, who, in searching for homes of their own, carved out the various branches and rivers entering the water body, which are named after them (Waihou, Waimā, Utakura, Wairere, Ōmanaia, Whirinaki, Ōrirā, Mangamuka, Wairupe, Motukaraka, Ōhopa, Waireira, and Motukauri). Each taniwha’s adventure has a story. I only recount two here. Waihou was the first to leave his home, burrowing inland and flooding the land, battering down a solid hill to create what is today called “The Narrows,” skirting around a high wall of mountains, and eventually waddling up a slope, where he went to sleep. The circular hole he made with his tail filled with rainwater and eventually drowned him, creating the lake known as Ōmāpere. Another of Waihou’s brothers, Utakura, almost reached his brother, creating the Utakura river valley, the outlet to Lake Ōmāpere, in the process. However, he was foiled by the high drop of the Korotangi Falls, and makes his home today in the pool below the falls. Another narrative as to the origins of Lake Ōmāpere has it that the taniwha in the lake is called Takauere (Bay of Islands Minute Book 1929:â•›6). According to this account the lake was once a swamp rich in eels. Ngātikoro and his son Tara went there to eel. It grew dark and the son, thinking his father had left, set fire to the bush, and his father was accidentally burned to death. Takauere, another son of Ngātikoro, had died before the fire, and his body was buried in the swamp on a



Embedded in place

rock called Paparoa, on which a tree grew. But later when people went looking for the body they found nothing, and because of this Takauere became clothed with the mana of a taniwha and the place became tapu, “taboo”. There are a number of other versions of the origins of Takauere but all agree that Takauere could be found at Ngāwha geothermal springs as well as at nearby Lake Ōmāpere, though some state that he is present throughout the north.

Named places at Lake Ōmāpere In 1879 the land court heard testimony to consider the ownership of the Ōmāpere block of 768 acres, through which ran the outlet streams to the lake, as well as ditches that had been dug to facilitate eel catching (Northern Minute Book 1879:â•›455–466). All the claimants were closely related, most tracing their ancestry to Tūiti and his wives, Marohāwea and Moengāroa, through various children who were involved in catching eels in the streams, and giving names to places: knowledge about the names, the people, and their activities “mirror” each other. The places named were the outcome of events happening in the mid-seventeenth century, and involve Tūiti’s children Rangahana and Korohue by Marohāwea, and Tike, the son of Ruitaia, Tūiti’s son by his second wife, Moengāroa. Other people involved claimed through Tūtahua, a daughter of Tūiti, or through ancestors who had been granted use rights. Our family are descendant through lines from both of Tūiti’s wives. In the records fifteen places are named: six eel weirs, three ditches dug to channel eels, three natural outlet streams to the lake, two burial places, and one cave or rock shelter. One claimant stated that his ancestors, Rangahana and Korohue, worked at “eel catching at Omapere” (Northern Minute Book 1879:â•›453). Rangahana and his wife resided there and Rangahana dug the ditch called Ngā Matawhero o Rangahana, because when digging the ditch his face (mata) became red (whero) from exertion. Another eel weir was named after the flax that grew there (Te Harakeke), and another, Ahipārera, was named because ducks (pārera) were cooked (literally burned) there. The weir (pā), Rāhiripāroa, was a long one (roa), and was named after Rāhiri, most probably named after the ancestor from whom all the northern tribe, Ngapuhi, trace ancestry. It was close to a place where Tike was exposed on a rock when he died, before he was interred in the burial ground Te Rere o Tike, so named because this was where Tike fell into the water. The cave initially was named Te Ana a Rākeia after a person, but was renamed Te Ana Kāuta, because it was the kitchen (kāuta) where eels were cooked. The natural stream names were Waipapa (which has many possible meanings, including a stream across a rock flat or plain, but also a place of water, or a second growth of timber), Te Kūaha (a gateway or entrance), and Waihōanga (sandstone water).

93

94 Brian Murton

This brief example illustrates how Māori made a living in their environment and, in doing this, dwelt in it. The places could be named by those who were there, and the act of naming places not only brought them to life, but also the ancestors who had made and named them. The Māori body/self was locked in the world of concrete work, especially through the everyday experience of obtaining food, movement and other activity. Much of this involved habitual activity, the result of socialization from a very young age in a world where the presence of others (people and general forms of life) were taken for granted, where the boundary between one’s perception of oneself and all others is unarticulated.

6.

“Simultaneous landscapes”

An appropriate way to think of the matrix of named places is as a “simultaneous landscape,” whereby place names, the markers (tohu) of the landscape, open directly onto cosmology as well as to the ancestors, history, and the everyday (Simmons 1996:â•›23). From this perspective, the body/self is an integral part of a close-knit harmony of organic parts united to the cosmos and world. Place names are imagined as the “footsteps of the ancestors” and are sign posts, survey pegs of memory, to the events and processes by which the world was created and the sequences associated with their continuous activity, and as such, place names need to be remembered and spoken in order for places to continue to be made. If the name is remembered it can release whole parcels of history to a tribal narrator and those listening. The daily use of place names means that history is always present, always available. The paper has been concerned with how, from a dwelling perspective, Māori represent places, with how they “make” places. Although some evidence indicates that Māori in the past did “gaze” at their surroundings and did have visual forms of representation, the concept of “landscape” as it is commonly used in Western scholarly and popular representations is inappropriate. What replaces the profound visuality of Western ways of understanding the world for Māori is language, especially sound and speech. Māori represent the world, indeed create it, in speech and through the act of naming. Naming places, I have argued, is but part of an overall naming process, which for Māori involves genealogy as the way of knowing the phenomenal world. For Māori, just as for Martin Heidegger (Malpas 2006:â•›51–52), existence cannot be construed as coming before (whether temporally or ontologically) the encounter with other things or other persons. Existence is “there” and is not something separate from the place, the world in which it finds itself. Names are part of the things they label, and place names illustrate the unity of language and



Embedded in place

place. Naming places is an integral part of active perceptual engagement with the animate and inanimate world for Māori. For Māori the world is full of active entities with which people engage. This kind of engagement has been called the “participation mystique” by the French anthropologist Lucien Lévy-Bruhl (Abram 1996:â•›57). Similarly, Cajete (2000:â•›27, 184) stresses that many American Indian languages are verb-based, reflecting the perpetual creative motion of the world; and Thornton (2008:â•›80) echoes this when he notes that Athabaskan languages are characterized by a grammatical emphasis on the verb and on complex prefixing and classificatory structures, which allow whole phrases to be built out of a single verb stem. However, according to Benton (personal communication, 29 December 2008), while a reasonable case can be made for classifying many American Indian languages as “verb-based”, it would be stretching the bounds of credulity to place Māori and other Polynesian languages in that category. Nonetheless, Māori viewed the world as being in perpetual creative motion, as being alive. After all, everything in it is a relative, connected through genealogy, and existing simultaneously. In using a kaupapa Māori approach, a way of framing and structuring, I have based my exploration of named places on relevant dimensions of mātauranga Māori, Māori ways of knowing. This has enabled me to identify pertinent parallels with ideas emanating from phenomenology, and especially from a dwelling approach. Even in contemporary times, in Māori worldview all physical landscapes are inseparable from ancestors, events, occupations and cultural practices (Hoskins 2008:â•›28). As the “Te Aranga Maori Cultural Landscape Strategy” (Te€Aranga Steering Committee 2008:â•›1) points out: As Maori we have a unique sense of our “landscape.” It includes past, present and future. It includes both physical, and spiritual dimensions. It is how we express ourselves in our environment. It connects whānau, flora, fauna, through whakapapa. It does not disconnect urban from rural. It transcends the boundaries of “land”scape into other “scapes”; rivers, lakes, ocean and sky. It is enshrined in our whakapapa, pepeha, tauparapara, whaikōrero, karakia, waiata, tikanga, ngā kōrero a kui mā, a koroua mā, and our mahi toi.

References Abram, David. 1996. The Spell of the Sensuous. Perception and Language in a More-than-human World. New York NY: Vintage Books. Adkin, G. Leslie. 1948. Horowhenua. Its Maori Place-names and their Topographical and Historical Background. Wellington: Department of Internal Affairs.

95

96 Brian Murton

Andersen, Johannes. 1942. Maori Place-Names. Wellington: Polynesian Society of New Zealand. Appleton, Jay. 1975. The Experience of Landscape. London: John Wiley and Sons. Barton, Phillip L. 1998. Maori Cartography and the European Encounter. In The History of Cartography, Vol. Two, Book Three: Cartography in the Traditional American, Arctic, Australian, and Pacific Societies, David Woodward & G. Malcolm Lewis (eds), 493–536. Chicago IL: University of Chicago Press. Basso, Keith H. 1988. “Speaking with names”: language and landscape among the Western Apache. Cultural Anthropology 3(2): 99–130. Basso, Keith H. 1996a. Wisdom Sits in Places: Language and Landscape among the Western Apache. Albuquerque NM: University of New Mexico Press. Basso, Keith H. 1996b. Wisdom sits in places: Notes on Western Apache landscape. In Senses of Place, Steven Feld & Keith H. Basso (eds), 53–90. Santa Fe NM: School of American Research Press. Bay of Islands Minute Book. 1929. Native (Maori) Land Court, No. 11, Omapere. Beattie, J. Herries. 1945. Maori Place-names of Canterbury. Dunedin: Otago Daily Times and Witness. Beattie, J. Herries. 1948. Otago Place-names. Dunedin: Otago Daily Times and Witness. Belgrave, Michael, Kawharu, Merata, Williams, David & Kawaharu, I. H. (eds). 2005. Waitangi: Perspectives on the Treaty of Waitangi. Melbourne: OUP. Best, Elsdon. 1925 reprinted 1977. Tuhoe, the Children of the Mist: a Sketch of the Origin, History, Myths and Beliefs of the Tuhoe Tribe of the Maori of New Zealand. With some Account of Other Early Tribes of the Bay of Plenty District. Auckland: A. H. & A. W. Reed. Best, Elsdon. 1942 [1977]. Maori Forest Lore. Wellington: Government Printer. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: CUP. Burenhult, Niclas & Levinson, Stephen C. 2008. Language and landscape: A cross-linguistic perspective. Language Sciences 30(2/3): 135–150. Byrnes, Giselle. 2004. The Waitangi Tribunal and New Zealand History. Melbourne: OUP. Cablitz, Gabriele H. 2006. Marquesan. A Grammar of Space. Berlin: Mouton de Gruyter. Cablitz, Gabriele H. 2008. When “what” is “where”: A linguistic analysis of landscape terms, place names and body parts in Marquesan (Oceanic, French Polynesia). Language Sciences 30(2/3): 200–226. Cajete, Gregory. 2000. Native Science. Natural Laws of Interdependence. Santa Fe NM: Clear Light Publishers. Casey, Edward S. 1993. Getting Back into Place: Toward a Renewed Understanding of the PlaceWorld. Bloomington IN: Indiana University Press. Casey, Edward S. 1996. How to get from space to place in a fairly short stretch of time: phenomenological prolegomena. In Senses of Place, Steven Feld & Keith H. Basso (eds), 13–52. Santa Fe NM: School of American Research Press. Casey, Edward S. 1997. The Fate of Place: A Philosophical History. Berkeley CA: University of California Press. Casey, Edward S. 2001. Between geography and philosophy: What does it mean to be in the place-world? Annals of the Association of American Geographers 91(4): 683–693. Casey, Edward S. 2002. Representing Place. Landscape Painting and Maps. Minneapolis MN: University of Minnesota Press. Clark, Kenneth. 1956. Landscape into Art. Harmondsworth: Penguin Books.



Embedded in place

Corner, James. 1999. Introduction: recovering landscape as a critical cultural practice. In Recovering Landscape. Essays in Contemporary Landscape Architecture, James Corner (ed.), 1–26. New York: Princeton Architectural Press. Cosgrove, Denis E. 1998. Social Formation and Symbolic Landscape. Madison WI: University of Wisconsin Press. Creswell, Tim. 2003. Landscape and the obliteration of practice. In Handbook of Cultural Geography, Kay Anderson, Mona Domosh, Steve Pile & Nigel Thrift (eds), 269–281. Thousand Oaks CA: Sage. Davis, Te Aue, O’Regan, Tipene, Whiting, Cliff & Wilson, John. 1990a. He Kōrero Pūrākau Mo Ngā Taunahanaha a Ngā Tūpuna. Place Names of the Ancestors. A Maori Oral History Atlas. Wellington: The New Zealand Geographic Board. Davis, Te Aue, O’Regan, Tipene & Wilson, John. 1990b. Ngā Tohu Pūmahara. The Survey Pegs of the Past. Wellington: New Zealand Geographic Board. Deloria, Vine & Wildcat, Daniel. 2001. Power and Place: Indian Education in America. Golden CO: Fulcrum Publishers. Durie, E. T. 1994. Custom Law. Working paper. Wellington: Waitangi Tribunal. Durie, Mason. 1994. Whaiora. Maori Health Development. Auckland: OUP. Fair, Susan W. 1997. Inupiat naming and community history: The Tapqaq and Saninaq coasts near Shishmaref, Alaska. The Professional Geographer 49(4): 466–480. Halbert, Rongowhakaata. 1999. Horouta. The History of the Horouta Canoe, Gisborne and East Coast. Auckland: Reed Books. Heidegger, Martin. 1962. Being and Time. New York NY: Harper and Row. Heidegger, Martin. 1971. Building dwelling thinking. In Poetry, Language, Thought, translated and with an introduction by Albert Hofstadter, 141–160. New York NY: Harper Collins. Heidegger, Martin. 1977. The age of the world picture. In The Question Concerning Technology and Other Essays, translated and with an introduction by William Lovitt, 115–154. New York NY: Harper and Row. Hokianga Harbour Wairere Boulders. n.d. (20 July 2010). Hoskins, Rau. 2008. “Our faces in our places”: Cultural landscapes – Maori and the urban environment. In Re-Thinking Urban Environments and Health, The Public Health Advisory Committee, 27–33. Wellington: Ministry of Health. Howitt, Richard & Suchet-Pearson, Sandra. 2003. Contested cultural landscapes. In Handbook of Cultural Geography, Kay Anderson, Mona Domosh, Steve Pile & Nigel Thrift (eds), 557–569. Thousand Oaks CA: Sage. Ingold, Tim. 2000. The Perception of the Environment: Essays in Livelihood, Dwelling, and Skill. London: Routledge. Insull, Herbert A. H. 1952. Marlborough Place Names. Wellington: A. H. & A. W. Reed. Irwin, Kathie. 1994. Maori research methods and practise. Sites 28: 25–43. Jay, Martin. 1992. Scopic regimes of modernity. In Modernity and Identity, Scott Lash & Jonathan Friedman (eds), 178–195. Oxford: Blackwell. Jones, Pei Te Hurinui & Biggs, Bruce. 1995. Nga Iwi o Tainui: The Traditional History of the Tainui People. Auckland: Auckland University Press. Kelly, Jan. 1999. Maori maps. Cartographica 32(2): 1–30. Kelly, Leslie G. 1949. Tainui: The Story of Hoturoa and his Descendants. Wellington: The Polynesian Society. Law Commission. Te Aka Matua o Te Ture. 2001. Maori Customs and Values in New Zealand Law. Study Paper. Wellington: Law Commission.

97

98 Brian Murton

Little Bear, Leroy. 2000. Jagged world views colliding. In Reclaiming Indigenous Voice and Vision, Marie Battiste (ed.), 77–85. Vancouver BC: University of British Columbia Press. Macnaughten, Phil & Urry, John. 1998. Contested Natures. Thousand Oaks CA: Sage. Malpas, Jeff. 1999. Place and Experience. A Philosophical Topography. Cambridge: CUP. Malpas, Jeff. 2006. Heidegger’s Topology: Being, Place, World. Cambridge MA: The MIT Press. Mark, David M. & Turk, Andrew G. 2003. Ethnophysiography. Paper presented at the Workshop on Spatial and Geographic Ontologies, 23 September, 2003 (prior to COSIT03), in Ittingen, Switzerland. Mark, David M. & Turk, Andrew G. 2004. Ethnophysiography and the ontology of landscape. Proceedings of GIScience, Max Egenhofer, Christian Freksa & Harvey Miller (eds), 152– 155. Santa Barbara CA: Regents of the University of California. McKinnon, Malcolm (ed.). 1997. New Zealand Historical Atlas. Ko Papatuanuku e Takoto Nei. Wellington: Department of Internal Affairs. Merleau-Ponty, Maurice. 1962. The Phenomenology of Perception. London: Routledge and Kegan Paul. Mitchell, J. H. 1944. Takitimu. Wellington: A. H. & A. W. Reed. Mitchell, W. J. T. 2002. Imperial landscape. In Landscape and Power, W. J. T. Mitchell (ed.), 5–34. Chicago IL: University of Chicago Press. Momaday, Scott. 1974. Native American attitudes to the environment. In Seeing with a Native Eye: Essays on Native American Religion, Walter Capps (ed.), 79–85. New York NY: Harper and Row. Myers, Fred. 1986. Pintupi Country, Pintupi Self: Sentiment, Place, and Politics among Western Desert Aborigines. Washington DC: Smithsonian Institution Press. Neich, Roger. 1994. Painted Histories. Early Maori Figurative Painting. Auckland: Auckland University Press. Northern Minute Book. 1879. Native (Maori) Land Court, No. 3. Omapere. Olwig, Kenneth R. 2002. Landscape, Nature, and the Body Politic. From Britain’s Renaissance to America’s New World. Madison WI: The University of Wisconsin Press. Patterson, John. 1992. Exploring Maori Values. Palmerston North: The Dunmore Press. Perrett, Roy W. 2003. Nga whakaaro Maori: Maori philosophy. In From Africa to Zen. An Invitation to World Philosophy, Robert C. Solomon & Kathleen M. Higgins (eds), 255–268. Lanham: Rowman and Littlefield. Phillips, F. L. 1989. Nga Tohu a Tainui= Landmarks of Tainui: A Geographical Record of Tainui Traditional History. Otorohanga: Tohu Publishers. Pihama, Leonie. 2005. Asserting Indigenous theories of change. In Sovereignty Matters. Locations of Contestation and Possibility in Indigenous Struggles for Self-determination, Joanne Barker (ed.), 191–210. Lincoln NE: University of Nebraska Press. Reed, A. W. 1974. A Dictionary of Maori Place Names. Wellington: A. H. & A. W. Reed. Relph, Edward. 1981. Rational Landscapes and Humanistic Geography. London: Croom Helm. Roberts, Roma Mere. 1998. Indigenous knowledge and Western science: Perspectives from the Pacific. In Science and Technology, Education and Ethnicity: An Aotearoa/New Zealand Perspective [Royal Society of New Zealand Miscellaneous Series 50], Derek Hodson (ed.), 43–77. Wellington: Royal Society of New Zealand. Roberts, Roma Mere, Haami, Brad, Benton, Richard, Satterfield, Terre, Finucane, Melissa L., Henare, Mark & Henare, Manuka. 2004. Whakapapa as a Maori mental construct: Some implications for the debate over genetic modification of organisms. The Contemporary Pacific 16(1): 1–28.



Embedded in place

Roberts, Roma Mere & Wills, Peter R. 1998. Understanding Maori epistemology. A scientific perspective. In Tribal Epistemologies. Essays in the Philosophy of Anthropology, Helmut Wautischer (ed.), 43–77. Aldershot: Ashgate. Royal, Te Ahukaramu Charles. 1998. Te Ao Marama – a research paradigm. In Te Oru Rangahau. Proceedings of the Maori Research and Development Conference, 7–9 July 1998, Te€Pumanawa Hauora (ed.). Palmerston North: Massey University. Royal, Te Ahukaramu Charles. 2003. The Woven Universe. Selected Writings of Rev. Maori Marsden. Otaki: The Estate of Rev. Maori Marsden. Salmond, Anne. 1985. Maori epistemologies. In Reason and Morality, Joanna Overing (ed.), 240–263. London: Routledge. Schama, Simon. 1995. Landscape and Memory. London: Harper and Collins. Shirres, Michael P. 1997. Te Tangata: The Human Person. Auckland: Accent Publications. Simmons, David. 1987. Maori Auckland, Including the Maori Place Names of Auckland Collected by George Graham. Auckland: The Bush Press. Simmons, David. 1996. The spiritual landscape. Lecture, University of Auckland. Smith, Graham Hingangaroa. 2000. Protecting and respecting Indigenous knowledge. In Reclaiming Indigenous Voice and Vision, Marie Battiste (ed.), 209–224. Vancouver BC: University of British Columbia Press. Smith, Linda Tuhiwai. 1999. Decolonizing Methodologies. Research and Indigenous Peoples. New York NY: Zed Books. Smith, Linda Tuhiwai. 2000. Kaupapa Maori research. In Reclaiming Indigenous Voices and Vision, Marie Battiste (ed.), 225–247. Vancouver BC: University of British Columbia Press. Stafford, Don M. 1982. Te Arawa: A History of the Arawa People. Wellington: A. W. Reed. Stafford, Don M. 1994. Landmarks of Te Arawa, Vol. 1: Rotorua. Auckland: Reed Publishing. Stafford, Don M. 1996. Landmarks of Te Arawa, Vol. 2: Rotoiti, Rotoehu, Rotoma. Auckland: Reed Publishing. Stewart-Harawira, Makere. 2005. The New Imperial Order. Indigenous Responses to Globalization. New York NY: Zed Books. Tau, Te Maire. 1999. Matauranga Maori as an epistemology. Te Pouhere Korero 1(1): 10–24. Tau, Te Maire. 2001. The death of knowledge: Ghosts on the plains. The New Zealand Journal of History 35(2): 131–152. Te Aranga Steering Committee. 2008. Te Aranga Maori Cultural Landscape Strategy. <www. ngaaho.maori.nz>. Thornton, Thomas F. 2008. Being and Place Among the Tlingit. Seattle WA: University of Washington Press. Tūroa, Taimoana. 2000. Te Takoto o Te Whenua o Hauraki. Hauraki Landmarks. Edited and additional materials by Te Ahukaramu Charles Royal. Auckland: Reed Publishing. Vycinas, Vincent. 1969. Earth and Gods. An Introduction to the Philosophy of Martin Heidegger. The Hague: Martinus Nijhoff. Ward, Alan. 1999. An Unsettled History: Treaty Claims in New Zealand Today. Wellington: Bridget Williams Books. Weiner, James. 1991. The Empty Place. Bloomington IN: Indiana University Press. Weiner, James. 2001. Tree Leaf Talk. A Heideggerian Anthropology. New York NY: Berg. White, David. 1978. Heidegger and the Language of Poetry. Lincoln NB: University of Nebraska Press. Williams, H. W. 1844 [7th edn, reprinted 1992]. Dictionary of the Maori Language. Wellington: GP Publications.

99

100 Brian Murton

Wiri, Robert K. J. 2001. The Prophecies of the Great Canyon of Toi. A History of Te Whāiti-NuiA-Toi in the Western Urewera Mountains of New Zealand. PhD dissertation, University of Auckland. Woodward, David & Lewis, G. Malcolm. 1998. The History of Cartography, Vol. Two. Book Three: Cartography in the Traditional African, American, Arctic, Australian, and Pacific Societies. Chicago IL: The University of Chicago Press.

chapter 5

Philosophical issues in ethnophysiography Landform terms, disciplinarity, and the question  of method Bruce B. Janz

Ethnophysiography is a nascent discipline, one which draws on at least a halfdozen existing disciplines. These disciplines exist in productive tension, a  tension which produces a range of possible answers to some central questions. These questions include: What is being analyzed? What issues arise when gathering data? How are questions framed to access data? What is the goal of ethnophysiography? How is human meaning connected to human expression, in the context of place language? What are the implications of applying the reconstructed data? And, if ethnophysiography is to be seen as a nascent discipline, how does it relate to other disciplines? This chapter expands on these questions that will help to develop and strengthen the concepts at the center of ethnophysiography, give it an identity, and suggest further research possibilities.

0.

Introduction

Ethnophysiography is a nascent disciplinary structure that has its roots in an interdisciplinary research program, as opposed to being a refinement of a method within an existing discipline. It is in the process of developing a set of methods, a collection of objects of analysis, and a set of desired outcomes. It draws on (at least) geography, linguistics, philosophy, anthropology, psychology, and computing, using the methods and assumptions of each to build a new space which allows researchers to inquire about the relationship between natural spatial features and the ways in which those features are parsed, captured in language, and logically structured (Mark & Turk 2003). Characterizing ethnophysiography in this way already takes a stand on the nature of the enterprise. Disciplines, even nascent ones, allow a diversity of methodological and theoretical approaches. If we characterized this as a research

102 Bruce B. Janz

project, we would implicitly be assuming a more or less unified methodology and theoretical approach, not to mention agreed-upon desired outcomes. Framing ethnophysiography as an emerging discipline is risky (not many candidates for disciplinarity actually achieve that status), but it is much more interesting, I would argue, and truer to the most interesting aspects of the material as it exists. While we may want to resist problematic and controversial elements of discipline construction, such as the formation of a canon or the designation of disciplinary founders, my interest is more in the ways in which disciplines can embody, foster, and make productive significantly different methodologies and outcomes in ways that most interdisciplinary research programs do not. Tensions may exist in research programs, but the imperatives of grant and foundation funding tend to homogenize those projects in ways that disciplines are not homogenized. In this chapter, I wish to isolate and consider some methodological questions in existing practice within ethnophysiography. These questions give rise to some theoretical issues as well, but I will bring those up in the context of method, rather than the other way around. New disciplines tend to arise as researchers from various fields have a sense that some object of analysis is not being properly served within the confines of existing disciplines, but the methods of those disciplines in cooperation still have something to offer to address the overlooked object (areas such as Women’s studies and African-American studies are good examples here). As Jill Vickers points out (Vickers 2003), interdisciplinary activity can have one or more of several goals – cooperation, cross-fertilization, fusion/synthesis, explanation (of some object of study), transformation of knowledge, or the transformation of society. Ethnophysiography is an attempt at (at least) the synthesis of knowledge from different disciplines, produced in different ways for different purposes, and an attempt at explanation of the logic of forms of geographical representation across cultures. Vickers argues that interdisciplinary projects bear the marks of their formation, particularly in projects that are culturally contextualized. So, the structure of this chapter will follow some methodological turning points identified within ethnophysiography as it is practiced. These turning points are questions, and depending on how the questions are answered, different methods and outcomes emerge. The questions are as follows: 1. 2. 3. 4. 5.

What is being analyzed? What issues arise when gathering data? How are questions framed to access data? What is the goal of ethnophysiography? How is human meaning connected to human expression, in the context of language about landscape?



Philosophical issues in ethnophysiography 103

6. What are the implications of applying the reconstructed data? 7. If ethnophysiography is to be seen as a nascent discipline, how does it relate to other disciplines? There will only be space to address a few issues with each question. This will be enough, though, to allow us to tease out some fault lines within this field. These fault lines are not meant to divide some practitioners against others, or to suggest that one choice is the right one and the others are not, but rather to make this discipline as productive as possible. Productivity within a discipline occurs not when there is a technology of knowledge, that is, a well-oiled machine in which each part works toward a common goal, but rather when there are differing ways of understanding the object of investigation and the appropriate tools for research. To use an example, anthropology’s theoretical advances came more from its multiple theoretical approaches (both in terms of the disparate areas within anthropology and in terms of theoretical approaches such as structuralism, functionalism, Marxism, phenomenology and so forth). It was one thing to apply tools to a new cultural group or a new archaeological site, and another thing to debate about how that group or site should be interpreted, what counts as an observable entity, and so forth. Disciplines bear within them those productive differences. These differences can be relatively benign, understood as differences of approach or emphasis, or they can be deep and acrimonious, to the point that those in some camps reject that others are even part of the discipline at all (for some, the rift between analytic and continental philosophy has had this character). The productive outcome of these differences, though, has been to clarify and define methods, goals, and assumptions, and advance the discipline. It is worth noting in passing that this chapter can be seen in tandem with another one on methodology in ethnophysiography (Turk et al. in press). That chapter works from specific projects, and focuses on significant questions in the collection of data. What follows here is meant to extend some of the issues raised there, and flesh out the methodological process and identify turning points from beginning to end of the process.

1.

What is being analyzed?

Is the object of ethnophysiography the analysis of landscape, or is it the analysis of the meaning and organization of landscape? This makes a difference, since the first assumes that we start from a given world that precedes cultural mediation, and even if the only access we have to that world is a set of cultural forms, the ultimate goal is to describe the world beyond the cultural forms. If, on the other

104 Bruce B. Janz

hand, the goal is to describe the meaning and organization of space or place, then we are rooted in culture, and even if there is something beyond that which is being described, the point is not to try to describe it apart from the cultural forms. This distinction clearly makes a difference for later modeling techniques, but it even makes a difference for the early parts of the field research. To clarify this, consider the difference between older versions of structuralist anthropology and more recent phenomenological anthropology. Both involve fieldwork, and superficially at least that fieldwork may seem to be similar in both cases. But the point of doing fieldwork in a structuralist theoretical agenda was to establish the invariant foundation of the culture, that which made it stable and provided the “grammar” for meaningful action. Individual actions were assumed to exist for that purpose. If you asked an individual why he or she did something, though, they would not give you an account involving the stability of social structures, but rather an account that involved individual interests, desires, and so forth. They may be unaware of those larger structures. Engaging in fieldwork, then, amounted to a kind of hermeneutic of suspicion, that is, disbelief in the explicit account that an informant would give, and belief in an account unavailable to most informants. Phenomenological anthropology, on the other hand, has no such hermeneutic of suspicion. There is not an underlying invariant structure unavailable to the informant. And, as a researcher, assuming there is one would render the real universalizable experience (ironically) unobservable. How is this relevant? If we assume that the object of analysis is space or place, we may well have a hermeneutic of suspicion toward the actual informants. Their way of parsing landforms will necessarily be partial and contingent, and the only observer able to truly see reality would be the one who can put that partiality together with other partial accounts and, through some Platonic dialectic, inch towards the real. But if the object is the meaning and organization of space and place, there is no Platonic form that all partial accounts lead to. And this distinction will make a difference when gathering data. Just as with structuralism, assuming that all accounts are contingent means that when the researcher has a sense of what the real is, deviations from informants will be seen as errors. Under phenomenology, on the other hand, deviations from the researcher’s expectations may be seen as new and useful information in a diachronic or changing system. This distinction leads to another question, still about the nature of the object of analysis. To what extent do the objects of analysis depend on either their integration into a cultural space or into a viewer’s world? Michael Worboys points out the importance of nearness as a feature of any comprehensive ontology of space, which would necessarily at least minimally place an experiencer into relation with the objects experienced (Worboys 2001). Martin Raubal (and before



Philosophical issues in ethnophysiography 105

him, Barry Smith) points out that it is not just landform terms that are at issue, but their use within culture, such as wayfinding, that matters (Raubal 2001; Smith 2001). And David Mark and Andrew Turk point out that spiritual and metaphysical aspects of peoples’ connection to the land are not incidental, but central to understanding landform terms (Mark & Turk 2003). For more on the implications of phenomenology for the collection of data concerning landscape, please see the chapters by Turk and by Murton in this volume. All of these argue for an approach to data collection that has some element of phenomenology involved, even if there are elements of empiricism as well. For further discussions of connections between phenomenology and landscape, see Turk (this volume) and Murton (this volume). The complexity of the object of analysis goes further even than this if we consider the nature of the relationship between space and place. Michel de Certeau speaks of space as a “practiced place,” by which he means that places are turned into spaces through the practices of the people who inhabit them (de Certeau 1986:â•›117). A street, to use his example, is a place which is “read,” and as such transformed, by different walkers. The street at 10 a.m. on Monday is a different space than the same street at midnight on Friday, and that difference may come complete with different names for objects in the place. If this is true, then ethnophysiography will have to take into account the fact that the same cultural place may be many spaces, depending on the practices, groups, and times of the day, month, or year, among other variables. This is more than just identifying the landform terms that different age groups or genders might use, it is a matter of recognizing meaningful constructions of space by whoever might be there to read them. It is worth noting that the relationship between space and place could be taken differently than I have outlined it here. Elsewhere (Janz 2005), I posit four approaches to place in the academic literature across multiple disciplines. De€Certeau falls into what I call the “symbolic and structural” version of place. Those who draw on phenomenological and hermeneutical sources approach place with the assumption that it is the modality of human experience loaded with meaning, and space is a relatively abstract, schematic or derivative term. We could also look to the “social constructivist and Marxian” approach, typified by Henri Lefebvre, and the “psychological and determinist” approach, often used within environmental psychology, as other ways of thinking about the relationship between space and place. The point here is that already we have different assumptions about how place works, and how it relates to space, across different disciplines and sometimes within the same discipline.

106 Bruce B. Janz

2.

What issues arise when gathering data?

In a certain sense, the issue of data gathering is the province of the disciplines involved in the process. Anthropology and linguistics, particularly, have wellÂ�established methods for fieldwork research. So, the issue here is not to question those methods, but rather to ask how that material can be gathered in light of the overall research program. How does the fact that ethnophysiography is not just anthropology or linguistics affect how those well-established methods do what they do? Are there different assumptions that must be brought to the observations? Does one, for instance, have to be more deliberate about the potential use of this material in a formal ontology, or in a cross-cultural project, than one might if the goal is just an ethnography or the analysis of a specific language? How does the structure of fieldwork have to change to accommodate the eventual use of the data and the communities who will use it? A central issue, one already recognized by many working in ethnophysiography, is that of mereology, or the question of the relationship between wholes and parts. In other forms of cultural naming, the issue of what constitutes a discrete object may be uncontroversial, and the issue then becomes how names emerge, relate to each other, carry cultural meaning, and so forth. In the case of geography, the nature of the discrete objects is not at all clear. What is an object, as part of the whole landscape? The nature of the parts may well differ in different cultures, and so a study of naming will have to begin with the problem of how different cultures parse the physical world. But there is another whole/parts issue at stake. Given that ethnophysiography is irreducibly about culture, the hermeneutical circle between wholes and parts applies. Briefly, this refers to the fact that any understanding within a cultural situation requires that we have both an understanding of the whole, that is, the overall system in which parts make sense, and the parts which assemble to create the whole. Each of these is needed as a prerequisite for the other. We might be tempted to suppose that, while the parts are clearly based on convention (they must be, given the mereological problem just mentioned), the whole is “natural,” since no matter how the parts are divided up and named, it is one natural world that is being divided and named. And yet, this goes back to the earlier issue of whether we are analyzing space or the meaning of space. If it is the first, we will tend to see the whole as natural; if it is the second, the whole is also conventional (even though it describes a shared reality). We might call the whole the “world,” in the culture’s sense of world, and the parts are landforms. The whole is not just the addition of all the parts together, but is the totality of meaning that a culture has about its physical space. Analysis of the landforms will require some sense of what it means to be competent in a world, that is, to move through it and interact with



Philosophical issues in ethnophysiography 107

others in it. To some extent, in fact, it may require some sense of what it means to be creative in that world. Asking about what exists in a cultural scape may have to be supplemented with asking about what is not in the cultural scape. Are there terms for what is not in the world, what is wholly other? Such spaces often find definition in terms familiar to the world itself, even as the West defined Africa in the 19th century as the “dark continent,” the inchoate and wild other to Europe’s order and civilization.

3.

How are questions framed to access data?

Questions of mereology do not exhaust the issues around the collection of data. There are the classic problems in fieldwork, of how to make questions intelligible in a cultural context that is not the researcher’s. The goal of ethnophysiography is not to produce an ethnography, though, but to produce a data set. Anthropological method tends toward producing knowledge in narrative form, and linguistic fieldwork tends to produce knowledge about the structures of communication within a dynamic cultural context. The desired goal in ethnophysiography is different from both of these. One issue that will be even more significant in this context than its original anthropological context is the emic/etic issue, or the issue of knowledge from the inside vs. knowledge from the outside. Are we looking for knowledge that is recognizable to the meaning-structures of participants, or knowledge that ties to something outside of that? Structuralism, for instance, is generally an etic form of knowledge, particularly when focusing on la langue. Hermeneutics at least strives for emic knowledge, if only in the end emic knowledge of one’s own space. The issue is that, for the dual goals of knowing and doing to be successful, both forms of knowledge are essential. We must know about a culture’s sense of its own place-meaning – that is emic knowledge. But we must also make it interoperable with other forms of place meaning, including our own – that is etic knowledge. How do we move from one to the other? It is not as simple as it may seem. It requires a sort of dual consciousness. To use an example – if I was to ask someone why she was getting married, she might say that she loved her partner and wanted to spend her life with that person. That is an emic response – it appeals to knowledge within a culture about that culture’s practices. If, on the other hand, I was to ask a functionalist ethnographer why people get married, they might tell me that marriage contributes to the overall stability of society, and therefore is encouraged within that society. That is an etic response. Note that the person getting married may not know or care about the etic understanding at all. If that

108 Bruce B. Janz

person was also an anthropologist, she might be able to give both the emic and the etic response, but moving from one to the other requires a shift of perspective and engagement. If we move back to ethnophysiography, we can see the issue at stake here. If we gather a culture’s knowledge about place, either in its generic or proper form, we are accessing a body of meaning that is bound together with itself and its peoples’ experiences. It is possible to gain access to that meaning by those outside the group€– participant observation attempts to do this, and at its best is quite successful at it. But the issue here is the move from the emic to the etic and back again. One way to handle this would be dialectically. Paul Ricoeur (Ricoeur 1976) proposed something like this when he argued that every form of understanding began as pre-understanding, analogous to speaking (this is the emic moment), and then moves to textualization, analogous to writing (the etic moment), and finally back to comprehension, as the text that has followed the rules that all texts follow is released back to the inquirer. The emic becomes the etic, which again becomes the emic. Put in terms of ethnophysiography, the initial field research could be based in phenomenology, producing an initial “guess” about the emic, or insider meaning. Putting the data into an ontology is analogous to the move to the etic or outsider meaning, in the sense that the structure of the meaning could be judged by other structures from other cultures. This could raise questions of consistency, or pose other questions to ask to the informants. The entire formal structure could then be brought back to the original informants, along with the new questions produced by textualizing the data, and further study could then be done to see how the new information relates to their earlier pre-understanding of landform terms. A dialectical method such as this could harness the strength of empiricism with the insight of phenomenology.

4.

What is the goal of ethnophysiography?

Clearly, if this nascent discipline is driven by the desire or need to produce a particular outcome, this will affect the way that data is collected. If the goal was to produce an ethnography, or to do philosophy-in-place (that is, to examine the concepts that respond to the needs of a place, and which are created to fulfill new needs), the approach to research will be different than if the goal is to produce empirical knowledge in the traditional sense. It will be different again if the goal is to produce a model that holds the hope of at least interoperability, if not a matrix of general translatability across cultural contexts. Put most simply, ethnophysiography may have as a goal to describe a set of objects or a set of relations (or some other phenomena), it may have as a goal to interpret any of these (which raises the



Philosophical issues in ethnophysiography 109

question, interpret for whom, under what conditions?), or it may have as a goal to explain the data (that is, either to use a classical causal account which would show how the objects or relations came to be, or use a more structural account that would make clear a syntactical or semantic order), or it may attempt to do a combination of these. Ethnophysiography has several options for representation. The most obvious, and that which has the most research, is a formal ontology. I will have more to say about this in a moment. But there are other options. One which is already being formulated is the “Pictorial Landscape Dictionary,” which indexes landscape terms and toponyms with visual examples or exemplars. Ethnophysiography could also be represented through traditional ethnographies, particularly if those ethnographies were allowed to go both ways, that is, scholars in two different cultures wrote ethnographies intended for their own culture, focusing on landform terms in the other culture. This would establish interoperability between those two cultures (although not more than that, of course) but would not be based on a formal ontology which formatted the terms in an abstract matrix. Since ethnographies are stories told for an audience, setting up reciprocal ethnographies for separate scholarly audiences, and then using those as the basis for translation, would enable ethnophysiography to highlight the rhetorical, mythical, and historical elements of landform terms (depending, of course, on the kind of ethnography that is done). In other words, the ethnography could be seen as not just part of the data gathering process, but part of the modeling process as well. Another approach, again based in ethnography rather than ontology, would be to construct what I will call a “synthetic ethnography.” Tim Ingold’s Lines: A Brief History (Ingold 2007) is an example of this. In it, he starts from a concept rather than a specific culture, and draws on much more than the discipline of anthropology to tell his story. The purpose is not to attempt some cross-cultural history of ideas (as might be done if he had started from a thicker concept such as home or region), but to use what is essentially a feature of inscription on the world, and ask about the range of ways that we have of creating, interpreting and following these lines in, for instance, written language, music, weaving, mapping and wayfinding, and constructing genealogies. This kind of synthetic ethnography could represent ethnophysiographical information by using as thin a concept as possible, and use it as a connecting thread to generate new narratives from cultural informants. I mention these not to advocate them, but to merely point out that ethnophysiography and formal ontologies do not necessitate each other. There is no question that an ontology can exist without the data collection methods that ethnophysiography uses, and the opposite is also true. That means that a formal ontology is a choice, and that choice makes some things available, and allows others to recede into the background.

110 Bruce B. Janz

The formal ontology has received the most attention as a model, no doubt because it yields the most immediate and practical outcomes, and in a brief chapter the most we can do is to point to some discussions that already exist and raise a few of the many issues that could be raised. Barry Smith in particular has sketched out in numerous places what an adequate ontology would look like in the case of geography (a significant consideration, given that issues of mereology, or part/whole relations, and the nature of borders have a particular character in the geographical that differs from other realms). David Mark, sometimes with Barry Smith and sometimes with Andrew Turk, is working out ways to link the ontology with cultural landform terms. There are questions that can be raised about ontologies, even beyond the obvious one of whether they can adequately capture a culture. Let us suppose that, for the intended purpose, an ontology is theoretically adequate to represent landform terms and the worlds in which they have meaning. In other words, the meaningful aspects of the world in question are in principle representable using a semantic network that relies on metatagged elements plus some learning capability. There are still questions about both the means of representation and the results we may intend for the matrix. Do we (or can we) metatag reality the same way we can tag created artifacts? If we suppose an Aristotelian version of reality, we might think that is possible, and metatags would be analogous to Aristotle’s concrete universals. But there is potential danger with that approach – we could easily mistake the contingent for the necessary, and we might mistake a particular for a universal. In other words, tagging may work better for worlds which are not culturally contingent than it does for ones which not only assign different terms for things, but which may differ on what counts as a “thing,” and which may have variability even within the cultural group based on (for instance) the position of some meaning-experiencer within the group. Particularly in the case of toponyms, there may be subcultures within a cultural group that assign very different senses to landforms from the “mainstream” of a culture, and that could connect land elements together or take them apart in ways differently from others within the culture, not to mention differently from other cultures. The profile of a range of hills could be regarded discretely by some, while others may treat it as one element through an anthropomorphization such as “reclining woman.” Ontologies can certainly capture some of these sorts of features, but the question is, is there a limit to what the ontology can do? Can it capture historical contingency, subcultural complexity, and competing political claims to land which might result in different names or descriptions? And, if it can, does it even make sense to ask this of the ontology? How fine-grained can the ontology become before it becomes unwieldy? Is there, in effect, always a trade-off in constructing an ontology, between the reflection of



Philosophical issues in ethnophysiography

the subtleties of meaning in a shifting historical and cultural world and the usefulness of the ontology as a tool for interoperability between cultures? A problem with conventional methods of populating an ontology is that most of the data comes filtered through researchers. In itself, this is not a problem, but it becomes a problem if there is no correcting mechanism apart from that of other researchers. Ideally, an ontology would be dynamic, in the sense that it would be able to absorb and evaluate new data from sources other than researchers (e.g., the original subjects themselves). By “evaluate,” I mean that it would come with a statistical model to assess the likelihood of a new piece of information fitting the existing structure. That could be done based on the number of informants who would agree on a term, or on the ways it might cohere with other information and corroborate existing information. This is by no means a perfect approach, of course, for like any system based on coherence it would become increasingly conservative over time, and there may be problems with the number of informants available to corroborate the data. There are also conceivably ethical issues to address with such statistical corroboration – what if some terms or names are protected or restricted even within a group? But it would be one way of providing a corrective mechanism for ethnophysiographical methodology that would not already be committed to the success of the existing method or project. It is important to note that none of this suggests that ontologies should not be used. In fact, they are becoming more sophisticated all the time, and as a modeling device they can accomplish a great deal. Like any model, though, an ontology comes with assumptions and limits. Finding ways to model ethnophysiographical data apart from ontologies could serve to make the ontologies more robust. Alternative forms of representation could serve to orient an ontology toward the most useful level of representation (that is, most useful to those striving for interoperability, as well as those from within a culture who may be more interested in a faithful representation). This is what I meant earlier, about decisions being made at points along the methodological way. Cultivating multiple modes of representation would make all the modes stronger and more nuanced, in the same way that any discipline becomes stronger when there are multiple internal rival methods and tools.

5.

How is human meaning connected to human expression, in the context of language about landscape?

A central feature of ethnophysiography must be that it is “ethno,” that is, that it is rooted in the cultural expressions of landform terms. This recognition separates it from a descriptive geography that ignores the problems of mereology, and also

111

112 Bruce B. Janz

might ignore the fact that even if we are able to label landforms across cultures, that does not take into account the fact that some of these landform terms matter in different ways to different cultures. This second issue may seem like an issue of application rather than research, but in fact, it will be relevant to research at least in the sense that the data sets are being generated from humans, embedded in culture, who are already engaged in the application of those terms. Since the question of human meaning will have to be addressed right from the beginning, and not just as a practical afterthought, we will need to think about how this affects the methods of ethnophysiography from the ground up. If we suppose that a classic positivist approach will work, in which the objective world is abstracted from human concerns and analyzed and described, we will necessarily have to take a stand on the question of human and cultural meaning, that it is irrelevant to the enterprise. As such, there could be no “ethno” in ethnophysiography, or at least, the cultures being used as the source of the data set would have to be regarded as a static, publically available object. But ethnophysiography seems to have built in from the beginning a resistance to true positivism. That is not the central issue. We do have another issue, though, which is the tension between empiricism and phenomenology. These two represent different ways of thinking about the relationship between meaning and what is observed. Both rely on the observation of the real (as opposed to something like idealism, for instance, or some deep constructivism which would diminish observation in favor of some privileged perspective). Empiricism begins from description of the sensory, but as Hume pointed out, that description may well be limited, and may be unable to account for certain contents of consciousness. This is a central theme of Hume (2000). Phenomenology begins from experience rather than sensation, and assumes that both the sensory input from the world as well as the knower receiving that input are secondary to the experience itself, and need to be bracketed in favor of examining the experience. Husserl’s goal was a universal phenomenology, a Cartesian starting point that would transcend culture; for further discussion, see, for instance, Husserl (1960). Heidegger, on the other hand, posits a phenomenology that recognizes that we as knowers always already find ourselves in a world of meaning (cf. Heidegger 1962). It is worth thinking a little about the place and potential of various forms of phenomenology within ethnophysiography, and also the possible limits of the approach. As already indicated, phenomenology has been applied to several related theoretical approaches and methological tools since Husserl (we will disregard earlier uses of the term in Kant, Hegel, and others). Husserl’s goal was to describe the universal structure of consciousness as it is constituted by the givenness of the experience of phenomena (which is the only kind of consciousness available to us). Accomplishing this required an increasingly complex set of steps, as



Philosophical issues in ethnophysiography 113

consciousness was weaned away from uncritical metaphysics, psychologism, naive empiricism, and so forth in order to return “to the things themselves.” Husserl’s dream of the explication of the universal structure of consciousness was revised (Heidegger), critiqued and extended (Derrida), focused on subjectivity (Sartre, Marcel) and different aspects such as embodiment were foregrounded and further developed (Merleau-Ponty), but the commitment to the primacy of experience remained central. Phenomenology, particularly in its hermeneutic variation, has a great deal to offer ethnophysiography. If phenomenology pushes the researcher to a kind of self-critical modesty, which brackets the unspoken and unrealized assumptions we bring to observation (always a problem, but especially in cross-cultural situations), hermeneutics begins from the idea that we are always already in a meaningful world, standing in the midst of a milieu of concepts and practices which both allows us to become manifest in the world and obscures other aspects or possibilities of ourselves and others from our view. The real potential of phenomenology is in its self-critique, that is, in the mechanisms inherent in it for reframing questions about another culture. It is in principle possible to construct a picture of another culture which is correct, but which is not in fact true. The Ptolemaic geocentric view with its deferent and epicycle model, for instance, could be amended and patched over time to yield a tolerably good match with observations, but it took a fundamental recognition that something outside of the observations, something brought by the researchers, stood in the way of a more adequate account for the things themselves to be made apparent. Phenomenology and hermeneutics enable researchers to question their own questions and interrogate their own assumptions, as much as it enables the interrogation of another culture. Indeed, questioning oneself and questioning another cannot be taken apart. While classical scientific method rooted in positivism has mechanisms for self-correction, phenomenology is more rigorous in this. It enables critique not only of the method, but of the researcher’s position. Critique does not imply finding fault with the lifeworld of the researcher, but it means recognizing the specific nature of the partiality of that lifeworld, and, in so doing, formulating better questions to ask of other lifeworlds. The tension between empiricism and phenomenology in approach mirrors the tension between the emic and the etic, mentioned earlier. Empiricism gives us a view from the “outside” in the sense that observation precedes meaning and is available to anyone. Phenomenology gives us a view from the inside, in the sense that metaphysical commitments need to be bracketed and the meaning of the experience becomes central (although it does not give inside knowledge in the sense of particularized knowledge, at least if the goal of universality is still held). Each

114 Bruce B. Janz

of these could serve as the basis for viable models, and they could be combined as I suggested earlier. Clearly, toponyms for geographical forms are embedded in contingent history. We might imagine landform terms to be completely divorced from such history, but even if someone thought this was true in European languages, there is no guarantee that it is true elsewhere. What is more interesting, though, is the possibility of a sliding scale from purely descriptive terms to terms which are necessarily dependent on or related to other cultural values or concepts. Descriptiveness, in other words, may well depend as much on contingent narrative cultural history as it does on picking out features of landforms, or recognizing relationships or temporal elements within and between landforms. And the continuum may not run between generic terms and proper names to have a basis in narrative. It is quite possible that landform terms could reference spiritual or metaphysical beliefs, about beings who dwell in the kinds of places named in particular ways. In other words, landforms that to outside eyes might look morphologically identical could nevertheless have different names based on some narrative content known by a particular culture. A place might, in other words, have a genius loci which renders it specific in a culture, without making it a conventional proper name. In other words, in the methodological path from data collection to analysis to the construction of an ontology, we might imagine that there is a fairly direct empirical path, or we might see there being cultural content at each point along the way that would require a kind of reflexive turn on the method of the researcher. We might suppose that frequency of use might correlate to importance or centrality to a culture, but even that does not necessarily hold. If the place term is descriptive, it might, but if the place name is narrative or cultural, an infrequently used term could designate sacredness or uniqueness rather than lack of importance. And, in some cases, cultural and spiritual topics may be inappropriate to discuss with outsiders, or may be limited to one gender group, ceremonial situation, time of day or year, and so forth. These are things that researchers in linguistics would of course recognize, but the issue here is not linguistic research but ethnophysiographical. In other words, if we see the goal of this research not as the description of the structure of a language, but the construction of a matrix of interoperability or translation, we would have to deal with the problem of what frequency or lack of frequency means.

6.

What are the implications of applying the reconstructed data?

If there is a fork in the disciplinary road with the question of the ways the data is rendered intelligible, there is a further fork when we think about the application



Philosophical issues in ethnophysiography 115

of the data. Werner Kuhn argues that “the tasks (or activities) to be supported by a GIS should determine the entities that are admitted to its languages” (Kuhn 2001:â•›615). If that is true, this has the dual effect of making an ontology (supposing that that is the model) less abstract and rigid, but also potentially more driven to preselect appropriate data and, potentially, ignore what does not already fit into a particular use or goal. What is the goal of ethnophysiography – to know or to do? In other words, does this method ensure the construction of a body of knowledge about the comparative ways which cultures have of coding their space? Or does this method aim to provide an interoperability matrix (or even a glimpse at a universal), anchored by the presumed objectivity, or at least intersubjectivity, of the land? Is there a functional difference between interoperability and universality? Is the model a linguistic one, in which a culture uses a subset of all possible human sounds (called “phonemes”) to produce language? To what extent will cultural meaning be translatable across borders if we have a formal ontology? It may seem that the two goals of knowing and doing can be accomplished together, and to some extent they can; but it is worth noting that having knowing as a goal will affect the entire process of questions and data interpretation, as will the goal of doing, that is, the production of an interoperability matrix. Doing requires that the provisionality of meaning be constrained – the contextualization of terms cannot go on forever – whereas knowing in principle allows meaning to exist more fluidly and contextually. A matrix requires some level of stability, and if that is the goal of research from the beginning, that will affect how questions are asked and what kinds of variability are allowed for. If the goal is to know, there are several options about what is being known. Knowledge might be about particulars, that is, about how specific cultures represent landscape in their imaginative universe. Secondly, the knowledge might be about commonalities. That is, one might be trying to make an argument about what is shared in cultural representations of landscape. Or, thirdly, the knowledge might be directed at demonstrating that there is a unified substructure to language. Given the variability of language in most situations, one might suppose that charting the diverse representations of landscape will give evidence of an invariant universal. It is worth noting that this might just as well be an assumption as a possible outcome. If, on the other hand, the goal is to do, there are some other possibilities. The desired outcome may be to create a matrix of interoperability. In other words, one may want to find an ontology robust enough to be able to contain the diverse representations of landscape in one matrix. That matrix would potentially enable localized translations, that is, translations between two language games, based on

116 Bruce B. Janz

a set of equivalences in either objects or relations. Or, alternately, the desired outcome may be universal translatability. This is a step beyond interoperability. One way to think through the philosophical issues in ethnophysiography is to regard it as a set of methods inherited from other disciplines, and which were developed as answers to different questions than what is being asked now. These methods do not seamlessly dovetail together, but exist in tension. Ideally, that tension would be seen as constructive. Method, of course, leads to an outcome, and, as has just been indicated, there may be different outcomes depending on the researcher. In particular, there may be a difference between those who see ethnophysiography as way of knowing something (about specific cultures or about a putative universal substrate, accessible through place words), and those who see it as a way of creating the basis for a matrix of interoperability or translatability that can connect GIS data to cultural forms, that is, as the basis for doing something. We might see these two goals as related – first we know and then we do. But superficially, at least, we could see these two reversed – in some cases, we do first, and then we know. In this case, some of the data gathering about cultural place words may occur in the context of already trying to implement an interoperability matrix, and then finding the limits of it. But more important than that: it is possible that the goal of doing (creating a matrix) may stand at odds with the goal of knowing (accessing cultural knowledge). Or rather, prioritizing the goal of doing may determine the kind of knowledge that we have access to in a culture. What use might the results from ethnophysiography be put to? Certainly, we can imagine that a catalog of landforms across cultures could be useful in producing a translation table for recognizing and naming significant entities within a domain. GIS systems could begin to link together data from different cultural spaces, which could provide the basis for comparisons of land use. There are potential dangers, though. If we imagine that data generated by ethnophysiography is “mapped” in a manner that approaches a universal form, we might come to believe that the knowledge has no philosophical presuppositions attached to it. The fact that we can translate generic landform terms across cultures tells us nothing about how cultures value those landforms, or how they fit into a culture’s self-identity, or even what a culture thinks about issues of ownership, stewardship, and use of the land. These might seem like ancillary issues to the concerns of ethnophysiography, but they are not. Universalized “realist” projects which purport to describe reality in a dispassionate manner have often been used as pretexts for the imposition of external narratives on other cultures. V.€Y. Mudimbe (1988) aptly describes the “invention of Africa” as the result of a scientific gaze that was insufficiently aware of its own assumptions and desires.



Philosophical issues in ethnophysiography 117

This is another decision in the methodology of ethnophysiography. It seems that as this discipline develops, there will be a methodological strand which will emphasize the production of data, the formalization of that data, and the practical application of the results. And there will be a critical side, which will emphasize the token over the type, the place of the cultural narrative in relation to the description.

7.

If ethnophysiography is to be seen as a nascent discipline, how does it relate to other disciplines?

David Mark and Andrew Turk use the analogy of other “ethno” scientific disciplines, such as ethnobotany or ethnozoology, as the models for this new discipline (Mark & Turk 2003), even as they recognize differences between these other disciplines and ethnophysiography (ibid.). The two disciplines just mentioned are interesting choices – neither of them have the problems of mereology that were outlined earlier, and neither have had the expressed goal of building a formal ontology or providing the basis for interoperability. They have, however, been seen as the means to a scientific end, which is to translate folk knowledge about the natural world into something scientifically testable and, hopefully, usable in a pharmaceutical or other commercial product. There is another “ethno” discipline worth mentioning: ethnophilosophy. It is a contested term within African philosophy. Paulin Hountondji coined it to describe what he saw as the prevalent mode of African philosophy in the mid-20th century, which saw philosophy in Africa as an anonymous, collectively held set of beliefs (Hountondji 1996). He argued that we would not regard this kind of activity as philosophy in Europe or anywhere else, and neither should we in Africa. African philosophy should have the same kind of rational, critical, individualized character that any other philosophy has. The reason for raising this is that the term “ethno” could be seen as a generalizing one, and one which assumes a particular character of culture. It may, for instance, assume that culture is static, and that it is a repository of relatively selfcontained wisdom. The formation of a new discipline invariably involves adapting methods from existing disciplines. New disciplines are routinely criticized as superficial compared to existing disciplines, in part due to the work it takes to adapt existing methods and generate formative texts and thinkers. Jill Vickers’ paper, mentioned at the beginning of this chapter, categorizes interdisciplinary efforts into various categories, but she stops short of addressing the emergence of new disciplines out

118 Bruce B. Janz

of those various formations. And yet, clearly it happens, and has happened since the explosion of discipline formation that began in the late 19th century. It is worth noting that not all new disciplines necessarily emerge as standalone university departments. A single department may well house multiple disciplines – consider studio art and art history within art departments, for instance, or creative writing, literature, and rhetoric/composition in English departments. So, the issue of disciplinarity is not the same as the issue of university structure. The issue in nascent disciplinarity is the emergence of multiple theoretical models and methodologies, which compete for legitimacy (the grounds of which might be quite different in different cases) or divide “territory” of study (as, for instance, in anthropology’s or geography’s subdivisions). The point here is that ethnophysiography has the potential to emerge from its existing status as a research program to being a nascent discipline, in the sense that it can harness the difference inherent in its multiple methodological provenances and construct a variety of productive strands of research. The choices that have been outlined at different stages of the method are not ones to be solved by simply agreeing on which is the right one, but rather, by pursuing all angles.

8.

Conclusion

This survey of questions in methodology barely scratches the surface. The point has been to clarify some options at various points along the way, and to broaden the discussion from focusing solely on the nature of formal ontologies to the entire production of knowledge: from data collection to interpretation of the data, and to the building and application of models or expressions of that knowledge. If there is a central issue that all practitioners within this nascent discipline will have to take seriously, it will be to address the ways in which concepts and methods developed within particular disciplinary spaces, for other purposes, can be altered to be appropriate in this setting. All disciplines will find themselves operating outside of their comfort zone – anthropological fieldwork will not lead only to the production of an ethnography, linguistic fieldwork will be taken beyond the point of the description and classification of a system, GIS experts will have to think carefully about the meaning of the data and not just its properties, and so forth. But there are other issues that could be raised. How do we deal with place names, or toponyms, and what is the relationship between them and generic landform terms? What continuities and ruptures are there between ethnophysiography and the earlier idea of ethnogeography (itself a term that refers to multiple research programs)? Will the use of a formal ontology make possible a semantic web, that is, the ability to dynamically search and construct questions that have



Philosophical issues in ethnophysiography 119

not been already placed within the system by a researcher? What are the differing philosophical assumptions within different existing ontology languages? The fact that there are so many questions is a positive sign of a robust and interesting nascent discipline. It has been the intention here to sketch out a broader account of the method of ethnophysiography than has appeared in most work to this point, and suggest productive directions of research. As more data is collected from more cultures, no doubt new questions will emerge that will help to further refine the various strands of viable research.

References de Certeau, Michel. 1986. The Practice of Everyday Life. Berkeley CA: University of California Press. Heidegger, Martin. 1962. Being and Time. Translated by John Macquarrie and Edward Robinson. London: Basil Blackwell. Hountondji, Paulin. 1996. African Philosophy: Myth & Reality, 2nd edn. Bloomington IN: Indiana University Press. Hume, David. 2000. A Treatise of Human Nature. Oxford: OUP. Husserl, Edmund. 1960. Cartesian Meditations: An Introduction to Phenomenology. The Hague: Martinus Nijhoff. Ingold, Tim. 2007. Lines: A Brief History. London: Routledge. Janz, Bruce. 2005. Walls and borders: The range of place. City and Community 4(1): 87–94. Kuhn, Werner. 2001. Ontologies in support of activities in geographical space. International Journal of Geographical Information Science 15(7): 613–631. Mark, David & Turk, Andrew. 2003. Ethnophysiography. Paper presented at the Workshop on Spatial and Geographic Ontologies, 23 September, 2003 (prior to COSIT03), in Ittingen, Switzerland. Mudimbe, Valentin Y. 1988. The Invention of Africa. Bloomington IN: Indiana University Press. Raubal, Martin. 2001. Ontology and epistemology for agent-based wayfinding simulation. International Journal of Geographical Information Science 15(7): 653–665. Ricoeur, Paul. 1976. Interpretation Theory: Discourse and the Surplus of Meaning. Fort Worth TX: The Texas Christian University Press. Smith, Barry. 2001. Objects and their environments: from Aristotle to ecological ontology. In Life and Motion of Socio-economic Units, Andrew Frank, Jonathan Raper & Jean-Paul Cheylan (eds), 79–97. London: Taylor & Francis. Turk, Andrew, Mark, David, O’Meara, Carolyn & Stea, David. In press. Geography – documenting terms for landscape features. In Oxford Handbook of Linguistic Fieldwork, Nicholas Thieberger (ed.). Oxford: OUP. Vickers, Jill. 2003. Diversity, globalization, and “growing up digital”: navigating interdisciplinarity in the twenty-first century. History of Intellectual Culture 3(1). (1 January 2010). Worboys, Michael F. 2001. Nearness relations in environmental space. International Journal of Geographical Information Science 15(7): 633–651.

chapter 6

‘Land’ and life Ethnoecology and ethnogeography  as complementary approaches  to the analyses of landscape perception Chris S. Duvall

Understanding how people classify physical geographic features is necessary for identifying cross-cultural geographic concepts necessary for successful communication of landscape knowledge. Identifying cross-cultural geographic concepts will require development of the field of ethnogeography, which employs ethnographic methods to analyze geographic knowledge. This chapter analyzes physical geographic knowledge in the Maninka language of southwestern Mali, and compares Maninka knowledge to that of other cultural groups. The results suggest that broad physical geographic concepts may be shared pan-environmentally, but that most physical geographic knowledge is contained in culturally specific classifications within a broad cross-cultural framework. Academic geographers should expect only broad correspondence between their categories of physical geographic variation and those of people who classify biophysical features according to local knowledge systems.

1.

Introduction

Indigenous peoples manage diverse ecosystems using sophisticated knowledge of the biophysical environment. This local knowledge of the biophysical environment (hereafter, ‘local knowledge’) is mediated through everyday language. Terms that carry ecologically significant meanings often carry meanings in other, seemingly unrelated contexts, ranging from politics to the cultural history (Richards 1993). Such contextually layered meaning, often overlooked by outsiders, is what makes local knowledge gain and retain relevance for its users (Agrawal 2002). Yet outsiders often emphasize limited aspects of local knowledge systems in order to underscore practical applications these may have in natural

122 Chris S. Duvall

resource management. The practical value of local knowledge is not in question, but overemphasizing its potential applicability in specific contexts leads to incomplete characterization of local knowledge (Agrawal 2002). Many researchers have employed an ethnoscientific approach to studying local knowledge, by analyzing particular aspects of local knowledge systems comparable in referential extent to specified scientific fields. Ethnobotany, ethnozoology, and ethnopedology have received much attention, while few have described local knowledge of climate (e.g. Osunade 1994; Ovuka & Lindqvist 2000) and habitat variation (e.g. Fleck & Harder 2000; Shepard et al. 2001). Recently, increased attention has been given to local knowledge of broad portions of the biophysical environment, including landscapes (e.g. Burenhult 2008; Johnson 2000). This is the referential frame Blaut (1979) suggested for ‘ethnogeography.’ Blaut’s concept of ethnogeography is underdeveloped, although his approach avoids a priori compartmentalization of indigenous knowledge through imposed correspondence with technical scientific fields. The theoretical purview of Blaut’s ethnogeography overlaps ‘ethnoecology,’ the study of the interrelated, culturally determined spiritual beliefs that guide resource use, the bodies of knowledge underpinning resource use, the practices of resource management (Barrera-Bassols & Toledo 2005; Toledo 2000). Ethnoecologists see local knowledge as site-specific and complexly layered, and the desire to maintain local knowledge in its environmental and cultural context explicitly motivates ethnoecological research. However, ethnoecologists have subtly continued to compartmentalize local knowledge according to technical scientific criteria, through their explicit desire to identify practical applications of local knowledge (Barrera-Bassols & Toledo 2005). This goal arguably is at cross purposes with that of maintaining local knowledge in sociocultural context (Agrawal 2002) and has led ethnoecologists to privilege certain research questions over others. Specifically, a central goal of ethnoecologists has been to compare local knowledge to particular domains of scientific knowledge (Barrera-Bassols & Toledo 2005), especially soils, although ethnoecologists have not shown and do not argue that scientific knowledge provides a standard for assessing other knowledge systems. Comparison of local knowledge systems across cultures and environments has not been prominent in ethnoecology. Cross-cultural comparisons of biophysical knowledge has been considered mainly by anthropologists using linguistic analyses of folk taxonomy, an approach Blaut (1979) promoted. These analyses have revealed that many ethnolinguistic groups have highly detailed knowledge of specific environments, including intricate classifications of environmental features. Folk taxonomies of plants and animals show, in many cases, universal similarities (Berlin 1992), although biophysical variation between environments and socioeconomic variation between



‘Land’ and life 123

peoples lead to differences in the detail with which specific groups distinguish biophysical features (Holman 2005), including soil (Williams & Ortiz-Solorio 1981; WinklerPrins & Sandor 2003). Does local knowledge of landscapes also vary with environment and culture? I contribute to our understanding of cross-cultural variation in physiographic knowledge by analyzing knowledge of landscape features amongst Maninka farmers in southwestern Mali, and comparing this with published information on other local knowledge systems. Following Blaut (1979), I assert that studying folk taxonomies can help identify widely shared geographic concepts because the necessary linguistic analyses expose patterns of perception and cognition (Berlin 1992; Ellen 1993). However, a folk taxonomy approach to studying landscape knowledge would benefit from ethnoecology’s careful assessment of biophysical phenomena€– just as ethnoecology would benefit from geography’s understanding of place and space. By detailing the taxonomy of Maninka physiographic knowledge, I show that delimiting portions of local knowledge based on external criteria can mischaracterize how people perceive physical geographic variation.

2.

Research setting

During January–December 2004, I conducted field research with Maninka farmers in Solo village (Figure 1), which was established > 400 years ago (Samaké et al. 1986) in the Bafing area of southwestern Mali. About 250 people live in Solo, subsistence farmers who rely on rainfall to grow millet, peanuts, and several minor crops. Residents also hunt and gather wild foods, for home consumption or to supply local markets (Horowitz et al. 1990; Samaké et al. 1987). Farmers follow a complex land management system to maximize crop security under constraint of the region’s variable precipitation regime (Koenig et al. 1998; Laris 2002; Samaké et al. 1987). Most residents have lived in Solo their entire lives, although many spend rainy seasons in small farming hamlets dispersed 5–10 km around Solo. The Bafing landscape is topographically complex. Sandstone plateaus, often incised by deep, narrow gorges, rise 100–300 meters above colluvial slopes and undulating plains (Michel 1973). Several sandstones and other fine-grained sedimentary rocks occur in the area, each having distinct characteristics (Varlet et al. 1977). Sandy soils dominate the land surface, and ferricrete crusts occur widely (Dames & Moore 1992). The edges of these crusts are steep and gravelly, while their upper surfaces may be barren, exposed hardpans or shallowly covered by silty to clayey soil (Michel 1973). Several dolomite intrusions outcrop to form steep, rounded inselbergs surrounded by silty soil.

124 Chris S. Duvall

Research Area M A U R I T A N I A Solo

S E N E G A L

a

M a n i n k a

m

0

50 km

a

GUINEABISSAU

B

BURKINA FASO

Ma

G U I N E A

nk

a u

G H A N A

J

Inset map of research area Manding spoken as first language by majority of population Jula Distribution of dialects mentioned in text

o

SIERRA LEONE

l

nk

ra

Ku

ni

a

THE GAMBIA

Ba

n

fin

a

gR

.

n

M A L I

Mandinka

Manantali Reservoir

C O T E D' I V O I R E LIBERIA

N 0

250 km

Figure 1.╇ West Africa showing research area and Manding language area.  Only dialects mentioned in text are shown.

Vegetation variation corresponds to hydrological and edaphic variation (Lawesson 1995). The area is in the broad Sudanian woodland band extending across West Africa (White 1983). Different woodland associations occur in sandy and silty soils, while grassland types occur in sites with shallow or waterlogged soil (Nasi & Sabatier 1988). Sites with deep soil are favored for agriculture. In sites that have been undisturbed for several decades, forest vegetation forms, dominated by woodland species. Along permanent or seasonal drainages, gallery forest occurs, dominated by species characteristic of more humid climate zones (Duvall 2001; Lawesson 1995). Hydrological resources are not diverse (Dames & Moore 1992). The most important source of water for humans is hand-dug wells. Rainfall is highly seasonal; nearly all precipitation falls between July and October (Barth 1986). There are two semi-permanent creeks, many seasonal drainage channels, and several permanent springs near Solo. The original course of the Bafing River (which was dammed in 1987 to form a major reservoir) is about 20 km from Solo, outside its traditional territory. For most residents the river (including the reservoir) is beyond their normal sphere of experience.



‘Land’ and life 125

3.

Research methodology

3.1

Orthography and language

In this paper, I use terms in English, French, and Maninka and Bamanan, closely related Mande dialects spoken widely in western West Africa. These dialects belong to the Manding branch of the Northern Mande sub-group of Mande, a subfamily of the Niger-Congo languages. Differences between Manding dialects are mainly systematic changes in pronunciation, reflecting historic or geographic separation of Manding populations (Bird 1982). I interviewed consultants in Maninka. The words I collected are written in bold italics, following Bird’s (1982) orthography. For clarity, I provide only singular Maninka nouns, even if a plural is given as a gloss or implied by context; the Maninka plural marker is ‑lu (Bird 1982). Covert concepts are presented as English terms and written in [brackets]. Previously published terms are italicized in quotes, and written as in the cited sources. English glosses are given in (parentheses). Glosses for Maninka words come from field research, while glosses of previously published words come from the cited sources.

3.2

Data collection and analysis

Participant observation offered opportunities to learn the use of Maninka terms, while ethnographic interviews clarified observations. I participated in about 600 hours of relevant conversation, including interviews of 35 consultants, male and female, aged 13–80. After eliciting enough terms to develop a functional vocabulary, I sought to understand relationships between labeled physiographic features and the taxonomy these relationships create. In interviews, I elicited terms for physical features my interviewee and I observed together, then asked how observed features differ from ones I already knew. I also asked consultants to indicate physical features for which I had collected labels, but which I had not knowingly observed. We discussed observed features so that I could determine precisely what a feature term indicated, and collect terms for related features. I consistently used specific question formats to facilitate analysis. Data analysis was qualitative. First, basic grammatical classes often suggested broad conceptual categories. Second, comparative questions – especially triadic and dyadic comparisons – and classification questions – ‘which of this set of features are most similar?’ – enabled identification of covert taxa (Berlin et al. 1968; Kay 1971). For instance, I asked consultants to group landforms. Consultants consistently grouped elevated landforms separately from depressions, but no

126 Chris S. Duvall

consultant provided a term for ‘elevated landform’ or ‘depression.’ The saliency of elevation versus depression means that these categories must be understood as covert to represent conceptual relatedness accurately. Third, listening to conversation semantics and subsequently asking about word meanings allowed identification of polysemous terms. Finally, near the end of data collection, I developed taxonomic structures to represent the relatedness of concepts (Berlin 1992; Kay 1971). I developed the taxonomic structures iteratively based on focused discussions with key consultants. Consultants mostly agreed in their descriptions and classifications of physical geographic features – I sought fairly basic, generalized knowledge – but, as described below, some aspects of my data were not unanimous. I did not assess knowledge variation between social groups.

4.

Maninka classification of the biophysical environment

4.1

Broad structure

The Maninka biophysical environment is conceptually bifurcated into [the biospiritual environment] and [the physical environment]. This paper focuses on a narrow aspect of [the physical environment]; for a full analysis see Duvall (2008). For my consultants, the [physical] and [biospiritual] are inseparable, as in other local knowledge systems (Johnson 2000; Rappaport 1979; Shipton 1994). In the Maninka case, [the biospiritual environment] comprises all beings – hadama‑ dèn (humans), jine (spirits), [animals], and [plants] – and their possessions. [The physical environment] is composed of physical features that indicate natural resources or hazards. Some features are owned, occupied, or otherwise possessed by powerful beings, especially humans and jine. The concept of possession, which connects physical things to beings, causes individual people to see spiritual and social meaning in physical features and can transform these into components of [the biospiritual environment]. Components of [the physical environment] are considered neutrally available for use to all people, but [biospiritual] features are accessible only to people who have socially granted use rights, or who have the spiritual knowledge and power to overcome or appease jine. Anthropogenic physical features are important indicators of natural resources, but these features can also carry attributes of possession, which can limit the accessibility of associated resources. Ownership can lapse for enduring artifacts like abandoned settlements, so that they may become simply physical features. However, due to their past association with humans – including perhaps subaga (sorcerers) or disguised jine (spirits) – these features must be encountered cautiously.



‘Land’ and life 127

ala ka baara

dugu ju kòrò (‘deep subsurface of the ground’) ba ju kòrò (‘riverbed’)

ju kòrò (‘deep subsurface’) san, san hutuma (‘sky’) + 3 more levels funteno (‘temperature’) dugu (‘earth’)

funteno (‘hot temperature’) nènè (‘cold temperature’) sumaya (‘coolness’) [water bodies] + 4 more levels

kuru funteno (‘humidity over damp rock’) bin funteno (‘humidity over damp grass’) dugu funteno (‘heat of the ground’) nining (‘shade under trees’) siniware (‘shade from clouds’)

dugukolo (‘ground’) + 3 more levels [vegetation] + 4 more levels [features made by animals]

mògò ka baara

[landforms] + 3 more levels siya (‘lair’) nyaga (‘nest’) kome (‘salt lick’) [land cover] + 3 more levels [artifacts] + 3 more levels

Figure 2.╇ Main categories of the Maninka physical environment. Line formatting indicates taxonomy for the three subdivisions of [the physical environment]: solid lines = ala ka baara; dotted lines = [features made by animals]; dashed lines = mògò ka baara. Categories without shading belong uniquely and totally to ala ka baara; lightly shaded categories belong uniquely and totally to [features made by animals]; darkly shaded categories belong uniquely and totally to mògò ka baara; categories outlined in gray subsume categories that include features belonging to more than one of the three primary subdivisions. Commas separate synonymous terms. For categories followed by braces, internal taxonomies are shown in the figures indicated.

4.2

The physical environment

[The physical environment] consists of all non-living, physical features of the environment and comprises three major categories (Figure 2). Ala ka baara (the work of Allah) and mògò ka baara (the work of humans) differ, but these two, along with [animal-created features], share some subordinate categories. Ala ka baara and mògò ka baara are unproductive secondary lexemes; baara does not carry a broader meaning equivalent to [the physical environment]. Ala ka baara includes physical features created by ala (Allah), the omnipotent spiritual force. Possession of these features is generally impossible. Ala designates the omnipotent force in the syncretic Maninka belief system, heavily influenced historically by Islam (Brun 1907; Tauxier 1927). Perceiving the physical environment as a powerful god’s creation seems common in local knowledge systems (Forde 1954; Johnson 2000). Mògò ka baara comprises anthropogenic physical features, which may be part of either [the physical environment] or [the biospiritual environment], depending on their ownership status. [Features created by animals] is of limited extent, including only the few enduring features created by animals. Abstractly, animals possess these features, but such possession means little to humans and accessibility to resources in these features is unrestricted.

128 Chris S. Duvall

In the following paragraphs, I describe the narrow aspect of ala ka baara that comprises landscape features. For a full taxonomy of Maninka physiographic knowledge, see Duvall (2008). Ala ka baara directly includes four categories. Features composing dugu (earth) are above those composing ju kòrò (the deep subsurface) and below those composing san (the sky). Dugu, the focus of this paper, comprises physical features associated with the ground, including water bodies and certain microclimatic features. Below, I use ‘terrestrial features’ to refer generally to components of dugu. The fourth feature is funteno (temperature), which is conceptually complex. Funteno permeates all components of dugu, san, and ju kòrò, and is mostly not salient in any particular geographic feature. In some situations, though, funteno is salient enough to create distinct features that are at once within dugu or san and distinct from observed celestial or terrestrial features. For instance, advection fog over damp grass or damp rock is seen as funteno, and several consultants observed, “kuru funteno mun kuru san to” (humidity over damp rock is in the air over the outcrop). Yet upon questioning, all consultants stated that funteno is neither kabo (cloud), kuro (haze), nor funio (air), because “funteno te san na” (temperature is not of the sky). The saliency of physical features subsumed in fun‑ teno is that the sensation these offer humans is distinct from surrounding celestial or terrestrial features. Dugu is the most developed of these categories and directly subsumes five categories in which physical features are differentiated based on topography, hydrology, ground characteristics, [vegetation], and [land cover], a synthetic assessment of site characteristics. In its general sense, dugu is an unusual term in a cross-linguistic context (cf. Burenhult 2008), because it seems to capture much of the meaning of the concept of ‘landscape.’ However, I present my analysis of dugu by focusing on its components, and return to the idea of landscape in my final discussion section. Components of dugu are highly meaningful as indicators of suitability for farming and settlement. Consultants classify [vegetation] by structure or composition (Figure 3). The category [compositional vegetation] potentially includes many subordinate categories because these are distinguished according to the most salient species (Sow & Anderson 1996). In practice, few compositional vegetation types are recognized, either field vegetation or stands of economically important wild plants. [Structural vegetation] types are either tu (vegetation with high stem density and high stature) or kèna ge (vegetation with low stature); people can see long distances in kèna ge, but not in tu. Short grasses dominate kèna ge, which is characteristic of many, but not all, kèna (clearings), a [land-cover] type discussed below. Tu vegetation has high densities of trees, bamboo, or tall grasses.



‘Land’ and life 129

[vegetation]

[compositional vegetation]

tiga he to (‘peanut field vegetation’) nyò he to (‘millet field vegetation’) sita he to (‘baobab grove vegetation’)

[structural vegetation]

soling he to (‘weedy sedge-dominated vegetation’) bin tu (‘grass tu, sutu (‘vegetation with bin jalen (‘dry grass stand’) stand’) high stem density and bin jukutun (‘tall grass stand’) high stature’) bo tu (‘bamboo bo tu (‘bamboo thicket’) thicket’) bo jalen (‘dry bamboo thicket’) yiri tu (‘treeyiri sutu (‘woodland’): vegetation dominated without many nambo (‘lianas’) vegetation’) kèna ge (‘vegetation with low stature’)

yiri sutuba (‘forest’): vegetation with many nambo (‘lianas’)

Figure 3.╇ Taxonomy of the covert category [vegetation]. For diagrammatic conventions, Figure 2.

tun (‘diggings’) [landforms]

wu (‘cavities’)

dinga (‘depressions’)

numubogo tun (‘blacksmith's clay pit’) tun ta (‘construction dirt pit’) gambun tun (‘smelter’) shi dumun tun (‘Ancistrotermes mound’) tun (‘termite mounds’) jemberem tun (‘Eubitermes mound’) bagabaga da ba tun (‘Odontotermes mound’) hurugun tun (‘erosion gully, cut bank’) wu (‘animal holes’) wu (‘crevice’) wòròn (‘pit’) hanhan (‘cave’) dinga (‘three-sided depressions’)

ji jigi silo (‘drainage channels’)

dinga (‘small area surrounded by low hills’) kubo (‘large area surrounded by high hills’) solon (‘large area surrounded by hills’) kunsa (‘head of drainage channel’) bilan da (‘mouth of drainage channel’) foula bèn woula (‘confluence’) ba bulo (‘river valley’) kò bulo (‘creek valley’) gouga (‘gorge’) hulung (‘gully’) wounjan (‘col’) hata hara (‘bedrock fracture in which water flows’) kuruma lanya (‘area between two outcrops’) hara (‘swale’)

Figure 4.╇ Taxonomy of wu (cavities), tun (diggings), and dinga (depressions).  For diagrammatic conventions, Figure 2.

Topography is classified in the covert taxon [landforms], which includes wu (cavities), [depressions], and [elevations]. Several criteria differentiate topographic features. First, many differ according to surface drainage, especially types of [depression] (Figure 4). Many depressions are types of ji jigi silo (drainage channel), distinguished mainly by side-slope form and density of drainage network. Second, microclimate also differentiates [landforms]. Several features have characteristic degrees of shading, such as types of wu (cavities) and types of [depression]. For example, hanhan (caves) contain large, permanently cool areas, while

130 Chris S. Duvall

[elevations]

koti (‘slope’) [topography most salient]

tinti (‘rise’) haya (‘drop’) konko koti (‘hill slopes’)

kuru koti (‘outcrop slopes’)

[landscape position most salient]

[substrate most salient]

konko sinbe (‘hill toeslope’) konko yerete (‘gentle hill slope’) konko sala (‘steep hill slope’) konko karoma (‘hill midslope’) konko mana (‘very steep hill slope’) konko kunte (‘hilltop’) konko julu (‘hill ridgeline’) konko nu (‘hill point’) kuru sinbe (‘outcrop toeslope’) kuru karoma (‘outcrop midslope’) kuru bake (‘bench’) kuru doki (‘outcrop meander’) kuru julu (‘outcrop ridgeline’) kuru kunte (‘outcrop top’) kuru nu (‘outcrop point’) kuru bòn da (‘pass’) kuru din (‘small outcrop’) kuru kilin (‘inselberg’) kuru nyahe (‘skree slope below cliff ’) ba gounga (‘island’) kò gounga (‘creek meander’)

doki (‘meander, peninsula’) gounga dakani (‘shoreline’) gurungurun (‘ford’) goroko (‘rapids’) gongoli (‘hillock’): soil not rocky or gravelly konko (‘hill’): soil extremely gravelly kuru (‘bedrock outcrop’) mana kuru (‘steeply sloped outcrop’) hata kuru (‘gently sloped outcrop’)

Figure 5.╇ Taxonomy of the covert category [elevations]. For diagrammatic conventions, Figure 2.

wòròn (pits) do not. Third, slope form differentiates elevated features (Figure 5). The concept ‘slope’ is covert to many people, but many younger men with knowledge of French through labor migration or radio label this concept koti, from the French côte (slope). Some slope classes differ based on how humans interact with them: a tinti (rise) is barely noticeable when walking, but a haya (drop) cannot be climbed or descended. Fourth, landscape position and substrate also differentiate [elevations]. For instance, both types of goungou (island; meander) have the same shape and soil characteristics as a gongoli (hillock), but the two categories are distinct because goungou occur along permanent water bodies while hillocks do not. Konko (hill) and kuru (bedrock outcrop) differ due to substrate, not slope form: konko are the edges of ferricrete crusts, while kuru are bedrock outcrops. Within the categories of permanent and seasonal, [water bodies] (Figure 6) may differ according to the duration water is present during the year or longer periods of time. For example, both gibingibin and ji ja balo are permanent pools in deep spots in creeks, but a gibingibin is less likely to dry in droughts. Many water bodies are distinguished by size: a ba (river) is larger than a kò ba (creek), and a sakanbe (spring) has more abundant flow than a tondi ji (seep). The origin of water bodies is also important: a kuru bake differs from other flowing water



‘Land’ and life 131

ji da (‘permanent water body’)

[subsurface origin] [surface origin]

sakanbe (‘spring’) tondiji (‘seep’) [flowing] dalo (‘pool’)

[seasonal water bodies]

ji sogo dula (‘puddle’) dunge (‘water in a pit’) lanya (‘seasonal flooding’)

ba (‘river’) kò ba (‘creek’) dalo la sigi (‘pond’) kò bilan da (‘water hole in a drainage channel’)

ji ja balo (‘water in a hole’) gibingibin (‘water in a crevice’)

kò (‘seasonal stream’) kuru kati (‘waterfall’) lanya (‘flooded shoreline’) lamboun (‘flooded drainage channel mouth’) kuru bake (‘sheet flow over sloped outcrop’) [belonging to a drainage channel]

Figure 6.╇ Taxonomy of the covert category [water bodies]. Ji da (permanent water body) and [seasonal water bodies] are directly subsumed in [water bodies]. For diagrammatic conventions, Figure 2. dugukolo (‘ground’)

bèrè (‘gravel’)

bèrè misèn (‘small-grained gravel’) huga kan bèrè (‘medium-grained gravel’) bèrè musa (‘large-grained gravel’) sumaya (‘moisture, nèma (‘soil moisture’) bogo fing, dugu fing (‘loam’) kombo (‘dew’) coolness’) bogo fing, dugu fing (‘loam’) nyama (‘healthfulness’) bogo (‘loam’) bogo fing, dugu fing (‘loam’) bogo (‘soil’) bogon ge (‘silty loam’) tun (‘clay’) dugu welen (‘laterite’) karankaran (‘clayey loam’) kenye fing (‘balck sandy loam’) kenye (‘sandy soil’) kenye welwn (‘red sandy loam’) kenyekenye (‘sand’) kuru (‘rock, stone’) nari kuru (‘dolomite’) [hardness most salient] kaba kuru (‘strongly cemented sandstones’) kenye kuru (‘decomposing dolomite or sandstone’) keleheti kuru (‘fine-grained sedimentary rock’) tuti kuru (‘ferricrete’) jaman kuru (‘clear quartz’) [use value most salient] sanu kuru (‘gold’) ta kuru (‘hard quartz’) san galima kuru (‘celt, Neolithic polished stone axe head’) [form most salient] kuru fing (‘rounded block’) kuru ge (‘angular block’) busu din kuru (‘cobble’) [landscape kuru makan (‘siltstone located in caves’) location most kuru sero (‘laterite located in the deep subsurface’) salient] kuru kare (‘iron-rich precipitates located on creek beds’)

Figure 7.╇ Taxonomy of dugukolo (ground). For diagrammatic conventions, Figure 2.

bodies because its water does not belong to a drainage channel, but comes from drainage through soil overlying exposed bedrock. There are six components of dugukolo (ground), which includes geologic and soil resources (Figure 7). First, nògò (organic matter) is surficial, decomposed litter that enhances the fanga (strength, chemical fertility) of soils. Second, nyama is a gaseous substance that emanates from the ground surface – especially soil€–

132 Chris S. Duvall

that controls the healthfulness of a land parcel. Nyama can be thought of as the spiritual component of the ground, and it can be good or bad depending on the site and the being exposed to it. Crops may grow in sites with nyama that is bad for people, but people cannot safely eat these crops. Third, sumaya (moisture) consists of both nèma (soil moisture) and kombo (dew), which is moisture that has ascended from the ground. Nyama and sumaya are mainly associated with soil, but exposed bedrock also has these characteristics. The final three components of dugukolo (ground) are more finely differentiated (Figure 7). First, there are three types of bèrè (gravel), distinguished by particle size. Second, bogo (soil) is classified based on arability, texture, and color. This category centers on bogo (loam). Bogo and kènyè (sandy loam) are preferred for farming; less preferred and non-arable soils are clayey or silty. This division reflects the demands of local staple crops: millet prefers well-drained soils, while peanuts cannot be easily dug from dry, fine-textured soil. Third, kuru (rock, stone) is classified according to which aspect – hardness, form, use value, or landscape location – is most salient. Of the 15 types of rock, only nari kuru (dolomite), kaba kuru (a geologically distinct sandstone), and jaman kuru (quartz) correspond to rocks recognized by geologists. The term jaman kuru derives from the French diamant (diamond), perhaps due to contact with French geologists who prospected for diamonds in the early 1900s (Varlet et al. 1977). San galima kuru (literally, thunderstone) form where lightning strikes the ground. Archaeologists call these celts or Neolithic polished-stone axe heads (Davies 1967). Clusters of features, not any single characteristic, differentiate types of kuru (cf. Hunn 1976). Many salient features cooccur because these are inherently related, such as how hardness leads to the typical shape of particular rock types. Nonetheless, one feature is considered most salient for each type of kuru, even if the mutual predictability of this and another feature means that the second is as characteristic as the first and explicitly recognized as such. Although these separate classifications of topography, hydrology, vegetation, and ground surface features are important, most consultants consider them altogether when classifying parts of the landscape. This synthetic view produces a separate classification of [land cover]. Land-cover types are either anthropogenic or a type of dan (non-anthropogenic land cover) (Figure 8). Dan is a concept laden with meaning, since jine (spirits) occupy parts of the landscape with nonanthropogenic land cover (Brun 1907; Cashion 1982). Dan is the root of danso (hunter) and dansoko (hunter’s prowess), words that imply mastery of the dan and its spirit occupants. Dan is subdivided between land cover for which vegetation, landscape position, or topography are most salient. Many land-cover terms come from terms for dominant soil or topographic features but are distinguished grammatically



‘Land’ and life 133

[anthropogenic land cover]

Manyang (‘fallow’) [settlement sites]

dan (‘nonanthropogenic land cover’)

kèna (‘clearing’)

[not a clearing]

kuru to (‘land cover of outcrops’) konko to (‘land cover of hills’) mako (‘creekside’)

tumbun (‘ruin’) bugu (‘hamlet’) dugu (‘town’) gasso (‘clearing’) furu (‘field’) hata ge (‘rock flats’) hara (‘seasonal marsh’) huga (‘ferricrete hardpan’)

bogon ge huga (‘silt-covered huga’) bèrèman huga (‘gravelly huga’) tutikuruman huga (‘ferricrete huga’) lemukan (‘arable woodland with sandy to silty loam soil’) bèrè kan to (‘arable woodland with gravelly soil’) nakate (‘arable woodland with high water table and dark soil’) fara fing (‘arable woodland with loamy soil’) kuru fing to(‘arable woodland with rocky, loamy soil’) karankaran kun (‘non-arable scrubland with clayey loam soil’) bogon ge to (‘non-arable woodland with silty, waterlogged soil’) kuru ge to (‘non-arable woodland with rocky, sandy soil’) kakakure (‘non-arable wooded grassland with shallow sandy soil’) kenye he to (‘non-arable woodland with very sandy soil’) busu din kuru he (‘non-arable woodland with extremely cobbly soil’) bèrè kun (‘non-arable woodland with extremely gravelly soil’) huga tirigiyon (‘patchy, non-arable woodland on edge of huga’) huga sinbe (‘non-arable forest at edge of huga’) kuru kunte to (‘flat hilltop with woodland and shallow soil’) kuru karoma la (‘steep slope with boulders and few trees’) kuru sinbe he (‘arable toeslope with few boulders and woodland’) wounjan he (‘steep slope with boulders and forest’) konko karmoa la (‘steep slope with gravel and woodland’) konko sinbe he (‘non-arable toeslope with gravel and woodland’) wawara (‘creekside where water flows over rock flats’) nama mako (‘creekside with rocks covered by slick mosses’) bo jalèn mako (‘creekside where water dammed by dead bamboo’)

Figure 8.╇ Taxonomy of the covert category [land cover]. [Anthropogenic land cover] and dan (non-anthropogenic land cover) are directly subsumed in [land cover].  For diagrammatic conventions, Figure 2.

as locative nouns requiring postpositions in all usages. Land-cover types labeled with such terms are not soil or topographic classes. For example, both kakakure and kuru ge to have kènyè soil, which is arable, but neither land-cover type is arable. Soil in a kakakure has poor moisture characteristics because it shallowly overlies bedrock; kuru ge to (literally, ‘at the place of the kuru ge’) sites are nonarable because the weedy grass ngòlò (Cenchrus ciliaris) dominates such sites. Other land-cover terms, such as lemukan (arable woodland with sandy soil), take postpositions only when used as the object of a clause/verb. Land-cover types for which vegetation is most salient are either kèna (clearings) or [not kèna], a covert category. Both categories have multiple subdivisions relating primarily to ground characteristics. All land-cover types associated with landscape position are types of mako (creekside), differentiated on ground and vegetation characteristics. Finally, land cover for which topography is most salient are associated with kuru (outcrops) and konko (hills). Although many of these cover terms are derived from slope terms, they are not topographic classifications but require postpositions in all uses.

134 Chris S. Duvall

5.

Ethnoecology and environmental variation

The Maninka system of landscape description shows several unusual characteristics when considered in a cross-linguistic context (cf. Burenhult 2008). First, this language domain is lexically rich, with a clearly structured and fairly saturated taxonomy. Importantly, though, my use of tree diagrams overemphasizes the structure of Maninka physical geographic terms, and underrepresents the ambiguity of many constituent terms and underlying concepts (Berlin 1992). Terms for physical geographic features tend to be inherently ambiguous, because they can refer to objects and/or places (Cablitz 2008; Smith & Mark 2001). This ambiguity is overtly marked in Maninka grammar and syntax, which is linguistically unusual. In particular, terms for land-cover types may function as objects or subjects, yet even when used as subjects, some land-cover terms require postpositions if these terms are derived from a label for a different physical geographic feature. This overt marker suggests that, ontologically, my consultants see the study landscape simultaneously as many parts and a gestalt whole (cf. Lyons 1977). Furthermore, since the interrelated concepts of land cover and other terrestrial features are components of the broad term dugu, this term can be proposed as a candidate for the concept ‘landscape,’ which is rarely labeled in studied languages (cf. Burenhult 2008). Further linguistic studies of Mande-speaking groups are necessary to test this possibility. Yet the salience of the synthetic category of [land cover] to my consultants shows that a holistic concept of physical geographic reality is highly important; the concept of ‘landscape’ is inherently holistic (Barrera-Bassols & Zinck 2003b; Goodenough 1966; Johnson 2000; Smith & Mark 2001). Maninka farmers in Solo have a more detailed classification of land-cover types than soil, vegetation, or other features. While soil characteristics (and other specific features) contribute to the perceived arability of a site, a synthetic view of all site characteristics is more salient. While certain soil types are favored for agriculture, many sites with favorable soil are non-arable due to constraints posed by topography, vegetation, or non-soil characteristics of the substrate. Similarly, some sites with less favored soils are arable because the total biophysical environment is suitable for certain crops. Soil characteristics vary within most land-cover types, but not enough for consultants to recognize soil-based subcategories. The only land-cover types for which a single substrate is diagnostic are non-arable, but not all non-arable land-cover types are associated with one substrate. The synthetic view of soil, vegetation, slope, hydrology, and microclimate embodied in Maninka land-cover categories provides a highly salient and useful indication of agroecological potential. Resource managers should use land-cover terms when communicating with Maninka farmers about land management (Laris 2002).



‘Land’ and life 135

The importance of land cover over soil, or any single feature, in classifying site arability is a widespread, underemphasized aspect of local knowledge systems (cf. Denevan & Padoch 1988; Fleck & Harder 2000; Frechione et al. 1989; McGregor 1994; Osunade 1987, 1988; Pulido & Bocco 2003; Shepard et al. 2001; Verlinden & Dayot 2005). Many works characterized as describing local soil types actually describe land-cover categories. For instance, Carney (1991:â•›40) describes how Gambian Mandinka farmers recognize “micro-environments” based on hydrology, topography, and soil. Soil plays a minor role in distinguishing these “micro-environments,” yet reviews of ethnopedology consistently categorize Carney’s paper as describing local soil knowledge (e.g. Barrera-Bassols & Zinck 2000; WinklerPrins 1999). While few linguistic studies have identified terms marking the concept ‘landscape,’ the widespread salience of land cover suggests that analogous concepts may be widely present. Land cover and soil are not interchangeable concepts in local knowledge systems, nor are land-cover categories simply a portion of local soil knowledge. Nonetheless, ethnoscientists frequently confound these concepts, suggesting inaccurately that local people do not differentiate soil from some or all other natural resources in a site. For instance, Barrera-Bassols and Zinck (2000:â•›19; see also 2003a) state, “there is no clear-cut distinction between soil and land characteristics” in local knowledge systems, and that “topography, land use, and drainage” are criteria used to classify soils, without citing specific studies. However, works in their annotated bibliography that apparently support these statements do not pertain to local knowledge of soils per se, but of land cover (e.g. Carney 1991; Kanté & Defoer 1996; Osunade 1988). In some primary works, land-cover terms are inaccurately applied to soils found in a given land-cover type: Tabor (1993:â•›47) translates “fouga” (= huga, ferricrete hardpan) as a specific soil type, but this is not a soil term (Figure 8). Conversely, some authors refer to soil when actually discussing a broader set of environmental features, comparable to Maninka [land cover] categories. For instance, both Osbahr and Allan (2003) and Osunade (1992) repeatedly state that farmers in their study areas examine ‘land’ characteristics – a range of biophysical features, especially vegetation – in determining arability, yet consistently describe this as ‘soil’ knowledge. Universally, soil characteristics are an important aspect of arability, but farmers use knowledge about more than soil in selecting arable sites. The ‘land’ characteristics they perceive could be described as ‘vegetation’ knowledge (e.g. Fleck & Harder 2000; Verlinden & Dayot 2005) as accurately as they are described as ‘soil’ knowledge. Elsewhere Barrera-Bassols and Zinck (2003b) show clearly how local knowledge may be partitioned to distinguish soil and land cover as separate, though related, aspects of the biophysical environment. They report how Purhépecha farmers in Mexico conceive “land” as an integrated whole composed of water, climate, relief, and soils. The Purhépecha

136 Chris S. Duvall

classify land according to how its four components interact at a given site; soil is only one of several variables that determine productive potential. The failure to clearly distinguish ‘soil’ and ‘land’ has caused misrepresentation of the conceptual extent and distinctness of analogous concepts in local knowledge systems. Additionally, past research has misplaced the concept ‘soil’ within folk taxonomies of physical geographic features. For the Maninka, bogo (soil) is one of several components of dugukolo (the ground); it is an intermediate-level taxonomic category (Figure 7). This finding contrasts with Williams and Ortiz-Solorio’s (1981) widely accepted position that ‘soil’ is a ‘kingdom’ in folk-taxonomical terms. Linguistic evidence poorly supports this position. Kingdoms are generally unlabelled (Berlin 1992). Thus, several authors have considered ‘soil’ an exception to this principle since ‘soil’ is a well-defined, labeled category in all studied languages. More parsimoniously, the evidence suggests that ‘soil’ is not a kingdom. Indeed, in most folk soil taxonomies, ‘soil’ is a primary lexeme that subsumes categories labeled mainly by other primary lexemes – such as ‘loam’ – that in turn subsume categories denoted by secondary lexemes – such as ‘sandy loam’. In such cases, ‘soil’ fits only the linguistic criteria for ‘life form,’ not ‘kingdom’ (Berlin 1992). Researchers who have misclassified ‘soil’ may not have studied broad enough samples of pertinent local knowledge systems to observe the concept’s full context. Some sources suggest, as shown here for the Maninka, that ‘soil’ is included in a broader conceptual category that also includes, at the minimum, ‘stone’ (Kanté & Defoer 1996; Romig et al. 1995; Ryder 1994; Sandor & Furbee 1996). For example, the Purhépecha recognize numerous soil types using secondary lexemes derived from the primary lexeme “echeri [soil],” and four types of stone using secondary lexemes based on “tzacapu [stone]” (Barrera-Bassols & Zinck 2003b:â•›239). These authors list both tzacapu and echeri under the heading “soil terms,” suggesting these are taxonomically equivalent categories subsumed in a category equivalent to ‘the ground.’ Knowledge of soils in relation to agricultural practice, an important research topic, must not be divorced from broader knowledge of natural resources farmers use in assessing arability. Soil is but one ground-surface feature farmers assess, and the ground surface is but one of several broad classes of physical feature that compose agroecological potential. Geographers should more widely build upon the work of ethnoecologists to develop Blaut’s ethnogeography, while also striving to ground ethnoecology in geographic theory. In particular, ethnoecologists have not drawn upon geographic theories of place. Yet elements of place contribute to rationales of resource use and management (Brookfield 1969; Nazarea et al. 1998), the core of ethnoecological inquiry (Toledo 1992, 2000). First, understanding culturally specific concepts of locale – the setting in which social relations and human-environment



‘Land’ and life 137

interactions occur (Williams 1973:â•›131) – is crucial for understanding how people perceive landscapes, and not just particular physical geographic features. Too frequently, ethnoecologists have focused on resources, like soil, and not landscapes. For my Maninka consultants, human-environment interactions exist in settings composed of all natural resources and hazards. The synthetic view of [land cover]€– essentially a classification of types of locale – is a more salient factor in landscape management than specific biophysical features. If ethnoecologists mean to understand how people “manage landscapes” as well as “natural resources” (Barrera-Bassols & Toledo 2005:â•›11), locale must become more central in studies of local knowledge. Additionally, sense of place – the local ‘structure of feeling’ arising from traditions and beliefs (Tuan 1977) – should also anchor ethnoecological studies, because it constitutes the site-specificity of the beliefs that guide resource use. For my consultants, the biophysical environment always expresses something about social and spiritual relations; biological, physical, human, and spiritual landscapes are inseparable. Ethnoecologists have recognized such inseparability, but must also recognize that it exists, in part, because of how places become meaningful to people. While ethnoecology could benefit from geographic theories of place, human geographers could also benefit from ethnoecology because studies of place have tended to analyze biophysical variation in more or less superficial manners (Stedman 2003). Geographers must impart theoretically sophisticated understandings of place to studies that maintain ethnoecology’s attentive analysis of physical geographic variation.

References Agrawal, Arun. 2002. Indigenous knowledge and the politics of classification. International Social Science Journal 54(3): 287–297. Barrera-Bassols, Narciso & Toledo, Victor M. 2005. Ethnoecology of the Yucatec Maya: Symbolism, knowledge and management of natural resources. Journal of Latin American Geography 4(1): 9–41. Barrera-Bassols, Narciso & Zinck, J. Alfred. 2000. Ethnopedology in a Worldwide Perspective: An Annotated Bibliography. Enschede: International Institute for Aerospace Survey and Earth Sciences. Barrera-Bassols, Narciso & Zinck, J. Alfred. 2003a. Ethnopedology: A worldwide view on the soil knowledge of local people. Geoderma 111(3–4): 171–195. Barrera-Bassols, Narciso & Zinck, J. Alfred. 2003b. ‘Land moves and behaves’: Indigenous discourse on sustainable land management in Pichátaro, Pátzcuaro Basin, Mexico. Geografiska Annaler 85A(3–4): 229–245. Barth, Hans K. 1986. Mali: Eine geographische Landeskunde. Darmstadt, Germany: Wissenschaftliche Buchgesellschaft.

138 Chris S. Duvall

Berlin, Brent. 1992. Ethnobiological Classification: Principles of Categorization of Plants and Animals in Traditional Societies. Princeton: Princeton University Press. Berlin, Brent, Breedlove, Dennis E. & Raven, Peter. 1968. Covert categories and folk taxonomies. American Anthropologist 70(2): 290–299. Bird, Charles S. (ed.). 1982. The Dialects of Mandekan. Bloomington IN: Indiana University African Studies Program. Blaut, James M. 1979. Some principles of ethnogeography. In Philosophy in geography, S. Gale & G. Olsson (eds), 1–7. Dordrecht: Reidel. Brookfield, Harold. 1969. On the environment as perceived. Progress in Geography 1: 51–80. Brun, Joseph. 1907. Notes sur les croyances et les pratiques religieuses des Malinkés fétichistes. Anthropos 2: 722–729, 942–954. Burenhult, Niclas (ed.). 2008. Language and Landscape: Geographical Ontology in Cross-linguistic Perspective [Language Sciences 30(2/3)]. Amsterdam: Elsevier. Cablitz, Gabriele H. 2008. When “what” is “where”: A linguistic analysis of landscape terms, place names and body parts in Marquesan (Oceanic, French Polynesia). Language Sciences 30(2/3): 200–226. Carney, Judith. 1991. Indigenous soil and water management in Senegambian rice farming systems. Agriculture and Human Values 8(1/2): 37–48. Cashion, Gerald A. 1982. Hunters of the Mandé: A behavioral code and worldview derived from the study of their folklore. PhD dissertation, Indiana University. Dames & Moore, Inc. 1992. Senegal River upper valley master plan study: Final completion report, Vol. 1 – synthesis report. Development project no. 625-0621. Dakar, Senegal: OMVS and USAID. Davies, Oliver. 1967. West Africa before the Europeans: Archaeology and Prehistory. London: Methuen and Co. Denevan, William M. & Padoch, Christine (eds). 1988. Swidden-fallow Agroforestry in the Peruvian Amazon [Advances in Economic Botany 5]. New York NY: The New York Botanical Garden. Duvall, Chris S. 2001. Habitat, conservation and use of Gilletiodendron glandulosum (Fabaceae-Caesalpinoideae) in southwestern Mali. Systematics and Geography of Plants 71(2): 699–737. Duvall, Chris S. 2008. Classifying physical geographic features: The case of Maninka farmers in southwestern Mali. Geografiska Annaler: Series B, Human Geography 90(4): 327–348. Ellen, Roy. 1993. The Cultural Relations of Classification. Cambridge: CUP. Fleck, David W. & Harder, John D. 2000. Matses Indian rainforest habitat classification and mammalian diversity in Amazonian Peru. Journal of Ethnobiology 20(1): 1–36. Forde, Daryll (ed.). 1954. African worlds: Studies in the cosmological and social values of African peoples. London: OUP. Frechione, John, Posey, Darrell A. & Francelino da Silva, Luiz. 1989. The perception of ecological zones and natural resources in the Brazilian Amazon: An ethnoecology of Lake Coari. Advances in Economic Botany 7: 260–282. Goodenough, Ward H. 1966. Notes on Truk’s place names. Micronesica 2(2): 95–129. Holman, Eric W. 2005. Domain-specific and general properties of folk classifications. Journal of Ethnobiology 25(1): 71–91. Horowitz, Michael, Koenig, Dolores, Grimm, Curt & Konaté, Yacouba. 1990. Resettlement at Manantali, Mali: Short-term success, long-term problems. Birmingham AL: Institute of Development Anthropology.



‘Land’ and life 139

Hunn, Eugene. 1976. Toward a perceptual model of folk biological classification. American Ethnologist 3: 508–524. Johnson, Leslie M. 2000. “A place that’s good”, Gitksan landscape perception and ethnoecology. Human Ecology 28(2): 301–325. Kanté, Salif & Defoer, Toon. 1996. How farmers classify and manage their land: implications for research and development activities. In Agricultural R&D at the Crossroads: Merging Systems Research and Social Actor Approaches, A. Budelman (ed.), 115–124. Amsterdam: Royal Tropical Institute. Kay, Paul. 1971. Taxonomy and semantic structure. Language 7: 866–887. Koenig, Dolores, Diarra, Tiéman & Sow, Moussa. 1998. Innovation and Individuality in African Development: Changing Production Strategies in Rural Mali. Ann Arbor MI: University of Michigan Press. Laris, Paul. 2002. Burning the seasonal mosaic: Preventative burning strategies in the wooded savanna of southern Mali. Human Ecology 30(2): 155–186. Lawesson, Jonas E. 1995. Studies of woody flora and vegetation in Senegal. Opera Botanica 125: 1–172. Lyons, John. 1977. Semantics. Cambridge: CUP. McGregor, JoAnn. 1994. Woodland pattern and structure in a peasant farming area of Zimbabwe: Ecological determinants and present and past use. Forest Ecology and Management 63: 97–133. Michel, Pierre. 1973. Les bassins des fleuves Sénégal et Gambie: Étude géomorphologique. Paris: ORSTOM. Nasi, R. & Sabatier, M. 1988. Projet inventaire des ressources ligneuses au Mali. Rapport de synthèse, première phase: Les formations végétales. Bamako, Mali: SCET/AGRI/CTFT/DNEF. Nazarea, Virginia, Rhoades, Robert, Bontoyan, Erla & Flora, Gabriela. 1998. Defining indicators which make sense to local people: intra-cultural variation in perceptions of natural resources. Human Organization 57(2): 159–170. Osbahr, Henny & Allan, Christie. 2003. Indigenous knowledge of soil fertility management in southwest Niger. Geoderma 111(3/4): 457–479. Osunade, M. A. Adewole. 1987. A viable method of land capability classification for small farmers. Journal of Environmental Management 25(1): 81–94. Osunade, M. A. Adewole. 1988. Nomenclature and classification of traditional land use types in south-western Nigeria. Savanna 9(1): 50–63. Osunade, M. A. Adewole. 1992. Identification of crop soils by small farmers of south-western Nigeria. Journal of Environmental Management 35: 193–203. Osunade, M. A. Adewole. 1994. Indigenous climate knowledge and agricultural practice in Southwestern Nigeria. Malaysian Journal of Tropical Geography 25: 21–28. Ovuka, Mira & Lindqvist, Sven. 2000. Rainfall variability in Murang’a District, Kenya: meteorological data and farmers’ perceptions. Geografiska Annaler 82A(1): 107–109. Pulido, Juan S. & Bocco, Gerardo. 2003. The traditional farming system of a Mexican indigenous community: The case of Nueva San Juan Parangaricutiro, Michoacán, Mexico. Geoderma 111(3–4): 249–265. Rappaport, Roy A. 1979. Ecology, Meaning, and Religion. Richmond VA: North Atlantic Books. Richards, Paul. 1993. Natural symbols and natural history: Chimpanzees, elephants and experiments in Mende thought. In Environmentalism: The View from Anthropology, Kay Milton (ed.), 144–159. London: Routledge.

140 Chris S. Duvall

Romig, Douglas E., Garlynd, M. Jason, Harris, Robin F. & McSweeney, Kevin. 1995. How farmers assess soil health and quality. Journal of Soil and Water Conservation 50(3): 229–236. Ryder, Roy. 1994. Farmer perception of soils in the mountains of the Dominican Republic. Mountain Research and Development 14(3): 261–266. Samaké, Maximin, Sow, Moussa, Sarr, Mamadou, Maiga, Fatoumata & Camara, Baba. 1986. Etude de l’économie domestique dans la zone du Barrage de Manantali. Rapport de mission de réperage. Phase II. Bamako, Mali: Institut des Sciences Humaines. Samaké, Maximin, Sow, Moussa, Sarr, Mamadou, Maiga, Fatoumata & Camara, Baba. 1987. Etude de l’économie domestique dans la zone du Barrage de Manantali. Phase III. Bamako, Mali: Institut des Sciences Humaines. Sandor, Jonathan A. & Furbee, Louanna. 1996. Indigenous knowledge and classification of soils in the Andes of southern Peru. Journal of the Soil Science Society of America 60(5): 1502–1512. Shepard, Glenn H., Jr., Yu, Douglas W., Lizarralde, Manuel & Italiano, Mateo. 2001. Rainforest habitat classification among the Matsigenka of the Peruvian Amazon. Journal of Ethnobiology 21(1): 1–38. Shipton, Parker. 1994. Land and culture in tropical Africa: Soils, symbols, and the metaphysics of the mundane. Annual Review of Anthropology 23: 347–377. Smith, Barry & Mark, David M. 2001. Geographic categories: An ontological investigation. International Journal of Geographic Information Science 15(7): 591–612. Sow, Moussa & Anderson, Jon. 1996. Perceptions and use of woodland by Malinké villagers near Bamako, Mali. Unasylva 47: 22–27. Stedman, Richard C. 2003. Is it really just a social construction? The contribution of the physical environment to sense of place. Society and Natural Resources 16: 671–685. Tabor, Joe. 1993. Soils of the lower, middle, and upper Senegal River valley. In Risk and Tenure in Arid Lands: The Political Ecology of Development in the Senegal River Basin, Thomas K. Park (ed.), 31–50. Tucson AZ: University of Arizona Press. Tauxier, Louis. 1927. La réligion bambara. Paris: Librairie Orientaliste Paul Geuthner. Toledo, Victor M. 1992. What is ethnoecology? Origins, scope and implications of a rising discipline. Etnoecológica 1(1): 5–21. Toledo, Victor M. 2000. Indigenous knowledge on soils: An ethnoecological conceptualization. In Ethnopedology in a Worldwide Perspective: An Annotated Bibliography, Narciso BarreraBassols & Alfred Zinck (eds), 1–9. Enschede: International Institute for Aerospace Survey and Earth Sciences. Tuan, Yi-Fu. 1977. Space and Place: The Perspective of Experience. Minneapolis MN: University of Minnesota. Varlet, M., Bouchind’homme, J. F., Sissoko, I. & Kissao, I. 1977. Rapport de grande réconnaissance, Mali-Sud: Régions de Kayes-Bamako. Marseilles, France: Compagnie Générale des Matières Nucléaires. Verlinden, Alex & Dayot, Bertrand. 2005. A comparison between indigenous environmental knowledge and a conventional vegetation analysis in north central Namibia. Journal of Arid Environments 62: 143–175. White, F. C. 1983. The Vegetation of Africa: Maps and Memoir. Paris: UNESCO/AETFAT/ UNSO. Williams, Barbara J. & Ortiz-Solorio, Carlos A. 1981. Middle American folk soil taxonomy. Annals of the Association of American Geographers 71(3): 335–358.



‘Land’ and life 141

Williams, Raymond. 1973. The country and the city. Oxford: OUP. WinklerPrins, Antoinette M. G. A. 1999. Local soil knowledge: A tool for sustainable land management. Society & Natural Resources 12: 151–161. WinklerPrins, Antoinette M. G. A. & Sandor, Jonathan A. (eds). 2003. Local Soil Knowledge: Insights, Applications, and Challenges. Special issue of Geoderma 111(3/4).

chapter 7

Landscape in Western Pantar, a Papuan outlier of southern Indonesia* Gary Holton

This chapter describes the landscape, streamscape, and seascape terminology of Western Pantar, a non-Austronesian (“Papuan”) language spoken in the Alor archipelago of eastern Indonesia. In Western Pantar reference to elevations is achieved through named places of habitation rather than through generic landform terms; water bodies are denoted according to their quality rather than their form; and seascape terms reflect a focus on intertidal foraging and minimal use of open sea resources.

1.

Introduction

Recent research into the cross-linguistic categorization of landscape reveals striking variation across languages (this volume; Burenhult & Levinson 2008). Landscape presents a continuous surface, and languages discretize that surface in different ways. Crucially, this process of landscape categorization is driven by the human experience of and relationship to the landscape, what Levinson (2008) has called human affordances. Understanding these cross-linguistic differences in categorization is not a simple matter of developing a mapping between landscape ontologies, for the underlying parameters on which the categorization is based may also differ both across and within languages. In the Western Pantar (WP) language, to be discussed in this chapter, some elevation terms focus on size, while others emphasize shape. Even more strikingly, WP terms describing water features forgo any reference to morphology and instead focus on the quality or nature of the liquid itself.

* The data presented in this chapter derive from the author’s field work in Pantar 2004– 2008, supported by US National Science Foundation grants SGER-0408448, BCS-0553754 and BCSâ•‚0756159, and ELDP field trip grant FTG0148.

144 Gary Holton

As I will argue below, landscape classification in WP is driven largely by cultural factors, namely the human experience of the landscape and the cultural construction of that experience. In some cases the evidence for this claim is quite pronounced. A quick glance at the inventory of WP toponyms reveals relatively few names for putatively universal geographic features such as mountain. Instead, place-naming strategy focuses on villages, gardens, rocks, trees, totemic sites, and other locations which are geographically unremarkable but culturally salient. Further, the WP classification of water quality clearly reflects a unique perspective on the utility of water as opposed to its geomorphic force. In other cases the role of cultural factors is more subtle. Although the WP category of mountain has a similar denotation to its English counterpart, the WP term differs substantially in usage. A seemingly simple question like “What is the name of that mountain?” evokes ambiguous responses in WP because the question mislocates the primary landscape feature as the mountain itself as opposed to a village or garden site which might be located on the slopes of the mountain. Thus, WP provides evidence that even when different landscape ontologies appear to correspond in their literal senses, their connotative senses in everyday language use may be quite different. The study of landscape categorization requires careful, detailed ethno-linguistic documentation. Before proceeding with the description of landscape categorization in WP I provide below some background on the language and its environment.

1.1

Prolegomenon

In describing the landscape of a far-off, little-known place there is always the danger of introducing too much exoticism. Indeed, to a certain extent any foreign place is by definition exotic. But the landscape of Pantar Island is truly out of the ordinary, an outlier in the extremes human habitation. This is not just true when viewed from outside: Pantar is seen as extreme and exotic even within the local regency and province. Civil servants express reluctance to take assignments on Pantar, and even ordained clergy assigned to Pantar routinely abandon their posts. More than half of the indigenous population has migrated elsewhere. Residence is often intermittent, with families maintaining households off-island and returning only at key points in the agricultural cycle.

. A regency (Indonesian kabupaten) is an administrative subdivision of a province, similar to a county. The Indonesian province of East Nusa Tenggara is divided into nineteen regencies and one municipal district.



Landscape in Western Pantar 145

The reasons for this are varied but are ultimately related to the presence of a low volcano which dominates the landscape, giving birth to sulfurous, poisonous creeks. Pantar is an extremely hot and humid place with little rainfall outside a brief torrential monsoon season. As a result there is no surface water other than these sulfur creeks, and subsurface water is similarly contaminated by volcanic brines. There is almost no cash economy and little access to mass communication or transportation. (Pantar is the largest island in Indonesia to lack an airstrip of any kind.) And yet, this situation arises not due to the external influences of war or natural disaster, but due to the inherent qualities of the landscape itself. From the perspective of a WP speaker this landscape is not at all exotic; it is simply home.

1.2

Language

Western Pantar (ISO 639-3 lev) is a Papuan language spoken by about 10,000 people on the western half of the island of Pantar, located at about 8 degrees south latitude and 124 degrees east longitude in the eastern Indonesian province of Nusa Tenggara Timur. WP is one of a small group of about 15–20 closely related languages which together form the Alor-Pantar (AP) group, spoken on Pantar and neighboring Alor, as well as on several smaller islands located in the straight between those two islands (see map Figure 1). Together with the three or four Papuan languages of Timor Island, the AP languages are generally assumed to belong to the Trans New Guinea family (Ross 2005); however, the evidence for this is scanty, based on a preliminary examination of pronominal forms. The AP languages may also be related to at least some of the Papuan languages of

Wersing Alorese

Adang

Nedebang

Pantar Island

Teiwa Western Pantar

Kabola Hamap

Blagar Blagar Retta Kaloa Koera Klon

Kamang

Abui

Kula Sawila

Wersing Kui

Alor Island

Tereweng

Figure 1.╇ The Alor-Pantar languages, Western Pantar on the far left (based on Grimes et al. 1997). Inset: Location of the Alor-Pantar language area within Indonesia.

146 Gary Holton

Timor, forming a Timor-Alor-Pantar group (Klamer et al. 2009). What is clear is that the (T)AP languages show no genetic relatives elsewhere within eastern Indonesia outside New Guinea, located some 1000 km to the northwest. This makes the AP group the most distant Papuan outlier group, entirely surrounded by Austronesian languages. At least five other Papuan languages are spoken on Pantar: Teiwa, Sar, Kaera, Blagar, and Nedebang. In addition, a single Austronesian language, Alorese (ISO 639â•‚3 aol; known locally as Bahasa Bara ‘Baranusa language’), is spoken indigenously on the island, primarily in the town of Baranusa, at the head of the large bay on the north side of the island, as well as on the coast to the west of Baranusa; on Kangge Island, off the west coast; and in the Tanjung Muna area on the northwest coast. Many speakers of WP have at least passive fluency in Alorese, and fluency rates increase with proximity to the Alorese-speaking regions. In addition, all but the youngest WP speakers speak either the local variety of Malay, standard Indonesian, or both.

1.3

Culture and economy

The economy of the island is based around subsistence swidden agriculture. The primary crops are dry-land rice, maize, cassava and sweet potato, with lesser contributions from millet. Fields are allowed to go fallow for three to five years. Then brush is cut and burned prior to planting at the beginning of the rainy season. Coconut palms, a source of cash income in other outlying regions of the Indonesian archipelago, are limited, owing both to a lack of water and to a recent blight, but the lontar palm (palmyra, Borassus flabellifer) is cultivated assiduously, primarily as a source of a fermented beverage known as tua, often consumed in lieu of water. Hunting of wild pigs and deer remains an important cultural activity but does not provide a significant source of food. On the other hand, animal husbandry, especially the raising of pigs, is a foundation of the local economy and tradition. Pigs are crucial to the annual rice harvest. Rice fields are owned by individual families or descent groups, but agricultural activities of clearing, planting, and harvesting are undertaken by larger work parties. The largest of these may involve as many as fifty adults from more than one clan. Work parties must be fed with a pig; a large work party may require two pigs. As a result, farmers adjust the size of their plantings according to the availability of pigs. Pigs are also necessary for certain ceremonies, including marriages, funerals and house-raisings. The meat of pigs and other animals is only rarely consumed outside of agricultural work parties and ceremonial occasions.



Landscape in Western Pantar 147

Cash crops play a limited role in the local economy. The most successful of these is the cashew tree (Anacardium occidentale), introduced in the mid 1990’s, but low yields, labor-intensive cultivation, and blight discourage development of a local market. Most plantations are limited to a handful of trees producing a cash income of less than US$50 annually. Recent efforts by the local government and outside NGOs have encouraged development of seaweed farming, but the collapse of the international market in 2008 has limited the viability of aquaculture in the region. Owing to the lack of local employment opportunities, many Pantar residents seek short-term employment in the regency capital of Kalabahi on the neighboring island of Alor. Outside of the three main agricultural seasons (field clearing/ burning, planting, and harvesting) as much as half of the adult population of west Pantar may be resident in Kalabahi. Residents also travel to Kalabahi for education. While a secondary school opened in west Pantar in 2006, many residents continue to send the children to Kalabahi for secondary school. Peripatetic travel to Kalabahi also provides a major source of news and information. Most west Pantar villages do not have electricity and hence have limited access to mass media. Satellite receivers have been installed in some villages, facilitating access to national television broadcasts, but local news is more difficult to obtain. A cellular phone tower was erected in 2007 in Baranusa, within range of some west Pantar villages, but as of July 2008 it was not operational, having suffered storm damage during construction. A shortwave radio at the middle school can sometimes be used to send messages to other islands. The most reliable means of communication with neighboring Alor remains hand-written notes or verbal messages carried by passengers traveling to Kalabahi.

1.4

Ecosystem, geography and climate

Pantar falls within the tropical dryland or savanna climatic zone, which is more similar to that of northern Australia than to the rest of Indonesia. Average annual rainfall is approximately 1000€mm, falling almost entirely within the months of January and February. This places Pantar among the most xeric regions in Indonesia. Where undisturbed forest still remains, the primary species is indigenous eucalyptus (Eucalyptus urophylla). Pantar can be divided geographically into two distinct geographic and climatic zones. The eastern portion consists of a peninsula approximately 30 km long and 10 km wide extending to the northwest, dominated by a steeply sloping ridge of mountains 500–700 m in elevation. This region is separated from the western region by lava flows which have created steep canyons and presented a barrier to

148 Gary Holton

LEMBATA ISLAND

PANTAR ISLAND

Western Pantar language area

20 km

Figure 2.╇ Google Earth image showing Pantar Island and the Western Pantar language area, looking north. The crater of Sirung Volcano is visible toward the southern end of the island (8° 29' 50" S, 124° 7' 48" E).

road construction. (To this day, the principal towns in the eastern and western regions are better connected to the regency capital of Kalabahi, on neighboring Alor Island, than they are to each other.) The western region of Pantar, which is essentially coterminous with the WP language area (see Figure 2), is characterized by a low plain of 100–200 m elevation, extending between an active stratovolcanic complex in the east and a smaller range of hills supporting active volcanic vents to the west. The entire region forms what was once a massive volcanic crater of some 15 km diameter, the southwestern portion of which is now submerged. While the presence of an active volcano is not unusual along the Sunda arc, the hydrologic system associated with Mt. Sirung is unique to the west Pantar region. Sirung is the source of a supersaline, acidic brine which surfaces at several locations across the western portion of the island (Poorter et al. 1989). Most notable are several hot springs, two of which are sources for surface creeks which run a course to the ocean. This brine results in the contamination of the water table, with the result that it is difficult or

. In fact, the Sirung brine is highly unusual worldwide, being one of the few natural acidic brines to occur in a subduction zone.



Landscape in Western Pantar 149

impossible to locate potable water in the region. Outside the brief rainy season, all wells are contaminated either by the volcanic brine or by seawater. As a result west Pantar presents an extremely dry and parched landscape.

1.5

Structure of the chapter

The rest of this chapter is organized as follows. Section 2 discusses toponyms and place-naming strategies, noting the preponderance of exophoric names and the paucity of geographically prominent named features. Section 3 describes elevation terms and shows that even where generic elevation terms exist, they lack the salience associated with less geographically prominent features such as gardens and villages. Section 4 briefly describes the classification of vegetated areas, noting the close correspondence between the terminology in this domain and the degree of human modification of the land. Section 5 describes stream and water terms. Here I show that WP lacks a generic river term or indeed any stream term which emphasizes the morphology or dynamics of water flow. Instead, WP water terms emphasize quality, distinguishing varying degrees of water acidity irrespective of morphology. Section 6 describes seascape terminology. The relative paucity of such terminology is hypothesized to reflect the low cultural saliency of the ocean and its resources. Finally, Section 7 concludes with an areal perspective, demonstrating that at least some typologically unusual features of WP landscape categorization can be found in neighboring Alor-Pantar languages as well.

2.

Toponyms

In considering the conceptualization of landscape in WP, the role of toponyms cannot be overlooked. Pantar is a storied landscape; the land is alive with names. Garden plots, trees, rocks, water sources, houses – even pools in the reef have names. The names link these places to history, real or imaged, evidencing a deep relationship between WP speakers and the land they inhabit. Names are often obscure to outsiders unfamiliar with local history but immediately relevant to WP speakers. For example, the name Ta Haila denotes a prominent red sand beach located toward the end of Puntaru Lagoon. The lagoon is remarkable in that it is fringed by beaches of many different colors of sand, reflecting different types of mineralization. A mere 200 m or so in length, the deep red rocks and coarse sand of Ta Haila stand out on the horizon. Ficus trees provide morning shade. But it is not the prominent red color or the inviting shade which are most notable to a WP

150 Gary Holton

speaker. The name Ta Haila means literally ‘source of building blocks.’ This name refers directly to an important legend in which a traditional chief ’s house, or kabi, was erected on this location. A large arrangement of red rocks just above the waterline is said to be the remains of this house. The beach now known as Ta Haila is thus the beach from which the building rocks for this house were gathered. The degree of transparency of such toponyms is of course debatable. As a name, Ta Haila may serve simply as a waypoint, a tool for conveying information about the landscape. For example, a speaker might instruct someone to feed the male pig which is tied up just inland from Ta Haila. Thus instructed, a speaker may walk to Ta Haila without consciously recalling the story of the building of the chief ’s house there. Yet, upon reflection the association with history is inescapable; every WP speaker knows the story. Even where WP place names do not reach deep into history, they still attest to the connection between WP people and the land. For example, the name of a prominent hill in the northern part of WP territory is Is Habbang, literally ‘banyan village.’ The literal meaning of this name derives from the metonymic relationship between the hill and the village which was formerly located at its base. The reference could in theory be disambiguated in context, but there is often no need to do so. The village is the hill, and the hill is the village. What is significant about the hill is that there is a village located nearby. In this sense the name Is Habbang also reveals history – a modern history of a village which once existed but is still very much remembered. As we shall see in the following section, this fine attention to history in the naming of places turns out to have great bearing on the landform terminology more generally. Let us first consider the structure of WP toponyms. Like other proper names in WP, toponyms are almost exclusively binomial, usually consisting of two nongeneric parts. In a preliminary survey of 202 toponyms in the vicinity of Puntaru, only six names can be said to be monomorphemic. These are: Buggu, Bakka and Mallung, three gardens near Puntaru; Wassir, a large valley to the north of Puntaru; Gurung, a small bay at the mouth of a stream; and Malua, the old name for Alor Island. Another 74 of these toponyms consist of a monomorphemic or binomial term together with a generic: habbang ‘village’ (24 names); gamma ‘point’ (6 names); hoang ‘beach’ (5 names); wee ‘fishpond’ (4 names); bi’ang ‘plain’ (2€names); halia ‘water’ (14 names); bila ‘ridge, cliff ’ (7 names); and kukka ‘mountain, island’ (12 names, though see discussion below). The remaining terms are binomial without a generic. Examples are given in Table 1. It should be emphasized that even those toponyms which do contain generics are often exophoric. For example, Hauwe Kassing, literally ‘split rock,’ refers to a beach, not a rock. The name itself evokes a reference to a legend said to have



Landscape in Western Pantar 151

Table 1.╇ WP binomial toponyms lacking a generic component Toponym

Literal translation

Designation

Bo Lolang Di Muggung Hauta Haulang Kuang Ra War Si Domma To Yasa Ta Haila

waves follow throw down thunder hanging coconuts drum fire shines ornament thick lontar palm fruit bad brick source

village name rocky point at south end of Puntaru lagoon garden site cultivated field river valley water source Beach

taken place on that beach. In the legend the protagonist splits a drum in half with a sword, and the drum then turns to stone. The split rock still found on Hauwe Kassing is said to be the petrified stone of the legend. In the same vein, Lallang Habbang, literally ‘Lallang’s village,’ refers not to a village but to a water well. An additional level of binomial structure exists in which two toponyms€– binomial or not – are recited together in a parallel structure. Parallelism is a pervasive areal feature of the languages of southern Indonesia: a pair of words or phrases having similar phonological, metrical, grammatical and semantic structure is pronounced together under a single intonation contour, often with exophoric reference (Fox 1988; Grimes et al. 1997). While often described as a crucial feature of ritual language, parallelism in WP is common in everyday language, especially the recitation of toponyms. Many WP toponyms are known formally as parallelisms consisting of two binomial toponyms. For example, Siarang Dai – Aname Dai, literally ‘multitude divide, people divide,’ referring to the place from which according to legend the various dialects of the Western Pantar language diverged, following settlement of Pantar Island. Often such parallelisms will use an alliterative structure, as in the repetition of the word dai in the preceding example, but other structures are possible as well. The name Mau Kaling – Mua Babar, literally ‘Mau Kaling sweeps the earth,’ makes poetic use of the near homonymy between Mau, a personal name, and mua ‘earth.’ This particular name denotes a volcanic erratic boulder of approximately one meter diameter which is located near the bank of a sulfurous creek. Toponyms in WP may also serve to structure discourse. Names are often memorized in a sequence reflecting a course taken by an historical or mythological figure, as in the following excerpt from a story which tells the origin of a particular dance form known as Soli-Meli. This brief excerpt traces a string of ten names from Towang Kalla to Manaung – Beda Gauwang. Note also that all but one of these names are parallelisms consisting of two names for a single place.

152 Gary Holton

(1) Excerpt from the Soli-Meli story by Gerson Lau Muri a. maum bunni me yang dekang Towang Kalla me level garden at return.from.above descend NAME to ‘[from] the garden over there [they] descended to Towang Kalla’ b. ma dekang Tai€Kassi – Bi€Dewang yang pi come descend NAME return.from.above descend Bukke – Reng€Ara NAME ‘came down to Tai Kassi – Bi Dewang, back down to Bukke – Reng Ara’ c. si Ta Baiyang – Mulli Gawang go NAME ‘went to Ta Baiyang – Mulli Gawang’ d. middang raung Ko Bunni – Bila Wang return.from.below ascend NAME ‘climbed back to Ko Bunni – Bila Wang’ e. pi Lang Suki – Serang Banang descend NAME ‘went to Lang Suki – Serang Banang’ f. tang golang raung ta Dowang Mali – Sela Bunni and.then ascend already NAME ‘then up to Dowang Mali – Sela Bunni’ g. Tobu Mali Wenang – Wenang Gakka Lesi NAME ‘[continue to] Tobu Mali Wenang – Wenang Gakka Lesi’ h. ma Ki Gaiti – Paling Gaiti, Manaung – Beda Gauwang come NAME, NAME ‘came to Ki Gaiti – Paling Gaiti and Manaung – Beda Gauwang’

In this stretch of discourse the toponyms serve as waypoints along a mental journey which is structured with liberal use of spatial reference terms. The narrative begins with the spatial deictic maum ‘at the same level’ and then proceeds to use nine spatial motion verbs, including cislocative yang ‘return from above’ and middang ‘return from below,’ as well as translocative pi ‘descend nearby,’ dekang ‘descend’ and raung ‘ascend.’ The only other verbs in this excerpt are the generic motion verbs ma ‘come’ and si ‘go.’ Binding place names together with spatial terms, such toponymic discourse embeds a human connection into the landscape. The landscape is not merely an abstract collection of features – mountain, hill, river, etc. – but rather a storied place through which humans have passed since the beginning of time. As with the route descriptions recorded by Wassmann (1997) for the Papuan language Yupno, toponymic discourses lack



Landscape in Western Pantar 153

the two-dimensional precision of a cognitive map, yet they function nonetheless to provide a human structure to the landscape.

3.

Elevations

The most common way of referring to raised elevations and convex landforms in WP is with the locational noun bila. This term is sometimes translated as ‘ridge’ or even ‘cliff ’; however, it is more often found with the third singular possessive prefix as gabila, meaning ‘the area uphill of it.’ In this usage, gabila contrasts with galawang ‘the area downhill of it.’ This is illustrated graphically in Figure 3 with the noun yattu ‘tree.’ The term bila thus refers more generally to any upland or upslope area, not necessarily a convex form. This term is common in toponyms, both those referring to large peaks, such as Was€Bila ‘sun ridge’ (907€m) and Pu€Bila ‘eucalyptus ridge’ (676€m), which dominate the western part of Pantar Island, as well as low lying areas such as Boling Bila ‘least valued moko drum ridge’ (25€m). Many secondary elevation terms in WP are based on metaphorical extensions. For example, mugang may refer to a ridge which protrudes out into a valley, but it also denotes the bump on the head of a chicken or other bird. Some Pantar elevation terms have primary senses which refer not to elevation but to the shape of the land. Consider the term harang ‘upland,’ which contrasts with bi’ang ‘flat, plain.’ The distinction between these terms is one of aspect rather than elevation. The term bi’ang may refer to flat areas regardless of elevation. Thus, the toponym Diddi Bi’ang refers to the large outwash plain behind Diddi beach. This plain is actually located within a large valley only barely above sea level and hence lower in elevation than the surrounding ridges. But bi’ang may also refer to raised plains such as Dalawang Bi’ang, a large flat area

Yattu gabila

Yattu galawang

Figure 3.╇ Locational nouns.

154 Gary Holton

on a ridge of approximately 100 m elevation located 5 km inland from Diddi. In contrast, harang refers to steeply sloping areas such as the slopes of mountains. Unlike bi’ang, the term harang does not occur in toponyms. However, it does occur in the designation harang wala ‘upland people.’ In this usage it may contrast with bi’ang wala ‘plains people (farmers),’ or – more often – with bara wala ‘people of Baranusa, coastal people.’ In this latter sense harang wala has the connotation of ‘Christian’ as opposed to ‘Muslim,’ inferred from the geographic distribution of those two groups on the island. In any case, harang is perhaps better glossed as ‘sloping.’ The WP elevation term which comes closest to the sense of English ‘mountain’ is the word kukka. This term refers both to large mountain massifs as well as to islands. (Islands may also be referred to as ir buka, a compound formed from ir ‘state, condition’ and buka ‘trunk.’) The term kukka occurs most commonly as an optional component in toponyms, particularly in island names such as Uddu Kukka ‘Tereweng Island’ (literally ‘ashes island’) or Gale Awa Kukka ‘Pantar Island’ (literally ‘living form island,’ and the likely source of the colonial era term Galiyao for Pantar Island). It may be that ‘island’ is the primary sense of this term, as it is found only in the more recent of those toponyms referring to mountains. For example, the major volcano on Pantar is commonly referred to today as Sirung Kukka, an apparent calque of the Malay name Gunung Sirung (English: Mt. Sirung). However, the older name for this feature is Saré Buri Ara, literally ‘large torch (?),’ without the generic kukka. In any case only a handful of large peaks on Pantar may be referred to as kukka. Most smaller convex landforms are identified not by a generic term at all but rather by a toponym. Usually this is the same toponym used to denote a co-located current or former village site. Thus, as we saw in Section 2 above, Is Habbang may refer either to an historic village or to the small flat hill (elevation 340 m) near which that village was located, or to both. The generic kukka may be used to distinguish the landform itself from the village, as in Is Habbang Kukka, but such usage has the feeling of a neologism, a deliberate invention for a perceived need to disambiguate between village and landform. The generic kukka also occurs in the compound kukka haila meaning ‘base, foot of mountain.’ The term haila ‘main part’ is not strictly speaking a landscape term but rather forms a variety of compounds, not all of them referring to landscape. Compare: yattu haila ‘area underneath a tree’; ber haila ‘main point, topic’ (ber ‘words’); tawasing haila ‘incisor’ (tawasing ‘tooth’); tawar haila ‘molar’ (tawar ‘chin’); tauwe haila ‘base of ear’ (tauwe ‘ear’).

. The identification of Galiyao as referring to Pantar is established by Barnes (1982) and Dietrich (1984). Additional folk etymologies can be found in those sources.



Landscape in Western Pantar 155

While the generic elevation term kukka is sometimes found in toponyms, this usage is innovative and is largely restricted to citation forms. And even in citation forms the use of the generic kukka seems to serve to create binomial names from monomial ones. For example, the name Bo Yali is already a binominal form without the generic kukka, and hence the addition of kukka to form Bo Yali Kukka (?) seems strange. In contrast, the name De Kukka would fail to be binomial without the generic kukka included and hence seems more acceptable, at least in citation form. Nevertheless, in text kukka is used only as a generic. For example, (2)

kukka si tar gaddi ba raung moutain this what do conj climb a-baulang a-hinna-b hang ha-lama asp-fall asp-die-conj 2sg 2sg-go ‘you will fall to your death if you climb up that mountain’

In contrast, proper names of mountains occur in text without kukka. (3) ging Was Bila tang a-raung 3pl name on asp-climb ‘they will climb (onto) Was Bila (Mt. Basangloi, ‘sun cliff ’)’

More typically elevations are referred to metonymically by the name of a nearby location associated with the elevated location. This location may have a much greater salience than any of the associated elevations. To say that WP has a generic elevation term kukka meaning roughly ‘mountain, island’ is not the same as saying that this term is used in the same context as the English term ‘mountain.’ This point can be illustrated with a brief anecdote from my field work in Pantar. Over the many months I have spent in Puntaru I have spent many evenings sitting on the beach as the evening sun sets, contemplating the view shown in Figure 4. Often I would inquire casually as to the names of what I took to be two prominent peaks. Rising to a height of 1344 m and 938 m, these are by far the highest elevations on Pantar Island, and at a distance of a mere 8 km from Puntaru, these peaks dominate the landscape. Yet, there is much ambiguity and disagreement as to the precise name for these peaks. Most speakers use the binominal parallel term Tua Pakki – Bodda Bali, a term which follows a typical WP naming practice by combining names for two localities in a compound in order to denote a region surrounding those localities. It seemed only natural, to me, to assume that Tua Pakki referred to one peak and Bodda Bali to the other. But none of my WP speaking consultants accepted this simplification. Most felt that the name Tua Pakki – Bodda Bali refers not to either peak but to the entire region, including both peaks and the wide saddle between them. Others felt that the name referred

156 Gary Holton

Tua Pakki - Bodda Bali

De (Kukka) Bo Yali Tua Pakki

Bodda Bali

Figure 4.╇ Image looking south from Puntaru village to the mountain peaks on southern Pantar.

only to the saddle region. Some speakers used Tua Pakki to refer to only the taller of the two peaks, but this was acknowledged to be a clipping, a common practice with parallelisms. It was some months before I learned that the names Tua Pakki and Bodda Bali referred not to elevations but rather to localities (former settlements, now just garden houses) located in the wide saddle between the two peaks. This was not an example of hidden or obscure knowledge; everyone acknowledged it to be true. It was just considered too obvious to be immediately notable. In contrast, there was much less agreement regarding the names for the two peaks. After much consultation I settled on De and Bo Yali as consensus names, though there remain speakers who disagree. The fact that I arrived at consensus names for the two peaks only after much discussion and debate is not evidence of lack of local knowledge of the landscape, but rather of my own efforts to impose an etic distinction between the mountain and the village – a distinction which is not very salient in WP. In other words, where I saw a landscape, WP speakers saw a living, embodied place. The WP approach to convex landforms contrasts sharply to the situation with other landforms. For example, the roughly 400 hectare area directly south of Puntaru lying between the two sulfurous creeks of Masi Salamang and Masi Ke Baddang contains over twenty named gardens, the names and locations of which are undisputed by all speakers, in spite of their lower geographic salience as compared to the high peaks to the south. For WP speakers it is cultural salience which matters, not geographic salience. Gardens and their location are important to the day-to-day activities of WP speakers; mountain peaks are not. With this in mind I take a brief excursion to discuss the WP terms for vegetated areas.



4.

Landscape in Western Pantar 157

Vegetated areas

In spite of the small population size, the landscape of Pantar is intensely utilized by humans. The thin volcanic soils are quickly drained of nutrients, with the result that the shifting cultivation systems require that fields lay fallow for several years between planting. Hence, the total amount of agricultural land may be as much as five times as large as the amount of land under cultivation in any one year. Virgin forests are found only in the highest reaches and steepest slopes of mountains, and even these areas may be used for (selective) timber harvesting. Vegetated areas can be referred to as either alá ‘grass, grassland, overgrown area’ or wappang ‘forest’. The contrast between these two terms can be seen in their relative temporality. The vegetation in alá grows thick during the rainy season, but can be cut back and eventually dies in the dry season, at which time it is easily burned. In the areas known as wappang the vegetation is more permanent. It must be cut in order to be cleared, not just burned. Still, there is quite a bit of semantic overlap between the two terms. Thus, alá me, lit. ‘at the grass’, may also refer to forest, perhaps by analogy with the term alá gaume ‘interior,’ lit. ‘within the grass’. Conversely, wappang sassa, lit. ‘smashed forest,’ refers to a brushy undergrowth or thicket. Crucially, wappang sassa refers only to brushy areas which have not (recently) been used for gardens, whereas alá me may refer to recent growth in an area which has been recently cultivated. A more wild forest may be referred to as mutta ‘jungle’ (Malay ‘rimba’), a term which implies virgin or original forest. On the other hand, a cultivated field is referred to as bi’ang (also meaning ‘plain’, see above) or more completely bi’ang atera, lit. ‘arranged field’. This term is distinct from bunni ‘garden,’ which implies active cultivation. That is, bunni focuses on human interaction, whereas bi’ang focuses on physiography. The generic terms alá and wappang exclude certain floras which are considered as distinct from these types. While many different species of grasses may be denoted by the generic alá ‘grass’, the plants known as sabia, bittung, bimmali, and kumbabba are all types of mali ‘bamboo’ but not types of alá. Similarly, hasi ‘alang-alang grass’ (Imperata cylindrica), which is used as a roofing material, is not alá. Thus, the term alá is at least in part defined by the way in which humans interact with it. It is thus possible to classify WP vegetation terms according to the degree of human modification. The terms in Table 2 are organized roughly from virgin forest to active crop.

158 Gary Holton

← increasing modification

Table 2.╇ Stages in garden preparation wappang maggar waro lotta tupur bi’ang bi’ang atera bunni

‘forest which has not been used for a garden’ ‘to clear a field for planting’ (verb) ‘wood refuse which remains after clearing garden’ ‘grass refuse which remains after clearing garden’ ‘stubble in fields’ ‘field’ ‘arranged, prepared field’ ‘garden’

This scale reveals that WP terms denoting vegetated areas are conceived in terms of human affordances. WP economy is based upon subsistence agriculture; the most relevant feature of a vegetated area is the way in which it has been modified for agricultural use.

5.

Streams and water

WP presents an unusual system for classifying streamscapes and water. In many languages streamscapes are characterized in terms of their morphology, rather than their contents. That is, rivers may be classified according to size or rate of flow, but the contents are generally irrelevant to this classification. Some languages do distinguish dry or intermittent watercourses with distinct lexemes, but in these cases it is the presence or absence of water which is crucial, not the quality of the water. WP takes exactly the opposite approach to the classification of water and water features, choosing to focus on the quality of the substance itself rather than its morphology. In WP potable water from any source other than the ocean is denoted by the term halia. The term halia may be extended to cover a wide range of meanings, including ‘water’, ‘spring’, ‘water well’, ‘lake,’ and ‘lagoon’. The water in question may be either fresh or brackish, hot or cold. A well may be optionally distinguished by using the modifier gaiti ‘eye’; thus halia gaiti ‘water source’. Similarly, a hot water source or hot-spring may be distinguished with the modifier sosoli ‘hot’, while brackish water may be distinguished with the modifier magge ‘brackish’ or makara ‘salty’. There are no terms which distinguish such water features based on shape or flow. Thus, the lagoon at the mouth of the Wassir Valley is known as Halia Bakurang and the hot spring which bubbles out from the beach at low tide south of Puntaru is known as Halia Kabbarung, literally ‘water-jar water’. Both are known generically as halia. Further, to fetch water is halia kaising and to drink water is halia ba’ai. Water is water.



Landscape in Western Pantar 159

The term halia excludes seawater, which is distinguished as tawá. Seawater is not a type of water but rather a different substance altogether. To describe fetching water from a well or spring one says halia kaising, whereas to fetch water from the sea is tawá kaising. More significantly, the term halia also excludes certain types of water which are deemed non-potable due to the presence of acid brine. Two types of brine are distinguished in the lexicon: masi and matá. Though neither is potable, masi and matá can be distinguished in terms of corrosiveness. The term masi refers to an acidic supersaline brine which is highly corrosive and will cause clothing and woven items to quickly disintegrate. Contact with the skin causes a mild to severe chemical burn (a property which may be used beneficially as a treatment for tropical ulcers). The term matá refers to a supersaline brine which is less acidic and can be used for washing though not for drinking. The distinction between these two types of brine is made clear in the names for three sulfur creeks which flow across the landscape of west Pantar. The utter lack of other surface flowing water features in west Pantar lends these three streams a certain prominence in WP geography. Originating in natural springs, the flow volume of these streams is relatively small, perhaps less than 2–3 cubic meters per second. Except when briefly augmented by rainfall, the volume remains constant both across time and along the length of the stream, since no other water courses feed these creeks. The precipitating salts are clearly visible in the satellite image where these creeks enter the ocean (Figure 5). ra gA

si

Ma

n oa

H

g

an

si

Ma

S

m ala

g

dan

ad eB

si K

Ma

a Ar

a

iH

s Ma

Masi Kepasali

g an uw

Matá Masigai

Figure 5.╇ Google Earth satellite image showing locations of masi (top and center) and matá (bottom). The length of the longest masi (top) is approximately 6 km from its source to the coast.

160 Gary Holton

While at first glance these streams appear to carry water, closer inspection reveals their contents to be something else. One first notices the complete absence of vegetation along the banks of the streams; a dead zone extends 1–3 m on either side of the streams. The banks of the creeks are heavily mineralized with sulfur precipitates, and a faint odor of hydrogen sulfide gas emanates from the liquid. The residents of west Pantar generally tolerate a high level of brine and saltwater contamination in their drinking and cooking water, but no one in Pantar drinks the water from these creeks. Only halia is deemed potable. The presence of four distinct WP lexemes which translate as types of ‘water’ reflects the WP focus on quality of the substance rather than morphological characteristics. Further evidence for the lesser importance of morphology comes from naming practices. While the continuity of the sulfur streams is clearly recognized, each is designated by several different toponyms. For example, the middle masi in Figure 5, which flows for some 4 km from its northern source to the coast, is denoted by at least three distinct toponyms. At its source, it is known as Masi Hauwang Ara, lit. ‘big and far sulfur creek’. Along the middle of its course it is known as Masi Ke Baddang, lit. ‘crow sulfur creek’. And at its mouth, where it reaches the ocean, it is known as Masi Kepasali (literal meaning obscure). The reason for these distinctions becomes quite obvious when one considers the types of landforms to which these names refer. The area known as Masi Hauwang Ara is an area of bubbling hot springs and seeps with heavy mineralization deposits, covering approximately 2.5 ha. In contrast, the place known as Masi Kepasali consists of a steep-walled canyon where the creek meets the sea. In terms of human affordance, these are very different places. The former can be easily traversed on foot; the latter presents a difficult barrier to travel. In other words, masi and matá are not generic streamscape categories but rather names for different qualities of water. Streamscape toponyms in WP refer not to stream courses but to specific locales along these stream courses. The notion of “watercourse” which would distinguish these two streams is simply not present in WP. In principle it would be possible to refer to each of these watercourses with the generic salu ‘valley’ or with a descriptive term masi salili ‘masi flows’ or matá salili ‘matá flows’, but I have never heard such terms used to refer to masi or matá. The term is reserved for dry valleys which lack permanent streams (though they may contain water wells), and the term salili occurs only . The phenomenon of naming specific parts of a water course based on unique properties of that specific area may be more widespread. In Tanana Athabaskan, spoken in interior Alaska, at least eight streams have two or more names each in different locations along the stream. For example, the Chatanika River is known as Nonilen No’ ‘current-flows-across river,’ Dradlayi Nik’a ‘round whitefish creek,’ and Tthato’ Toya No’ ‘straight-stretch-under-rocks river.’



Landscape in Western Pantar 161

in the phrase halia salili, which describes large bodies of flowing freshwater (not masi or matá). To a certain extent the distinction between halia, masi and matá can be taken to reflect a kind of environmental reality. The WP region is characterized by a lack of water: there are no permanent flowing sources of fresh water. Water is found only in a number of hot springs and in subsurface aquifers fed by these springs, all of which are contaminated to a greater or lesser degree with sulfurous brine. But while the scarcity of water and the presence of acidic brine may be environmental facts, the choice to focus on water quality rather than on streamscape morphology is essentially a linguistic one. The WP language distinguishes water from non-water liquids based on their usefulness to humans. Halia can be consumed; matá can be used for washing; and masi must be avoided. No basic WP terms distinguish flowing water from that in a static body or found in a well. In WP the quality of a liquid dominates its form.

6.

Seascape

Before discussing WP seascape terminology we pause to examine the role of the sea in WP culture. Pantar is a small island. The western part of the island has a coastline of approximately 75 km, compared with an area of approximately 350€km2. No point in this region is more than a couple of hours walk from the ocean. However, WP culture remains essentially a farming economy (see Introduction), with almost no utilization of sea resources beyond the coastal reef. While many WP villages are today situated on the coast, including (clockwise from the east) Beang, Alimake, Koliabang, Puntaru, Boloang, Wolu, Nadda, Tulai, and Beangonong, these coastal villages are all recent settlements, reflecting for the most part migrations in the early 1950s. A Dutch military survey in 1910 reported that the southwestern coastal parts of the island were completely uninhabited (Anonymous 1914:â•›90). Each of these coastal settlements is composed of groups of clans, each of whom still to this day identifies with an “old village” located in the interior of the island. Some of these old villages are still used for traditional ceremonies and for gardening, while others are essentially abandoned. In this sense WP coastal communities remain inward-looking, focused on the interior of the island rather than on the inshore resources. Moreover, the major WP settlements – Kakamauta, Aramaba, and Kalondama – remain inland. That is not to say that WP territory did not traditionally include the coastal regions. On the contrary, long-term use of the coastal regions is attested to by oral history, toponyms, origin myths, and an elaborate terminology for coastal,

162 Gary Holton

intertidal, and pelagic biotic forms. However, WP speakers never developed or adopted a seafaring technology. The village of Puntaru, located adjacent to a reef which forms a natural harbor, has only a handful of dugout canoes, all of questionable seaworthiness. There are no sailing canoes capable of venturing far beyond the reef, and no motorized vessels. This is in sharp contrast to the adjacent Austronesian-speaking communities in Baranusa and Marica, which are centered almost entirely on seafaring and fishing. Indeed, fishermen from Marica occasionally visit WP villages to barter dried fish for local rice. This dried fish is the primary source of fish-based protein in the WP diet. The ocean in WP is distinguished as tawá, a term which contrasts with halia ‘water.’ The defining features of tawá are its expansiveness and its contrast with mua ‘earth; dirt’, dani ‘ground’; ir ‘area, condition’; and por ‘earth’. In WP folklore tawá is bad, a destroyer. Unfortunate people fall into the ocean (tawá-m baulung) or are swallowed up by the ocean (tawá-m muggung). The sea rises to inundate villages (tawá banuakang araung do habbang tang biring) and comes flooding into boats (tawá ma hai-m ipering). When the ocean waves are rough, the ocean is angry (tawá bo aroga). Hence, it is no surprise that seascape terminology in WP is extremely circumscribed. The most elaborate area of seascape terminology refers to the surface features of the shore. The shore above mean high tide is hoang, or beach. Tidal flats which are exposed at low tide are tamu, whereas a large flat reef is hubi and is an excellent place to search for intertidal life when the reef is exposed at low tide. The term hubi refers to physical shape rather than composition; the generic term for coral, whether a living reef or a rock extracted from a reef, is lassang. Any number of small pools, or wee, are exposed on a hubi when the tide is out, and these pools are extensively mapped and uniquely identified by toponyms. The roughly 1000€m long reef on the north side of Puntaru lagoon contains at least six named wee. Small reef fish and eels can be harvested from these wee by using a funnel-shaped basket (sar or pilang) to block off one of the entries to the “pond” while the tide recedes. Such natural pools are distinguished from mappu, man-made fish ponds built from rocks for a similar purpose. Along this same reef are located four named volcanic erratics which protrude from the reef surface. These features are known by the generic term hauwe, and their corresponding toponyms are invariably preceded by this term, as in Hauwe

. The reciprocal relationship between the WP speakers and the Austronesian speakers can be compared to that described for the Yup’ik and Chukchi of Chukotka (Kertulla 2000). The WP speakers specialize in farming, while the Austronesians specialize in marine resources.



Landscape in Western Pantar 163

Reef area

500 m

Figure 6.╇ Map of Puntaru lagoon.

Do Boling, literally, ‘far away least valued moko drum rock.’ These rocks serve as waypoints for navigating the reef surface. Crucially, these reef toponyms represent local knowledge. While the six named wee and four named hauwe just mentioned are widely known to residents of Puntaru, few residents can provide names for similar features on the reef near Koliabang, a short 3 km walk to the south. The primary motivation for naming reef features is to guide subsistence harvest activities on the reef. Since these activities take place at dusk or early evening, residents do not travel far from the village in order to harvest. Hence, there is little knowledge of reef toponyms outside the local area. While the reef, tide flats, and rocky points surrounding the lagoon are recognized as distinct landscape features, there is no generic term referring to lagoon. Admittedly, there are few lagoon-type features on Pantar, and even the lagoon at Puntaru is not well-defined as far as lagoons go (Figure 6). The opening in the reef is nearly as wide as the lagoon is deep, with the result that the lagoon offers only marginal protection from prevailing southerly swell, even at the lowest tides. Instead, the water inside the reef is known simply as tawá ‘ocean,’ undistinguished from the water outside the reef. Hence, my use of the term ‘Puntaru lagoon’ is a completely etic reference to a feature-type which is not locally distinguished. On . A moko (WP kuang) is a bronze kettle drum used as a form of bride wealth in marriage exchange throughout Pantar and neighboring Alor.

164 Gary Holton

the other hand, the area of Puntaru lagoon specifically is distinguished toponymically with the name Ang Blegur, literally ‘Blegur’s market,’ a reference to a barter market which has been held traditionally at this location. The paucity of seascape feature terms in WP reflects the relatively minor role of the sea in WP culture. Though residing within close proximity of the ocean, WP speakers have not adopted a maritime culture. WP speakers engage with the ocean only in order to practice subsistence harvest of intertidal organisms on the exposed reef. Knowledge of and recognition of subsurface features is extremely limited.

7.

An areal perspective

The data presented above show that WP landscape terminology and toponyms reflect the Pantar landscape and the WP speakers’ relationship to that landscape. While generic terms for elevations do exist, convex landforms such as mountains and hills are much more likely to be referred to using either (i) a locational noun denoting an area located uphill, or (ii) a toponym denoting a co-located village, settlement, or garden area. Vegetated areas are denoted by terms referring to cultivation, reflecting the importance of subsistence agriculture. Seascape and coastal terminology is limited, directly reflecting a cultural focus on subsistence agriculture and coastal foraging rather than active harvesting of ocean resources. Clearly the most striking feature of WP landscape terminology is the lack of a generic term for ‘river.’ Rather, WP provides names for various types of aqueous substances: halia ‘water,’ matá ‘slightly sulfurous brine,’ and masi ‘highly sulfurous brine.’ Here the essential property is the quality of the substance, not the shape or character of the landform in which the substance flows or stands. In WP ‘water’ is halia regardless of whether it is found in a lake, running down a mountain, in a bubbling hot spring, or in a drinking glass. The existence of distinct terms for different types of brine concentrations might be taken to be a mere case of mundane lexical specialization, risking a comparison with the trite “interestingness” of Eskimo snow vocabulary (Pullum 1989). However, the significance of WP streamscape vocabulary lies not in these distinct terms themselves but rather in the fact that WP chooses to focus on the quality of the substance rather than on the physical character of the landform created by that substance. From a geomorphology viewpoint halia, masi, and matá are essentially similar; they differ not in morphology but rather in chemistry. This focus on substance rather than form is not unique to WP but is found in at least some neighboring languages, in spite of their vastly different geographic and climatic settings. Given that the volcanic environment of sulfurous brines is unique to WP, it is not surprising that other languages of the Alor-Pantar family lack terms



Landscape in Western Pantar 165

for sulfurous brine or even for sulfur itself. On the other hand, there is some variation in the treatment of the ‘river’ term. Teiwa, spoken just 15 km from the WP region on the northern peninsula of Pantar Island, distinguishes yir ‘water’ from the generic landscape term bo’oi ‘river’ (Klamer 2011). Similarly, the Klon language of southwest Alor distinguishes araa ‘water’ from weer ‘river’ (Baird 2008). Teiwa bo’oi and Klon weer are part of the standard landscape terminologies, denoting reference to landscape objects and serving as the generic component of toponyms. In Klon one can weer agai ‘go to the river’ and also Hwak Weer mi taa’ ‘sleep at Hwak River.’ In contrast, WP uses the substance terms halia, masi, and matá to fulfill this function. There is no WP term cognate with Teiwa bo’oi or Klon weer. Elsewhere in the Alor archipelago we find streamscape terminologies more similar to that found in WP. Abui is spoken in the rugged interior mountains of central Alor, with many settlements located above 1000 m in elevation (the self-designation abui also means ‘mountain’). At this altitude mountain valleys are often shrouded in mist, and water sources are relatively abundant. The damp climate of the Abui language region contrasts greatly with the dry savanna of the WP region. Nevertheless Abui also lacks a distinct term for ‘river.’ There is no Abui term cognate with Teiwa bo’oi and Klon weer, and though Abui lu ‘valley’ may metonymically denote the river in the valley (compare WP salu), the more standard way to denote a body of water, flowing or not, is by combing the term ya ‘water’ with the augmentative wal (compare WP wala). Thus, ya wal ‘river, pond’ (Kratochvíl & Delpada 2008). The augmentative morpheme is found in other Abui landscape terms as well, such as tama wal ‘ocean.’ (Although the WP cognate wala does not occur with landscape terms, it serves a similar function, as in kuba wala ‘midwife,’ derived from kuba ‘old woman’.) Thus, in Abui a river – or any other body of water – is simply ‘water.’ These limited comparative data provide some evidence that the conceptualization of landscape in WP is more deeply embedded within the language family, reflecting a unique world view. Such a tentative conclusion must of course be viewed with great caution. As noted at the outset, an exotic landscape invites exotic linguistic interpretations; to avoid such pitfalls we must rely on solid ethnographic methodology. Enormous progress has been made recently toward the documentation of grammatical structures in the Alor-Pantar languages, yet the landscape domain remains only cursorily documented. Given the demonstrated categorical mismatch between landscape categories in WP and those in English, Indonesian, and other languages of wider communication, it is necessary to do more than simply elicit translational equivalents of a given set of landscape terms. Rather, as Burenhult and Levinson (2008) suggest, it is necessary to tease out the underlying indigenous landscape ontologies. Much work remains to be done.

166 Gary Holton

References Anonymous. 1914. De eilanden Alor en Pantar, residentie Timor en onderhoorigheden. Tijdschrift van het Koninklijk Nederlandsch Aardrijkskundig Genootschap 31: 70–102. Baird, Louise. 2008. A Grammar of Klon: A Non-Austronesian Language of Alor, Indonesia. Canberra: Pacific Linguistics. Barnes, Robert H. 1982. The Majapahit dependency Galiyao. Bijdragen tot de Taal-, Land- en Volkenkunde 138: 407–412. Burenhult, Niclas & Levinson, Stephen C. 2008. Language and landscape: a cross-linguistic perspective. Language Sciences 30: 135–150. Dietrich, Stefan. 1984. A note on Galiyao and the early history of the Solor-Alor islands. Bijdragen tot de Taal-, Land- en Volkenkunde 140: 317–326. Fox, James J. (ed.). 1988. To Speak in Pairs: Essays on the Ritual Languages of Eastern Indonesia. Cambridge: CUP. Grimes, Charles E., Therik, Tom, Dix, Barbara & Jacob, Max. 1997. A Guide to the People and Languages of Nusa Tenggara. Kupang, Indonesia: Universitas Kristen Artha Wacana and Alfa Omega Foundation. Kertulla, Anna. 2000. Antler on the Sea: The Yup’ik and Chukchi of the Russian Far East. Ithaca NY: Cornell University Press. Klamer, Marian. 2011. A Grammar of Teiwa. Berlin: Mouton. Klamer, Marian, Holton, Gary & Kratochvíl, František. 2009. The languages of Alor-Pantar (Eastern Indonesia): A (re)assessment. Paper presented at the 11th International Conference on Austronesian Linguistics, June 25, 2009, in Aussois, France. Kratochvíl, Frantisek & Delpada, Benny. 2008. Kamus Pengantar Bahasa Abui. Kupang, Indonesia: UBB-GMIT. Levinson, Stephen C. 2008. Landscape, seascape and the ontology of places on Rossel Island, Papua New Guinea. Language Sciences 30: 256–290. Poorter, R. P. E., Varekamp, J. C., van Bergen, M. J., Kreulen, R., Sriwana, T., Vroon, P. Z. & Wirakusumah, A. D. 1989. The Sirung volcanic boiling spring: an extreme chlorine-rich, acid brine on Pantar (Lesser Sunda Islands, Indonesia). Chemical Geology 76: 215–228. Pullum, Geoffrey K. 1989. The great Eskimo vocabulary hoax. Natural Language and Linguistic Theory 7: 275–281. Ross, Malcolm. 2005. Pronouns as a preliminary diagnostic for grouping Papuan languages. In Papuan Pasts: Cultural, Linguistic and Biological Histories of Papuan-speaking Peoples, Andrew Pawley, Robert Attenborough, Jack Golson & Robin Hide (eds), 15–66. Canberra: Pacific Linguistics. Wassmann, Jürg. 1997. Finding the right path: The route knowledge of the Yupno of Papua New Guinea. In Referring to Space: Studies in Austronesian and Papuan Languages, Gunter Senft (ed.), 143–174. Oxford: OUP.

chapter 8

Hawaiian storied place names Re-placing cultural meaning Renee Pualani Louis

“We live in a time of un-naming, in a time when old names for the land, names given in honor, happiness, and sorrow have been set aside for marketing jingles that commemorate little more than a desire for sales, for ka mea poepoe, the round thing, money” (DeSilva 1993). Hawaiian place names are storied symbols reflecting Hawaiian spatial and environmental knowledge. Performed in daily rituals they were a conscious act of reimplacing genealogical connections, recreating cultural landscapes, and regenerating cultural mores. This chapter highlights the sensuous nature of Hawaiian place names, examines the processes by which they are incorporated into the cultural landscape, investigates cultural conflicts and problems involved with naming places in the post-contact/moderncolonial era including the standardization of place names, and advances transmodern solutions that re-place the old names DeSilva refers to without fueling existing cultural conflicts.

1.

Introduction We live in a time of un-naming, in a time when old names for the land, names given in honor, happiness, and sorrow have been set aside for marketing jingles that commemorate little more than a desire for sales, for ka mea poepoe, the round thing, money. We who learn and love these old names are, therefore, people of two worlds, residents of rival geographies. We lead our everyday lives on the congoleum, concrete, and tiff-green crests of Hawai‘i’s Bay Views, Crest Views, Soda Creeks, and Enchanted Lakes. But when our souls wither and thirst, we seek nourishment in that other, deeper geography where the true names of our ‘āina are sung by the stones themselves, in what Ellen Pendergast has called “ka ‘ai kamaha‘o o ka

168 Renee Pualani Louis

‘āina – the astonishing food of the land,” that we, the stone eaters of this land, find sustenance and comfort, pride and purpose.  (DeSilva 1993)

My kuleana (life purpose/responsibility) and my passion relate to Hawaiian place names. Hawaiian place names are storied symbols reflecting Hawaiian spatial knowledge of the environment. Many Hawaiian place names performed in daily rituals were a conscious act of reimplacing genealogical connections, recreating cultural landscapes, and regenerating cultural mores. They constitute a critically important body of Hawaiian cultural knowledge and are used in all forms of Hawaiian cartographies. Hawaiian cartographies are interactive presentations of place as ‘experienced space.’ They situate mapping in the landscape and encode spatial knowledge into bodily memory via repetitive recitations and other habitual performances. When Hawaiian place names were incorporated into Western cartographic maps, they were transformed epistemologically. They went from representing place as a repository of cultural knowledge to representing place as an object on the landscape. My kuleana is to continue adding new dimensions to Hawaiians’ relationships with ‘place’ and specifically with their kulāiwi (ancestral lands) through the sharing of storied place names. This text will explore the nature of Hawaiian place names with a multisensory approach to understanding their power, pleasure, and pedigree. It examines the processes by which Hawaiian place names are incorporated into the cultural and cognitive landscape, presents a brief description of the cartographic culture clash in the post-contact/modern-colonial era, and advances transmodern solutions that re-place the old names DeSilva refers to without fueling existing cultural conflicts.

2.

Sensuous nature of Hawaiian place names

Hawaiian place names are more than just identification tags for the features and/ or phenomena of the physical world. They are also powerful cognitive mechanisms that unfold the richness of the Hawaiian cultural landscape, revealing as much about Hawaiian perceptions of the metaphysical world (their beliefs about their gods, their interactions with nature, and their cultural practices) as they do about the places and times to which they refer. By naming a place, whether it is an island, a forested region, a coastal area, a celestial expanse, or the lo‘i (taro pond) in their backyard garden, Hawaiians were placing parts of themselves onto the ‘placescape,’ inserting themselves into the world as they understood it. Thus, Hawaiian place names simultaneously provide a glimpse into the harmonious



Hawaiian storied place names 169

relationship Hawaiians maintained with the environment and reveal the cognitive clarity of the Hawaiian imagination. Like many place names, Hawaiian place names serve as constant reminders of past events, cautionary tales, and epic tragedies. However, Hawaiians carefully and mindfully wove many place names into their daily lives via various cultural practices. Many Hawaiian place names are often only understood with the mo‘olelo (historical account) that accompanies them and usually only by those within its genealogical proximity. They are cultural reference points, helping each succeeding generation find and define themselves. Knowledge of their meaning provides insight into the importance these place names had in shaping Hawaiian cultural identities. Sharing the names and meanings of places is more than a conscious act of cultural regeneration for Hawaiians; it is also an integral part of their cartographic tradition. Hawaiian cartographic traditions, like Western cartographic traditions, are social constructions of spatial knowledge systems; however, they evolved along a different course. Hawaiians encoded their knowledge of the environment into archival graphic forms such as ki‘i pōhaku (petroglyphs), kapa (bark cloth) designs, and various royal insignia – ‘ahu‘ula (feather capes/cloaks), mahiole (helmets), and kahili (royal standard). However, they also incorporated their spatial understandings into various cultural practices such as mo‘olelo (historical accounts), ‘ōlelo no‘eau (proverbs), mele (song), oli (chant), mo‘o kū‘auhau (genealogy), and hula (dance). This is a form of cartography known as ‘performance or ritual cartography’. From a Hawaiian point of view, place is a part of a larger order of a living Earth. Each place is alive and made up of distinctive sights, sounds, smells, sensations, and essences. Place has depth and is part of a heaven-land-ocean continuum that stretches between the physical and metaphysical planes. Hawaiians not only named various regions of the heavens, lands, and oceans, personalizing their island environment, they also recognized various kinolau (physical manifestations of gods) within the continuum. Lono, noted as the god of fertility, for example, was seen in the skies in the form of dark clouds or lighting. From the ‘āina [land], he would rise back up to the heavens as steam billowing from active volcanoes. On land, Lono would manifest himself as a pig or a kukui tree. In the ocean, he would take the form of the (Oliveira 2006:â•›223) humuhumunukunukuapua‘a fish. 

Many places had distinct names for the various winds and rains that sometimes went beyond describing their individual traits and would also identify specific phenomena associated with them.

170 Renee Pualani Louis

Ka makani kā ‘aha‘aha la‘i o Niua, for example, alluded to “the peaceful ‘aha‘aha breeze of Niua that drives in the ‘aha‘aha fish”. In this example, ‘aha‘aha refers to both the fish and breeze of the same name. Fishermen knew that when this breeze blew, it was the right time to launch their canoes in search of the ‘aha‘aha fish.  (Oliveira 2006:â•›273–274)

This approach also applied to other elements of nature like cloud types, lava formations, ocean wave movements, stars and other astronomical features. At this point it should be rather obvious that from a Hawaiian point of view the only way to truly ‘know a place’ is to interact with it and experience it sensually. While all human beings have similar sensory receptors with about the same physiological capabilities in which to receive information about the world, different cultures in different times mix the information received in varying proportions, placing greater emphasis on different sensory systems. Our sensuous experience and perception of the world “is grounded on previous experience and expectation, each dependent on sensual and sensory capacities and educational training and cultural conditioning” (Rodaway 1994:â•›5). This means that each culture’s concept of reality and their associated representations or presentations of their spatial environment are tremendously different. In fact, these representations or presentations need not be visual pictures; oftentimes a sound or a smell will suffice (Downs & Stea 1977:â•›23). Many Western-educated people acquire their spatial knowledge from many sources including archival documents such as maps. Polynesian-educated navigators acquire their spatial knowledge, among other things, from the rhythm and direction of ocean movements. According to Oliveira (2006), Hawaiians are capable of acquiring spatial knowledge from nine sensory systems, prioritizing metaphysical sensuous modes. Contrary to popular knowledge, this type of sensual participation is not ‘supernatural’ or ‘extra-ordinary’. It is a naturally conditioned response to ‘tuning in’ to the natural world (Cajete 2000:â•›20). At the most intimate depths of understanding and meaning, storied place names reveal the sensual nature of Hawaiian cartographic traditions. Through place names such as Kepuhi, Kamaiko, ‘Umiwai, Mokuoka‘e, and Kealakekua we learn how Hawaiians experience the world with all their senses and incorporate these experiences into their cultural landscape.

Kepuhi (sound) Kepuhi is the blowhole in the Kealakekua area of Hawai‘i Island. It acts as an early warning system for storms. After years of observation the community learned to listen to the way the waves crashing at Kepuhi sound. If you just hear the water rushing through the rocks, it is okay to go out for deep sea fishing. However, if it sounds like an underwater explosion of rocks slamming into each other, it is



Hawaiian storied place names 171

time to bring the canoes in and lash them down because rough weather will be approaching.

Kamaiko (smell) Kamaiko was a heiau (temple) for preparing the body for burial. It sat on the promontory over Palemanō and was given the name Kamaiko because rotting human flesh stench smelled the same as a rotting maiko fish. This heiau was not used by the Ali‘i (chiefly class). It was used by the everyday people in the community. The burial practices in this area entailed drying the body on a wooden cross until the flesh could be pulled away from the bones. While the prized bones were cleaned off and put into a basket and buried, the flesh was thrown into the ocean, creating a cyclical connection between Palemanō and Kamaiko. According to the community members, Palemanō describes the protected ledge directly beneath the promontory where manō (sharks) hovered and consumed the dried rotted fleshy remains of the dead. Many community members believed, and still believe, the sharks were or are their extended family members watching over them while they fished and protecting them from harm. This kind of belief system ensures cultural protocols of respect, and responsible ocean resource management practices are maintained. ‘Umiwai (taste) ‘Umiwai is an area with a brackish water well that caused those people not accustomed to drinking brackish water to gag. The name was meant to be an indication of the brackish water, although many people think ‘umi refers to either the number ten or an Ali‘i (chief) named ‘Umi. This is an excellent example of why the stories are so important. It is so easy to misunderstand the literal meaning of place names, especially when there are so many different ways to interpret them. Mokuoka‘e (sight) Mokuoka‘e is the name of the land area where Kamehameha victoriously battled his cousins Kīwala‘ō and Keōua to become Ali‘i Nui (High Chief) of Hawai‘i Island. It was named Mokuoka‘e because the ‘āina (land) became smeared with the rotting bloodied bodies of fallen warriors both literally and figuratively. It was an epic battle, fought in historic times and memorialized in the landscape. Kealakekua (essence) Lastly, Kealakekua is a significant place name for both the Ali‘i (chiefs) and Maka‘āinana (people who worked the land) social classes. It is most commonly known from the Ali‘i perspective as the trail the Makahiki (winter harvest festival) gods used to start their island circuit. Thus, the name is invested with the mana

172 Renee Pualani Louis

(powerful spiritual essence) associated with the annual Makahiki festival and its rituals. It honors the memory of Lono and provides the Ali‘i an opportunity to rest from warfare and enjoy the bounty of life. From the Maka‘āinana perspective, Kealakekua is the trail the shark god of Ka‘ū, Kua, walked between the shoreline and the mountain gardens, helping the people living in this area with their daily tasks. He was welcomed immediately and treated like family. He helped the farmers in their gardens and the fishermen with their catch. No one knew where he came from until it was time for him to leave. The people named the trail he walked between the shoreline and the mountain gardens after him, forever memorializing it in their cultural landscape. In this example, the mana of Kealakekua exudes from various perspectives and dimensions. Yet, all these dimensions involve metaphysical realities. Whether it is a reenactment of an ancestor elevated to Akua (God) status through generations of ritualized performances, or a god living and working among the people, the stories associated with Kealakekua acknowledge a metaphysical presence in the Hawaiian cultural landscape.

3.

Incorporative practices A place tells me who I am and who my extended family is. A place gives me my history, the history of my clan and the history of my people. I am able to look at a place and tie in human events which affect me and my loved ones. A place gives me the feeling of stability and belonging to my family both living and dead. A place gives me a sense of well being and knowledge that I am accepted by all who have experienced my place.  (Kanahele n.d.)

Place names and their cultural meanings are generated by the constant interaction of people and their environment. For generations, Hawaiians incorporated their spatial knowledge, moral underpinnings, and spiritual relationships into their heaven-land-ocean continuum or ‘placescape’. They conveyed this spatial knowledge via various performance-oriented cultural practices. These cultural practices employed metaphor and mnemonic devices, like storied place names, sounds (including pitch, rhythm, voice quality, and intonation), gestures (including facial expressions and body movements), and artistic designs, to relay their message to succeeding generations. For each Hawaiian, the ‘placescape’ became the symbolic archive for those stories that depict the “lives and deeds of the immortal beings from whom he himself is descended, and whom he reveres” (Tuan 2001:â•›158). Cajete refers to this as ‘ensoulment’:



Hawaiian storied place names 173

This projection of the human sense of soul with its archetypes has been called the “participation mystique,” which for Native people represented the deepest level of psychological involvement with their land and which provided a kind of map of the soul. The psychology and spiritual qualities of Indigenous people’s behavior reflected in symbolism were thoroughly “in-formed” by the depth and power of their participation mystique with the Earth as a living soul. It was from this orientation that Indian people developed “responsibilities” to the land and all living things, similar to those that they had to each other. In the Native mind, spirit and matter were not separate; they were one and the same.  (Cajete 1994:â•›186)

Through the use of kinolau (physical manifestations of gods), Hawaiians metaphorically recognized various topographic features and elements of nature as representations of deities. In fact, the landscape itself was intimately known as Papahānaumoku, the sacred earth mother that gives birth to islands. Every part of the landscape is considered fertile and capable of nourishing both body and imagination. It is Papahānaumoku that provides a Hawaiian ontological and epistemological framework for spatial knowledge. For example, the Hawaiian word ‘āina (land) has a deeper definition as ‘that which feeds the body’ and ‘inspires the mind’. The Hawaiian word na‘au is defined as both the ‘intestines,’ that body part which we feed, and the ‘mind,’ the source from which ancestral knowledge emanates. Add to this the Hawaiian concept of kulāiwi, those ancestral lands where the bones of kūpuna (ancestors) rest, and the metaphor of the ‘āina being imbued with ancestral knowledge that feeds our bodies and souls becomes even clearer. Recognizing the landscape as Papahānaumoku sets the stage for metaphoric understanding of nature. Several topographic features were associated with a single deity, like those identified in the Pele lore, such as: Haleama‘uma‘u as the home of the goddess Pele; Kauhi‘īmakaokalani as a dog demigod (also known as the Crouching Lion); and Kohelepelepe as the representation of Kapo’s vagina. This practice was not reserved for the major Akua (gods), it was also extended to demigods such as Kamapua‘a, who took the form of a … handsome man, single pigs and dozens of pigs, clouds (ao), plants (kukui, ‘uha=loa, ‘āma‘uma‘u fern, kūkae=pua‘a grass [literally ‘pig excreta’]), the small triggerfish humu-humu=nuku-nuku=a=pua‘a, as well as that of the great god Lono. (Pukui et al. 1974:â•›260–261)

Sometimes a single tree would be the symbolic manifestation of two gods. For example, the trunk of the ‘ohi‘a lehua tree was the physical manifestation of the god Kū but the blossoms were sacred to the goddess Pele.

174 Renee Pualani Louis

4.

Cartographic culture clash in post-contact/modern-colonial Hawai‘i The transition from an oral culture to a literate culture is a transition from incorporating practices to inscribing practices. The impact of writing depends upon the fact that any account which is transmitted by means of inscriptions is unalterably fixed, the process of its composition being definitively closed. The standard edition and the canonic work are the emblems of this condition. This fixity is the spring that releases innovation. When the memories of a culture begin to be transmitted mainly by the reproduction of their inscriptions rather than by ‘live’ telling, improvisation becomes increasingly difficult and innovation is institu(Connerton 1989) tionalised. 

By the time Europeans arrived on the shores of Hawai‘i a little more than two centuries ago, both Western and Hawaiian cartographies reflected the philosophical and technological advancement of each worldview. Western cartography reflected the technical progress of ‘accurate representation’ focusing solely on scientific advancement, supported, in European practice, by Descartes’ separation of the mind from the body. Hawaiian performance cartographies reflected the sensual, intimate, and multidimensional relationships Hawaiians developed with their natural and spiritual environments. These two cartographic traditions met on the shores of Kealakekua with the arrival of Captain Cook, precipitating an opportunity whereby differing understandings of space/place and their accompanying (re)presentations began to intermingle. After the return of Captain Cook’s ships to Europe, much of Hawai‘i’s modern history was carefully documented as ethnographic accounts in journals and other historical texts, including maps. Europeans began publishing maps and other images of the islands as part of these ethnographic accounts, for their own use. One of the first printed maps of Hawai‘i included a detailed plan of the bay which Hawaiians living in this area knew to be Kapukapu. However, on this map, the bay was given the name Kealakekua Bay and spelled phonetically according to what the English understood the Hawaiian names to be, in this case, Karakakooa Bay (see Figure 1) (Fitzpatrick 1986:â•›15–18). Cartographically speaking, maps like this one were made for European use and had not yet affected the cognitive cartographies of Hawaiians. Changes to Hawaiian cartographies began in missionary-run schools and culminated in the first Western style maps being made in Hawai‘i by Hawaiians. In the mid-1830s the need for maps in the classrooms for geographic education of Hawaiians was widely recognized and the most significant development in the effort to provide these maps began at the Lahainaluna seminary. As the head of seminary, Lorrin Andrews learned the process of copperplate engraving and printed his



Hawaiian storied place names 175

Figure 1.╇ First map of Hawai‘i compiled by Captain William Bligh based on the journals of Captain James Cook. Source: scanned image of original map for instruction on  historic maps in K-12 classrooms by The Newberry Library as part of a National  Endowment for the Humanities grant, Chicago, IL. Downloadable pdf available at  .

first world map in November 1834. The technique was taught to his students and they became proficient at it, producing five other maps by early 1836 (Fitzpatrick 1986:â•›107–109). In 1838, ten years prior to the Mahele, one of the most important maps of Hawai‘i was drawn and engraved by a student, Kalama (see Figure 2). According to Fitzpatrick this eight-sheet map of Hawai‘i is important for several reasons: it was made by Hawaiians; it reflects the proficiency which the students at Lahainaluna achieved in a short period of time; it corrected major errors in existing maps of Hawai‘i; and it was instrumental in preserving the names of basic Hawaiian land subdivisions (Fitzpatrick 1986:â•›112, caption for Figure 66). Fitzpatrick further postulates that during the Mahele, when property transactions required the name of the ahupua‘a in which the parcel was located, Kalama’s map

. This text will not attempt to discuss the horror known as the Mahele of 1848 in great detail. There are several excellent books, journal articles, and online resources available on this topic, including Chinen’s (1958) The Great Mahele: Hawai‘i’s Land Division of 1848, Kame‘eleihiwa’s (1992) Native Land and Foreign Desires – Pehea Lā E Pono Ai, Andrade’s (2001) Hā‘ena, Ahupua‘a: Towards a Hawaiian Geography, Osorio’s (2002) Dismembering Lahui: A History of the Hawaiian Nation to 1887, Herman’s (1999) “The Aloha State: place names and the anti-conquest of Hawai‘i” in Annals of the Association of American Geographers, and Wikipedia.

176 Renee Pualani Louis

Figure 2.╇ A portion of the first Western style map of the Hawaiian Islands made by a Hawaiian student, Kalama, at the Lahainaluna Seminary in 1838. Only two complete copies exist, one in the Hawai‘i State Archives and the other in the Royal Geographical Society in London. (Image from .)

was the only one available for reference and “could have been consulted by the king, government officials, and private individuals” (Fitzpatrick 1986:â•›112). During this time Hawai‘i underwent a massive eco-socio-political shift. The ‘āina concept changed dramatically from ‘that which feeds the body and inspires the mind’ to ‘an objectified commodity’. Through various legal acts by the Hawaiian government under the guidance of non-Hawaiian advisors, kulāiwi became allocated parcels of personal property. The shift to private property required the work of land surveyors. Surveying in Hawai‘i played a huge role in how the islands were represented. European surveying and map-making techniques were



Hawaiian storied place names 177

introduced in 1779 when Captain Cook spent several weeks at Kapukapu (see the detailed inset on Figure 1). However, maritime surveying is primarily concerned with charting coastlines and navigational hazards, whereas land surveying is concerned with the description of the extent of real estate (property boundaries) as well as topographic mapping. Land surveys were a necessary step for all land claims to be awarded title (Moffat & Fitzpatrick 1995:â•›53). In 1843 William Patterson Alexander joined the Lahainaluna seminary and taught surveying to many of the students. Both he and his students conducted “substantial amounts of surveying in support of kuleana claims” (Moffat & Fitzpatrick 1995:â•›54). Unfortunately, by the time the Hawaiian government survey was established in 1870, several problems existed resulting from boundary lines survey inaccuracies, leading to great difficulties with land ownership records. W.€P. Alexander was hired to provide a workable solution. He began by creating the most accurate triangulation network possible that other surveys could connect to. In Kona, this primary triangulation network was completed by Joseph Swift Emerson in 1883. While in the field, Emerson’s survey crew completed several sketched views of the area surrounding Kapukapu (see Figures 3–5).

Figure 3.╇ Emerson’s sketches 14 & 15 of Ke‘ei from Makolehale station looking southerly.

178 Renee Pualani Louis

Figure 4.╇ Emerson’s sketch 18 of Kealakekua Bay and bluff from Palemanō station.

Figure 5.╇ Emerson’s sketch 21 of the sea coast looking north from Lae o Kanoni station.

Such depictions pose well-known problems for Hawaiians and indigenous people as a whole. They portray the landscape as a vast open area, devoid of Hawaiian cultural features or activities. Occasionally, surveyors would jot notes about place names. However, Emerson was not one to take many notes that did not pertain to the business at hand. And we must keep in mind that the surveys were not meant to include information about culturally important features.



Hawaiian storied place names 179

Emerson’s only concern was to delineate a triangulation network on Hawai‘i Island that provided other surveyors the opportunity to accurately locate boundary lines between parcels. Like in most encounters of colonial ‘third space’, there has been significantly more of an effect on the colonized society than on the colonizing society. Much of this is a direct result of the inscriptive and authoritative practices of Western societies. While Hawaiians readily incorporated Western maps as another ‘tool’ in their performed presentation of place, Hawaiian spatial knowledge went largely unnoticed in Western maps except for the inclusion of place names. It is perhaps surprising that after all the change Hawaiian communities have endured there is still a vibrant Hawaiian culture being maintained. Some of the cultural elements are obvious and readily observable; others are only known in distinct circles or only shared with those who can demonstrate a trusted responsibility. The next section provides insight into how these culturally significant elements missing from Western maps in the form of storied place names are re-placed using transmodern solutions without fueling existing cultural conflicts.

5.

Transmodern solutions The transmodern is a shared knowledge space in which equivalences and connections between differing rationalities can be constructed. Communication, understanding, equality and diversity will not be achieved by others adopting Western information, knowledge, science and rationality. It will only come from finding ways to work together in joint rationalities and in knowledge spaces constituted through these joint rationalities.  (Turnbull 2003:â•›3)

The Hawai‘i Board on Geographic Names (HBGN) began working on a transmodern solution to this cultural cartographic conflict in 1999. They began with a process to diacritically correct those Hawaiian place names found on US Geological Survey (USGS) topographic maps. It was and continues to be exciting work because the corrections are being incorporated into the Federal Geographic Names Information System (GNIS) database. Each diacritically corrected place name is considered an ‘official name’ that must be used on all government publications, including maps (see Figure 6). While these orthographic corrections are a critically important first step in the process of incorporating more Hawaiian cultural ‘meaning’ into the maps, they are still tied to a Western epistemology of place/space. Hawaiian spatial knowledge presentation is interactive, multisensual, and multidimensional. The next logical step toward accomplishing this goal is to adapt Western technological

180 Renee Pualani Louis

Figure 6.╇ Waikīkī Digital Raster Graphic (DRG) clip. Left: 1983 series with no diacritical marks. Right: 1997 series with diacritical marks included.

innovations in digital multimedia to (re)present place in a more Hawaiian cultural context – as a repository for cultural knowledge. In 2007, the US Geological Survey (USGS) provided funds to begin working on a webpage modeled after the Coeur d‘Alene Tribe’s Native Names Project, but with more Hawaiian cultural nuances, such as genealogical connections, songs, chants, mythical and/or legendary accounts. The Hawai‘i webpage will provide both Hawaiians and non-Hawaiians, among other things, an opportunity to hear the Hawaiian pronunciation for Hawaiian place names via digital audio recordings, see pictures of the place via digital photo galleries, and learn many different stories associated with the places via digital video clips. It is designed similar to a wiki or other ‘crowdsourcing’ webpages, but with judiciously granted logins. All edits and additions to the webpage are to be stored temporarily awaiting HBGN approval. The project draws upon the expertise and resources of the HBGN board members to facilitate language and other cultural aspects of this project. After much deliberation, the HBGN chose the island of Kaho‘olawe (see Figure 7) for the pilot project for the next step in the process. Since the last native residents of Kaho‘olawe departed the island in the early part of the twentieth century, almost all of the Kaho‘olawe’s place names known today are derived from maps and written accounts, rather than from the mouths of those living on and familiar with the island. While the loss of innumerable place names and their meanings is tragic, it provides the pilot project with a practical starting point: concentrating on the technical methodology.



Hawaiian storied place names 181

Kilauea

Kaua‘i

Kalaheu

Ni‘ihau

Kapaa Lihue Koloa

O‘ahu

Mililani Town Waipahu

Kailua

Moloka‘i

Honolulu

Kahului

Lāna‘i

Lahaina Kihei A

Paia

Maui

Makawao

Kaho‘olawe Hawai‘i Waikoloa Village

Honokaa

Hilo Kailua Kona

Holualoa

Captain Cook

Mauna Loa

Keaau Mountain Pahoa View

Hawaiian Ocean View

Figure 7.╇ Location of Kaho‘olawe in the Hawaiian Islands.

Figure 8.╇ Hawaiian place names database table design view.

A simple database was created to record each place name (see Figures 8 and 9), which allows for easy updates. It is meant to compliment the GNIS database currently maintained by the US Geological Survey (USGS). There are twenty-five fields in the database. Many are meant to provide the information necessary to fill in the Hawai‘i Geographic Name Application form with relative ease, in the event community members want to add a place name to

182 Renee Pualani Louis

Figure 9.╇ Hawaiian place names database input form.

GNIS. Any information that can be updated is recorded in the database as a hyperlink address. The ‘explanations’, ‘cultural significance’, ‘historical information’, and ‘etymological information’ data fields are stored as Universal Naming Convention (UNC) hyperlink .pdf files (i.e. \\server\share\path\filename). This allows each of these documents to be updated as more information becomes available. The ‘map’, ‘images’, and ‘videos’ data fields are stored as Uniform Resource Locator (URL) (e.g. http://hawaii.gov/dbedt/gis/bgn), allowing these digital files to be added to as necessary. One of the most important data fields is ‘cultural significance’. It will include ethnographic information such as songs/chants that are about the place, and myths/legends/genealogies. The project will not infringe on copyrighted materials by posting .mp3 files or scanning entire books. However, it will transcribe/ scan documents considered public access from government agencies. Thus far 132 place names from Roland Reeve’s (1993) “The Place Names of Kaho‘olawe” have been transcribed into the appropriate hyperlink document (see Figures 10–11).



Hawaiian storied place names 183

Wai o Kanaloa

(Puna - Spring)

Explanation

The following is an excerpt from, The Place Names of Kaho‘olawe: A Study of Those Place Names Known from the Island of Kaho‘olawe (Reeve no date:199)

Inez Ashdown: – “Kana-Ioa was an ancient name of the island, and he was the god who visited there with Kane and made the hidden water spring, Wai o Kanaloa, that comes up out of the sea at Kanapo‘u area.” (Ashdown’s letter to Mary Kawena Pukui, dated 27th March, 1960, a copy of which is held among her papers at the Maui Historical Society: page 5)

Reeve, R. B. (no date). The Place Names of Kaho‘olawe: A Study of Those Place Names Known from the Island of Kaho‘olawe. K. I.C. Commission. Honolulu, State of Hawaii.

Figure 10.╇ Sample of ‘Explanation’ document. Wai o Kanaloa

(Puna - Spring)

Etymology

The following is an excerpt from, The Place Names of Kaho‘olawe: A Study of Those Place Names Known from the Island of Kaho‘olawe. (Reeve no date:199)

This place name comes to us from a letter written by Inez Ashdown to Mary Kawena Pukui and dated 27th March, 1960, a copy of which is held among Ashdown’s papers at the Maui Historical Society. In this letter she writes that; “Kanaloa was an ancient name of the island [Kaho‘olawe), and he was the god who visited there with Kane and made the hidden water spring, Wai o Kanaloa, that comes up out of the sea at Kanapo’u area.”(page 5) Ashdown does not tell us where exactly this brackish water spring bubbles up from the sea bed, but it would appear that it was located somewhere in Kanapou bay. She also provides us with no indication as to where or from whom she learned this place name. There does exist, of course, the possibility that “Wai o Kanaloa” is simply another name for the spring which Ashdown elsewhere refers to as “Punawai Honu (or Huna)” Or “Puna Pee” (see above). To further confuse the matter, Ashdown speaks of a “hidden water spring” apparently named Kanaloa which “comes up out of the sea” to the west of Kamohio bay. In a note on page 57 of her annotated copy of McAllister’s Archaeology of Kaho’olawe (on file at the Archives of the Bishop Museum) Ashdown states that the place name Kaneloa, situated to the west of Kamohio bay, should be, “Kanaloa not Kaneloa. Kanaloa was the very ancient god-name of Kahoolawe, referring to when he visited there with Kane. There is a hidden water spring so named there, that comes up out of the sea.” One can only wonder whether this is a different spring, or whether Ashdown has simply confused the location of “Wai o Kanaloa”.

Reeve R. B. (no. date). The Place Names of Kaho‘olawe: A Study of Those Place Names Known from the Island of Kaho’olawe. K. I. C. Commission. Honolulu, State of Hawaii.

Figure 11.╇ Sample of ‘Etymology’ document.

184 Renee Pualani Louis

Figure 12.╇ Hawai‘i Board on Geographic Names Hawaiian place names project website (Beta version).

The successful creation of the database and use of technology prompted the project to post all the Hawaiian place names reviewed by the HBGN on all islands. While most of the information appears to be nothing more than a place names data clearinghouse, two of the most exciting components yet to be added are the digital photos and video clips of community members sharing their understandings of the place names. The purpose of the digital video and photographic images is to show seasonal change to emphasize the dynamic dimension of the place, people interacting with each place highlighting the fact that place is not ‘empty,’ but has cultural significance and is ‘used’ space. This project’s webpage design is still under construction at this time, but the following screen shot gives an idea of what it will look like (see Figure 12).



Hawaiian storied place names 185

This project provides many opportunities to serve the Hawaiian community and society as a whole. First and foremost, it will encourage families to participate in, or be present at, the digital video recordings, thereby (re)creating a Hawaiian cultural practice of interactive learning. Next and just as important, it will encourage storytellers to record at the actual place, so participants can literally see it while listening to the story and begin to introduce other senses into the interactive learning, like the sounds that make that place unique. Lastly, it will allow for multiple stories about a place to be shared, accommodating differing experiences of place. For example, women relate to a place differently from men, and fishermen from farmers. Yet all of these experiences are valid. By using modern technologies the HBGN is creating a shared knowledge space capable of representing a Hawaiian epistemology of place and space as it pertains to Hawaiian place names.

6.

Closing remarks

The fate of Hawaiian place names has come full circle. Hawaiian place names began as a repository of cultural knowledge in a place where interacting and experiencing the world intimately and sharing stories rhythmically was as normal as it is natural. Upon transitioning from a predominantly oral culture to a primarily literate culture, the meaning of Hawaiian place names that resonated in the telling of stories, recitation of chants, and dancing of hula was overlooked. Thankfully, ours is a culture that has not forgotten all of our connections to place and the importance of our place names. With the help of compassionate government agencies and the dedication of a few individuals, Hawaiian place names have once again become the “astonishing food of the land” for not just the “stone eaters of this land” but also the greater community.

References Cajete, Gregory. 1994. Look to the Mountain: An Ecology of Indigenous Education. Durango CO: Kivakī Press. Cajete, Gregory. 2000. Native Science: Natural Laws of Interdependence. Santa Fe NM: Clear Light Publishers. Connerton, Paul. 1989. How Societies Remember. Cambridge: CUP. DeSilva, Kīhei. 1993. The Song of Stones. Liner notes, CD by Kawai Cockett, ‘O ka ‘Ohau Ku‘u ‘Aina Nani. Hilo HI: Ho‘olokahi Productions and Nani Mau Records. Downs, Roger M. & Stea, David. 1977. Maps in Minds: Reflections on Cognitive Mapping. New York NY: Harper & Row. Fitzpatrick, Gary L. 1986. The Early Mapping of Hawai‘i. Honolulu HI: Editions Limited.

186 Renee Pualani Louis

Kanahele, E. L. H. n.d. An introduction to wahi pana: A Native point of view. The Journal of the University of Hawai‘i Community College. Moffat, Riley M. & Fitzpatrick, Gary L. 1995. Surveying the Māhele. Honolulu HI: Editions Limited. Oliveira, Katrina. 2006. Ke alanui kīke’eke’e o Maui: na wai ho‘i ka ‘ole o ke akamai, he alanui i ma‘a i ka hele ‘ia e o‘u mau mākua. Unpublished Ph.D. Dissertation, Department of Geography, University of Hawai‘i at Mānoa. Pukui, Mary K., Elbert, Samuel H. & Mo’okini, Esther. 1974. Place Names of Hawai‘i – Revised and Expanded Edition. Honolulu HI: University of Hawai‘i Press. Reeve, Roland. 1993. The place names of Kaho‘olawe: A study of those place names known from the island of Kaho‘olawe. Consultant Report No. 16, prepared for the Kaho‘olawe Island Conveyance Commission, Honolulu. Rodaway, Paul. 1994. Sensuous Geographies: Body, Sense, and Place. London: Routledge. Tuan, Yi-Fu. 2001. Space and Place: The Perspective of Experience. Minneapolis MN: University of Minnesota Press. Turnbull, David. 2003. Masons, Tricksters, and Cartographers: Comparative Studies in the Sociology of Scientific and Indigenous Knowledge. London: Routledge.

chapter 9

Between the trees and the tides Inuit ways of discriminating space  in a coastal and boreal landscape Scott A. Heyes

This chapter provides an account of how three generations of Inuit conceptualize the environment in a spatial sense, and explores the extent to which this is bound up with Inuit belief systems. The discussions on Inuit notions of space are based on fieldwork and interviews that have been carried out with the Inuit of Kangiqsualujjuaq, a coastal community in Nunavik, Northern Quebec. With the support of illustrations and a spatial lexicon, I present how these Inuit conveyed to me the ways in which they discern, describe and discriminate features within and upon the land and coastal environment.

1.

Introduction

The Arctic region that is inhabited by the Kangiqsualujjuamiut, the Kangiqsualujjuaq Inuit, is vast, mountainous and relatively isolated. The township is situated on the estuarine banks of the George River, about 16 km inland from Ungava Bay. Approximately 750 Inuit live in the community. Due to its geographical location, travel between Kangiqsualujjuaq and Southern Canada occurs only by plane or by ship. The town was established in the 1960s by the Canadian government as a permanent settlement for otherwise nomadic bands of Inuit. Initially, it consisted of a school, a timber mill, a cooperative building and a few houses (Arbess 1966;

. This chapter is principally based on research I carried out from 2002 to 2007 for my doctorate. Further research was undertaken at Melbourne University from 2007 to 2009, the Arctic Studies Center at the Smithsonian Institution during 2010, and at the Frost Centre for Indigenous Studies and Canadian Studies at Trent University in 2010. . The suffix miut means “people belonging to” and can be used in other contexts such as the Nunavimmiut, which means “the Inuit people of Nunavik.”

188 Scott A. Heyes

Figure 1.╇ Location of Kangiqsualujjuaq within the Quebec-Labrador Peninsula, Canada. The village is situated in a sheltered cove called Akilasakallak, which overlooks Kangirsualujjuap Kuunga (the George River). The popular hunting locations of Killiniq, Abloviak Fiord and Nachvak Fiord are north of Kangiqsualujjuaq.



Between the trees and the tides 189

Heyes et al. 2003). Today, the community consists of over 100 houses and has many modern conveniences, including the Internet, grocery stores, and athletic facilities. Prior to moving to Kangiqsualujjuaq to form a permanent settlement, the Inuit lived in small hunting camps that were scattered along the eastern shores of the Ungava Bay coast. These camps ranged in distances of up to 150 km north and south of Kangiqsualujjuaq, extending from Kuujjuaq to Killiniq (Figure 1). Beyond the sheltered cove of the settlement, the land has remained relatively unchanged in the face of a growing community. The northern extent of the tree line (naapaatuit isuaat) wraps itself around the mountains of the township (the extent of the township is spatially referred to as nunalik), clinging to life in only a few areas were a microclimate exists to sustain it. A few kilometers north of the township, where service roads lead to popular fishing lakes, the land is elevated and windswept. It is along these ridges that the tree line abruptly disappears to make way for the tundra. While the Kangiqsualujjuamiut are one of the few Inuit groups that have an association with a woodland environment (napaark) in the Eastern Canadian Arctic, they are also amongst the few Inuit groups that have lived an existence based on the extreme variability of the tides. The coastline and estuarine areas in which they hunt and fish principally for seal and char are subject to some of the most extreme tidal fluctuations in the world, similar to those in the Bay of Fundy in Nova Scotia. Tidal fluctuations of up to 16 meters have been recorded in the George River (Hubbard 1906:â•›536; NGIA 2005). In summertime, these tides cause tremendous rips and eddies, making boating and landing a dangerous undertaking. In spring and winter, the tides shift giant boulders around within the intertidal zone as if they were pebbles. The ice that forms along the sea edge during winter is constantly broken off and reconfigured with the passing of each tide. A different picture of the coastline is presented on a daily basis. The Kangiqsualujjuamiut have gained a sustained appreciation and knowledge of this coastline over more than 1000 years (McGhee 1984) through

. With the support of the Canadian government, the Inuit from the Quebec-Labrador Peninsula started constructing the village of Kangiqsualujjuaq in 1962 and permanently settled there in 1964. Arbess (1966) indicates that the creation of this centralized community, while initiated by the government, was overwhelmingly supported by the local Inuit. . There are many Inuit societies with a woodland environment in Alaska and Southern Labrador. With the exception of Kuujjuaq and Tasiujaq, the neighboring Nunavik and Nunavut Inuit communities to the North and West of Kangiqsualujjuaq are well above the tree line.

190 Scott A. Heyes

dwelling, hunting and storytelling. This knowledge of the land has led the Kangiqsualujjuamiut to develop nuanced and sophisticated ways of naming and describing the spatial dimensions of their maritime landscape. These descriptions and names reflect the presence of coastal phenomena, seasonal variations, microclimates, and in some cases legendary figures and non-human entities. Although the Kangiqsualujjuamiut have ways to describe and discriminate the tundra and woodland environments, this chapter draws largely upon research that focused on Kangiqsualujjuamiut conceptions of the coastal realm. Research was concentrated on this setting, in a large part, because this space was observed in the field and recorded in oral histories as being more frequently occupied and discussed amongst the Kangiqsualujjuamiut than were the woodland and tundra environments.

2.

Notions of space in a landscape of variability An empty room still has the dimensionality of space to contain something.  (I. Rice Pereira 1956:â•›46)

Those individuals who have never been to the littoral and tundra regions of Kangiqsualujjuaq might, at first glance, describe the area as characterized by “an endless horizon,” “a never ending vista” or a “barren landscape”. To the uninitiated, the absence of distinctive geographic features or purposely built features on the landscape is likely to generate a feeling that the land is devoid of boundaries, edges, landmarks and place names, some of the many tools of orientation that non-Inuit use to construct and apprehend space (Lynch 1960). The metaphor of an “empty room”, which appears in the opening quote to this section and which might also be used by the uninitiated to describe Kangiqsualujjuaq, is introduced here as a way to discuss the ways in which the spatial characteristics of the Arctic might be conceived from a non-Inuit perspective. While the purpose is not to use the construct of an empty room to make direct comparisons between how Inuit and non-Inuit conceptualize Arctic space, this metaphor provides a useful starting point to explore the ways in which conceptions of space are framed by knowledge, understandings, interactions and connections with familiar places. The empty room construct, when applied to the Kangiqsualujjuaq setting and considered from a non-Inuit perspective, would suggest that the viewer is looking

. I refer to land in a broad geographical sense. In the context of the Canadian Arctic I use the term “land,” regardless of whether it is snow/ice free or or snow/ice covered, as encompassing the regions of the interior, the coastline and the near shore environment.



Between the trees and the tides 191

at the space using a set of cognitive and navigational devices that differ from Inuit ways of knowing and seeing this space. With no trees, buildings, roads, power lines, fences, signs, or lighthouses, a newcomer to the Arctic may feel as though he or she is stranded in an “empty room”, whose walls, while imperceptible, are precipitous and distant. Conversely, the Kangiqsualujjuamiut would argue that the expansive Arctic countryside and seaside are composed of a number of landscape and seascape units, much like a jig-saw puzzle, some of which have perceptible edges and others of which have intangible boundaries. In this landscape and seascape of variability, where tides, currents, ice, wind and snow are constantly reshaping the surroundings, the Kangiqsualujjuamiut have developed remarkable ways of constructing and discerning space. This is a product of a number of knowledge systems that they have developed and refined over many generations of living on the land for survival. These knowledge systems include: navigational and orientation methods that have been formulated on the basis of star positions, wind directions, and the angle of the sun and the moon in the sky (MacDonald 2000); traditional place names and travel routes (BriceÂ�Bennett 1977; Makivik 1985; Müller-Wille 1987; Müller-Wille & Weber 1983); terms for distinguishing geographical features and phenomena (Kohlmeister & Kmoch 1814; Schneider 1985; Saladin d’Anglure 1964); and purpose-built markers (Hallendy 2000; Heyes 2002; Lewis 1966). The land and the sea are not “empty rooms” to the Kangiqsualujjuamiut. The landforms and waters contain their creation stories, history, myths and ancestral legends (Anon 1833; Turner 1894; Hawkes 1916; Hantzsch 1932; McLean 1849; Heyes & Jacobs 2008). Cosmological ways of knowing the land and waters, which are not possessed by an uninitiated visitor, allow the Kangiqsualujjuamiut to communicate information about the environment among each other and provide for them a way to anchor themselves in their surroundings. More than containing the dimensionality of “something”, the Kangiqsualujjuaq surrounds are populated with imprints of human habitation, histories and stories.

2.1

Cosmological notions of sea space

Inuit oral histories, and the records of early explorers, missionaries and ethnographers that were associated with the Ungava Peninsula, indicate that the Kangiqsualujjuamiut had animistic and reciprocal belief systems that likely had a bearing on their conceptions of space. As the ethnographer Turner (1894:â•›194) . The term “units” in this context refers to generic parcels of space, as opposed to units of mathematically-oriented measurement.

192 Scott A. Heyes

observed of Ungava Inuit spiritual relations with the sea, “Every cove of the seashore, every point, island and prominent rock has a guardian spirit.” On the notion of reciprocity and spirits, an account by Anonymous (1833 para. 11) indicates that Kangiqsualujjuamiut cosmology was a product of the actions of a land-dwelling female spirit, Supperguksoak, and a sea-dwelling male spirit, Torngak. It is said that each of these spirits reigned over the animals within their kingdoms (ibid.). These cosmological accounts of spiritual agents operating in separate domains suggest that the Kangiqsualujjuamiut historically regarded the land and sea as finite spatial entities. Their cosmologically grounded beliefs of spatial regimes may have, by extension, been embodied in a physical reality. The notion that tangible concepts of discriminating space may have originated from Kangiqsualujjuamiut cosmology is further supported by an account of how the land and sea were associated with the afterlife: Those that lived a good life proceeded to the deep sea, where “free from care and toil, they fare sumptuously on raw flesh and blubber, in secula seculorum. The wicked … are condemned to take up their abode in a ‘sea of troubles’, where none of the delicacies enjoyed by the blessed are to be found; and even the commonest necessaries are procured with endless toil, and pain, and disappointment” (McLean 1849:â•›137). Like Anonymous (1833 para. 11), Kangiqsualujjuamiut informants indicated to McLean (ibid.) that the sea was also the dwelling place of the Tomakhs spirit or “dead men” (ibid.) and that they were able to “… indulge in the pleasures of the chase on their old element [land], whenever they please; and are often heard calling to each other while in pursuit of the deer” (ibid.). In addition to suggesting a spatial distinction between the land and sea, these descriptions suggest that when animism was common among the Kangiqsualujjuamiut, the sea itself was appropriated into three spaces: one for the living, one for the dead, and one for the spirits. Although these demarcations of sea space may no longer be conceptualized or internalized by the Kangiqsualujjuamiut, these accounts suggest that previous generations had cosmologically grounded ways of thinking about the sea as a series of spaces and that there may have been a time when these spaces were associated with coastal nomenclature and place names. Writing on Kangiqsualujjuamiut belief systems and toponymy before Christianity took hold in the 1800s, Turner (1894:â•›194) noted that “Every rise of land, every curve of stream, every cove in the seashore has a name descriptive of something connected to it, and the names are known to all who have occasion to visit the place.”

. Tomakhs presumably refers to the Tuurngait spirit to which the Kangiqsualujjuamiut refer today.



Between the trees and the tides 193

With this background in cosmologically grounded ways of thinking of space, I expand the discussion in the following sections to include an account of current ways in which the Kangiqsualujjuamiut spatially discern, describe and discriminate the land and coastal environment. The account is supported by Tables 1 and 2, which provide a lexicon of spatially-laden terms that the Kangiqsualujjuamiut use to describe horizontal and vertical features of the land and sea, respectively. Many of the Inuktitut terms that feature in this chapter are described in more detail within these tables. A set of illustrations of the coast has also been created to support the tables. Appearing as Figures 2–8, these illustrations resemble the geography of the Kangiqsualujjuaq coastline near the community. They should be regarded as representative only, as the illustrations condense features of the coast that naturally occur over a space of tens of kilometers, such as sea ice. Where possible, I augment the descriptions of the land and sea with Kangiqsualujjuamiut stories about specific geographical features.

2.2

Generational perspectives on nomenclature

The ensuing discussion is based upon the results of fieldwork conducted in Kangiqsualujjuaq and its surrounding areas from 2003–2010. It is based on information obtained from 34 Inuit men and women, belonging to six different family units, across three generations, in the community of Kangiqsualujjuaq. The three generations that constituted the cohort were formulated on the basis of age brackets. These were defined as the elderly (≥ 55), middle-aged (31–54) and youth (≤€30). The ideas presented in this chapter are drawn largely from information conveyed to the author by the elders and middle generations through interviews, participant observation and drawing exercises. Much of the information was conveyed either by active hunters as they recalled their experiences on the land, or on hunting excursions. The findings relating to the generational component of the study are documented in Tables 1 and 2 and augmented in Figures 2–8. By interviewing male and female residents across each of the three age brackets, the nomenclature project was designed to account for terms that might be gender- or age-restricted. Drawing also from a cohort of six family units, the project sought to account for any variations in nomenclature knowledge that might be related to the ancestral provenance of the speakers.

. Figures 2–8 were generated using information collected through interviews with the study participants and by illustrations of the coast that they provided as part of drawing exercises. Adobe Illustrator CS4 was used to set out these figures.

194 Scott A. Heyes

2.3

Nomenclature conventions and interpretations

The Inuktitut orthography and definition of spatial terms that appear forthwith are those described to me by the Kangiqsualujjuamiut study participants. These were retained to reveal any individual or generational disparities regarding the use and definition of terms for the same feature. Where applicable, the orthography of sea ice terms that were recorded in the most regionally-relevant Inuktitut dictionary, Ulirnaisigutit (Schneider 1985), appears in Tables 1 and 2 for comparative purposes. The terms provided by the study participants were not adjusted to conform to the orthography and definitions of Ulirnaisigutit for several reasons: (1) Inuktitut spelling conventions are far from standardized: the syllabary continues to lack the sophistication to capture all the sounds and subtleties of Inuktitut, which reflects the fact that Inuktitut remains primarily a spoken language; (2)€the experts who were consulted in the making of the Ulirnaisigutit dictionary were from different landscape and seascape settings to the one experienced in Kangiqsualujjuaq, which may suggest a regional bias of the recorded terms; and (3) Ulirnaisigutit was broad in scope – when it was devised it did not specifically focus on identifying terms relating to Inuit cognition of the environment. The absence of a larger vocabulary of environmental features in Ulirnaisigutit should not be regarded as a failing of this work. It is acknowledged that it was intended to be a broad reference, and not a lexicon of Inuit environmental terms. Accordingly, this chapter provides information about the spatial terms beyond that found in Ulirnaisigutit. I chose not to “standardize” the orthography provided by the study participants against the Ulirnaisigutit orthography because it is important to recognize the changing nature of the language and the growing use of different, yet related terms to describe the same spatial feature in many instances.10 Having discussed Inuktitut nomenclature conventions, I wish to address the way in which the orthography and definitions of terms were recorded during field and interview processes. As a novice of Inuktitut, I relied on the expert advice of local Inuktitut translators to transcribe the spelling and definitions of the sea ice terms in situations when participants did not commit the spelling and definitions of terms to paper. Recognizing that the translators themselves might adjust the . When comparing the participants’ terms with Schneider’s, it is important to point out that many terms identified by the study participants seem to involve distinctions between single (short) and double (long) vowels, as well as a distinction between q/k/t (singular/dual/plural) endings, and the omission of fricatives r and g before consonants. 10. The identification of multiple terms might be a corollary of the study participants having grown up in regions other than Kangiqsualujjuaq, where different ways of knowing and describing the land and sea might apply.



Between the trees and the tides 195

terms that were conveyed by participants to conform to their own understandings of the terms, I explained to them that the nature of the project was to record the terms verbatim. Much trust was placed in the hands of the translators to explain (to the non-English speaking study participants) what was being asked of the participants. Having worked with these translators on several occasions prior, I was confident of their ability to translate and transcribe the material in accordance with the project. To circumvent linguistic and cognitive misunderstandings between the translator, participant and myself, the project largely depended, however, on visual aids to elicit spatial nomenclature. This was a particularly effective method, as the location and description of spatial terms were often pointed out directly or drawn on the visual aids. These aids consisted of 11” × 17” templates that were composed of section, plan and perspective drawings of coastlines at various seasons. Upon these, the participants drew, described and labeled various phenomena, features and geographic units of the seashore environment.11 The exercise supposed that those who could describe more coastal features likely possessed a greater knowledge base of the land-water interface than did those who identified fewer phenomena. The names of coastal phenomena provided by each participant were categorized according to whether the nomenclature referred to tides, water, ice, snow, coastal forms, edges and regions, atmospheric conditions, vegetation, activities, animals, or vertical elements of the coast. Comparison of the naming conventions among the cohort was facilitated by tabulating the nomenclature data in a database, which enabled the data to be sorted by age, gender, family unit, or generation. With the methodology in mind, the descriptions of the spatial terms discussed in this chapter should be regarded as my interpretations. These have been formulated through: my interaction of the land and sea ice environment with the study participants; readings of oral histories and literature pertaining to Inuit knowledge of the land and sea; the oral and visual descriptions of these settings that have been conveyed to me by the participants through translators; and information conveyed to me by English-speaking participants.

11. I was aware that a pre-determined template might influence the way in which participants approach the drawing exercise, and that it might influence the layout, style, composition and content of the drawings. A pilot study, using a blank piece of paper, proved less successful than the template. To limit misunderstandings of the templates, a local translator was present to explain the nature of the templates in Inuktitut. Owing to the visual nature of the templates, the aim of the exercise was generally well understood by all participants. Cultural difficulties on comprehending the templates were not encountered. For further details of how the template was designed and considered, as well as the way the content was analyzed, refer to Heyes€(2007a).

196 Scott A. Heyes

3.

Sea ice, sea space

Based on the multiplicity of evidentiary types, it appears that the Kangiqsualujjuamiut maintain a spatial picture of the coastal zone that transforms with the seasons. The units used by them to describe sea space and the land adjoining the coast differ when the surrounds are frozen and when they are ice free. Taking into account the transitional periods of ice development and deterioration that occur in autumn and spring respectively, the Kangiqsualujjuamiut have effectively developed a system of discriminating geographic units within a land-water interface zone that is highly porous and variable. In addition to having two broad sets of horizontal units to describe sea space, one in summer and one in winter (Table 1), they have a set of terms to describe vertical units of sea space (Table 2) that also shrinks and expands in accordance with seasonal ice conditions. This stratified world of ice that constantly decays and accrues might appear as a featureless white blanket to the uninitiated. However, for the Kangiqsualujjuamiut who participated in this study, when they see ice, they also see different spaces. When questioned about the distinction between land and sea,12 the study participants consistently maintained that the land and sea were viewed as two distinct geographical entities. As discussed earlier, this distinction has roots in cosmological appreciations of the environment. Building on this, the Kangiqsualujjuamiut have stories of non-human entities that are associated with the sea but not the land. The feared Mitilik, a feathered creature with transformative powers, is an example of a non-human entity that occupied only the maritime environment. Stories tell of how hunters were once chased by a group of Mitilik on the sea ice, as a consequence of a hunter killing a Mitilik who had been disguised as a seal (Heyes 2007b). To escape their wrath, the hunters made their way towards land, for it is said that the Mitilik do not venture beyond the sea ice (ibid.). Other non-human entities associated with sea ice include: Sillegiksartoj, a male spirit that dwells upon the ice-field and regulates the weather (Schultz 1895:â•›128); the Tuurngaq (pl. Tuurngait), a spirit that hunts on the ice in a similar fashion to the Kangiqsualujjuamiut; the Nanukulluk, a marauding polar bear of giant proportions that creates waves by pounding its feet upon the sea ice; and Sikuliasiujuitu,13 a giant and legendary figure whose name means “unable to go on thin ice” (Heyes 2007b).

12. I do not wish to suggest here that the distinction be regarded as one of binary opposites. Further study is required to explore this idea; see Tuan (1974). 13. The root siku- refers to ice.



Between the trees and the tides 197

Given the number of stories on non-human entities and spatial domains, it might be argued that these formed the foundations from which sprung similar ways of knowing and seeing other spatial elements of the land and the sea. With many of the old stories no longer being actively transmitted among the Kangiqsualujjuamiut, it remains difficult to explore this idea in more detail. However, the responses from the study participants on spatial nomenclature of the land and sea strongly suggest that – in the absence of old stories – they still do not conflate the biomes of land and sea. When asked if they perceived the sea ice as an extension of the land at any point during winter or spring, they emphasized that it was not. Sea ice is tuvaq14 and the land is nunak. Drawing a distinction between land and sea when both domains are covered in snow and ice may be difficult for many non-Inuit. This is because the point of physical separation between land and sea is difficult to distinguish. The distinction between land and sea is also exemplified in the terminology used by the Kangiqsualujjuamiut to describe certain animal behavior. For instance, a sea mammal (especially a ringed seal) that is basking or resting on the surface of the ice is referred to as uttuq,15 whereas a sea mammal resting on the ice free shore would be referred to as siluk. If a Kangiqsualujjuamiut hunter were to hear either of these terms in conversation, he or she would immediately recognize the geographical whereabouts of the sea mammal. While land and sea are regarded as distinct spaces at a larger scale, we observe that both of these realms are broken into smaller spaces if we zoom in to each space at different times of the year. For example, the shoreline component of the sea ice realm is regarded by the Kangiqsualujjuamiut as another distinct geographical unit. In Inuktitut this unit is known as sikuutsuni tuvaq,16 which means “ice laden shore”. Within this space there are a number of sea ice features that possess specific names, as shown in Table 1 and illustrated in Figure 2 and partially in Figure 3. These include: qarvitinniq, which is the ice upon which Inuit travel by snowmobile when moving along the coast, and qainnguq, which is solid ice stuck to the shore. The shore, in these wintertime examples, is understood through specific qualities or physical characteristics that distinguish it from its immediate surrounds. 14. The Kangiqsualujjuamiut generally use tuvaq to refer to the sea ice surface, although it should be noted that in other Arctic communities this term refers to ice on a surface, whether the sea or a lake. When referring to frozen lakes, the Kangiqsualujjuamiut typically use the term sikuk, the generic word for ice. 15. Uttuq means literally “it warms itself ”. The reference to basking sea mammals is a metaphorical extension. 16. This term contains both “ice” roots siku- and tuva-.

Seal breathing hole

Big open hole in sea ice Edge of snowy area

kiggulik (Sch)

puturlak (Sch)

aput (snow) (Sch)

akluk 1 var. alluk; aalluk3 alt. puijiup anirtirivinga1; nulataq6; natsiup aklunga2; amuvinga kikkulik1

allutinni3 alt. kilaluk2

aputik killirut3

2

3

4

Ice crack that opens and closes with the tides

aajuraq (Sch) aajuraq1 2 3 4 5 6 var. aajurak4 5; aajuraq3 4 5 alt. qunnik1; qunniq3;  qupisimajuk5; siku qupisimajuk6

English equivalent description (by Inuit participants)

Edge

Hole/area

Hole/area

Hole/line

Land

Frozen sea

Frozen sea

Frozen sea

Geographical/ Spatial element/unit seasonal association (referent)

Note: The terms in this table support the illustrations in Figures 2–4 and Figures 7–8. Reference numbers have been provided to locate the terms on the figures. The term in bold typeface is either the term identified by more than one participant or the term provided by the most senior generation (where multiple terms and variations were provided by more than one generation). The original spelling of the terms used by the hunters has been retained. The author is aware that such spellings and definitions may not necessarily be consistent with those in Inuktitut dictionaries and those in common usage in communities other than Kangiqsualujjuaq.

1

Closest Inuktitut orthography equivalent and/or term from Ungava region (not necessarily from Kangiqsualujjuaq)

Gen A Male 55+ Gen A Female 55+ Gen B Male 31–54 Gen B Female 31–54 Gen C Male 0–30 Gen C Female 0–30 Variant, where applicable Alternative term, where applicable Equivalent term(s) from L. Schneider (1985) Equivalent term(s) from Kohlmeister & Kmoch (1814) Not recorded

Ref Inuktitut term # (by Inuit participants)

KEY 1 2 3 4 5 6 var alt Sch KK NR

Table 1.╇ Spatially-laden Inuktitut terms for land, sea ice and ice free features along the horizontal plane that appear in Figures 2–4  and Figures 7–8

198 Scott A. Heyes

Shore, frozen sea

Frozen sea

Edge Area

The lowest extent of the receding tide; the outer edge of the tinninik region on an ebb tide That which becomes an island when the tide recedes Edge of rough land Transitional space between ittiniq and tuvaq

Rough ice beyond the shore (seaward) that continues to Area build up from tidal movements. Interface between land and sea-ice Area Thin layer of ice that forms along sinaa overnight or during extremely cold weather; dangerous surface. Also forms against blocks of floating ice. Also ice which cracks in springtime Sea ice that piles up against the shore End of curved shore Cove, short inlet or long Fiord Long inlet Line of tidal water in its transitional phase

tinistsariirpuq (tide has finished going out) (Sch)

qikiqtaujaq (Sch)

NR

NR

ittiniq (Sch)

ivujut; sikuliaq; qinnuaq (Sch)

ivuniit (Sch) ilumunnatuq (Sch)

qamngaliqpuq (Sch)

kangirsuk (Sch)

NR

NR

10 iqqaauk3

11 ittini sinaanga2

12 ittinikitaa2

13 ittiniq1 2 3 4 5 6 var. ittiniit2 3 4 6; ittinik5;  ittinnik5; ittinniq3. alt. qaanuuk3

14 ivujuk1 6 var. ivujuq1 alt. qanittaq1; tuvairpuq2

15 ivuniq1

16 kangirsup qingua1

17 kangisuk1 3

18 kangisukutaaq1

19 kasak3 alt. katsutuk1; tiniktuk1

ingirranikisaup ungummuavinga2

9

Water’s edge

Edge

Area

Area

Edge

Surface/ substance

Area

Edge

Edge

Land

Shore

Frozen or ice free

Shore, frozen sea or ice free sea

Shore

Shore, frozen sea

Shore, frozen sea

Land

Frozen or ice free

Shore

Land/Shore

Offshore

imiq (water) (Sch)

imaup killanga1 alt. imarq killirut3;  imaup sinaanga1; sikuituk5

8

Area Area

Small mountain

Shore

Offshore area; distant seas; the sea area when land  disappears from the horizon; also first tide that reaches the shore after the sea ice has disappeared in spring

imarpik1

7

Edge

imaaluk (Sch)

ikpiapik1 2

6

High tide edge

ulitsariirpuq (the tide has finished rising) (Sch)

qarqaq (hill) (Sch)

igirnikisuk1 alt. ulitsariituk1 3

5

Between the trees and the tides 199

Area

Sand bars Edge of wooded area; e.g. where trees meet the tundra Woodland; stands of trees, esp. referring to trees along the George River corridor Round mountain Horizon Land

Extent of built-up area of community

Solid sea ice stuck to the shore. Inuit travel on this surface by snowmobile Flat rocks Flat rocks on shore

NR

napaartulik (Sch) nappartolik (KK)

NR

NR

nuna (Sch)

nunalik (Sch)

Point uivvaq (Sch); tikkerarsuk (a low point reaching into the sea) (KK)

Land near water

siuraq (sand) (Sch)

NR

NR

qainnguq (Sch)

maniraq (flat ice, soil, ground) (Sch)

maniraq (flat ice, soil, ground) (Sch)

24 marqaulu siuraulu akunninga1

25 naapaatuit isuaat1

26 napaark1

27 niaqunnakallaq1

29 nunak1 2 3 4 5 6 var. nuna1; nunaq1 2 3 5 6 alt. nunammarik3

30 nunalik1

31 nuvuk1 2 3 var. nunuk4

32 parniq1 2

33 qaingngu2

35 qairtuq1

36 qairtuq1 alt. qainguk3

34

qainnguq3

qaningituk1 3

An area of ice that consists of lines or bands of ice with open water between them

Point

Muddy area

maujaq (Sch)

23 marqaq1 2 4 var. marqak1 3 5 6

28 nunak alt. qapiarnituk1

Edge of muddy area

NR

22 marqaq killigu3 alt. tinniniup killinga1

Surface

Surface

Surface/ substance

Area

Area

Surface

Area

Area

Area

Edge

Area

Area

Edge

Area

Flat land

maniraq (flat ice, soil, ground) (Sch)

21 maniraqtuq1

Surface

Rough land with ice

NR

20 maniittuit1

Shore

Land/Shore

Shore, frozen sea

Land

Offshore; frozen sea or ice free sea

Frozen or ice free

Land

Land

Frozen sea or ice free sea

Land

Land

Land

Shore

Shore

Shore

Land

Land/shore/frozen

200 Scott A. Heyes

Big rocks on shore Island Rocky areas on shore

NR

qikirtaq (Sch) kikkertorsoak (KK)

ujaralaanguniq (rocky) (Sch)

40 qiguinaaluk1

41 qikittak1 alt. qikirtaq1; qikirtakallak2

42 qiquinaaluk1

Where ice area finishes

NR

46 sikuk killirut3

Sea edge Ice edge

kit (Sch)

NR

50 sinaa tariup1

51

Outer edge of tuvaq, where land fast ice meets open water

sinaaq (Sch)

49 sinaa1 2 4 var. sinaak4 5 alt. tuvaup sinaanga1; imaup sinaanga1

sinaanga2 3

Sea mammal that basks or rests on ice-free shore

silu (an animal found dead on  the shore or inland) (Sch)

48 siluk1

Ice laden shore

Ice, esp. freshwater ice

sikuk; siku; tuvaq (Sch)

45 sikuk1 2 3 4 5 6 var. nilaq1

ittiniviniit (Sch) 47 sikuutsuni tuvaq1 alt. tinujaumajuk1; tinutsimajuk1; kutjutuq1

Flat shore

maniraq (flat ice, soil, ground) (Sch)

44 sanirajak1

qumniq (Sch) 43 qunniq1 2 3 var. qunnik1 alt. qupisimajuk1; qupiniq killali2

Area

Area

Surface

Surface/ substance

Area

Area

Edge

Edge

Edge

Surface/ substance

Surface/ substance

Edge

Surface/ substance

Area

Ice crack that shows water; remains open during change Hole/line of tides. This term can also be used to describe a big crack in a rock face on a mountain

Sea ice that clings/attaches to shore. Inuit travel on this surface by snowmobile. It only becomes dislodged by  the biggest of the highest tides in June

alt. qingngusaliusijuq1

qangatsaniq (overhanging block of ice) (Sch)

39

Mountain

qarqaq (hill) (Sch)

38 qarqak5

qarvitinniq1

Frozen inlet

NR

37 qangngurusirtuq1

Frozen sea

Shore

Shore, frozen sea

Shore, ice free sea

Shore, frozen sea

Shore, frozen sea

Frozen sea

Shore

Frozen sea

Shore

Shore

Land/Shore

Shore, frozen sea

Land

Shore, frozen sea

Between the trees and the tides 201

qinuit (Sch)

var. tariak1

Edge

Water line of a tide that has partially receded Exposed area of tidal flat at low tide

Tidal beach Gravel areas

tinippaliajuq (the tide is going out) (Sch)

tininniq (Sch)

NR

tuapaq (Sch)

60 tiniriartuk3

61 tinninik1 2 3 4 5 6 var. tinniniq1 2 3 4 5 6 alt. tinittasuuk5

62 tinniniup sitjanga2

63 tuupauq3

65 uttuq1 2

uuttuq; parnguliaq (Sch)

tuvaq (Sch) 64 tuvaq1 2 3 4 5 6 var. tuuvuq 3 alt. tariup sikunga1; sikuk tariuk2; tariuk sikunga3

Sea mammal that basks or rests on ice surface; esp. ringed seal

Sea ice; general term. Often used to describe ice on  the seaward side of the ittiniq zone. Flat ocean, when frozen over

Edge

tiningajuq (the tide has gone out); Low tide edge; general term tininniq (Sch)

59 tininniup isua1 3 4 alt. tinisarinik1 3; tinsavirniq3; tininniup killinga1

Surface/ substance

Surface/ substance

Area

Area

Area

Surface

Edge

Area

Lake edge

57 tasiup

sinaanga1 Rocks

Lake

tasiq (Sch)

Area

ujaralaanguniq (rocky) (Sch)

Sea water; salt water

tariuq (Sch)

Surface

Area

Edge

NR

Sand

siuralik (Sch)

Beach; can consist of rock, sand or mud

Thin layer of ice that sometimes forms along the edges of seal holes during extremely cold weather; dangerous surface. Qunniq is also used to describe a crack

58 tininip ujarangit1 alt. ujarait3

56 tasiq1 2 3 4 5 6

55

tariuk1 2 3 4 5 6

54 siuraq1 2 5

sitjaq (Sch) 53 sitjak1 2 3 4 5 6 var. sitsak1; sitjuk1 alt. imaup sitjanga1; iimarsualuk5

52 siniritsijuk1 alt. qunniq2

Frozen sea

Frozen sea

Frozen or ice free

Shore

Frozen or ice free

Shore

Shore

Land/shore

Land

Land

Ice free sea

Land/shore

Shore, frozen sea or ice free sea

Frozen Sea

202 Scott A. Heyes

The shaded boxes indicate the terms used by lnuit hunters according to age and gender.

Legend:

39

Male Female

37

18

32 Parniq Inuktitut term Reference

Generation A age 55+ Generation B age 31-54 Generation C age 0-30

34

Imarpik

34 Qainngug

32 Parniq

20 Maniittuit

18 Kangisukutasq

7

Table 1. Horizontal Units

51 Sinaanga

40 Qiguinssluk

39 Quarvitinnig

37 Quangngurusirtuq

51

40

32

7

Figure 2.╇ Inuktitut nomenclature for horizontal units of the sea ice realm with an emphasis on the near shore components.

© Dr Scott Heyes, 2010 Graphics by Dr Heyes & Anna Sundman

Inuktitut no menclature for horizontal units of the sea ice realm with an emphasis on the near shore components.

Figure 2

20

Between the trees and the tides 203

12

11

47

The shaded boxes indicate the terms Male used by lnuit hunters according Female to age and gender.

Legend:

36

14 Ittiniq Inuktitut term Reference

Generation A age 55+ Generation B age 31-54 Generation C age 0-30

50

13 Ittiniq 14 Ivujuk 43 Qunniq 47 Sikunutsunituvaq 49 Sinaa

3 Allutinni 11 ttini sinaanga 12 ttinikitaa

2

2 Akkuk

43

1 Aajuraq

Table 1. Horizontal Units

1 14

52

65 Uttuq

64 Tuvaq

55 Tarluk

52 Sinintsijuk

50 Sinaa tariup

3

64 65

55

36 Quairtuq

49

Figure 3.╇ Inuktitut nomenclature for horizontal units of the land and sea ice realm with an emphasis on the most seaward components.

© Dr Scott Heyes, 2010 Graphics by Dr Heyes & Anna Sundman Hand drawn illustrations by Kangiqsualujjuamiut hunters

Figure 3 Inuktitut nomenclature for horizontal units of the land and sea ice realm with an emphasis on the most seaward components.

13

204 Scott A. Heyes



Between the trees and the tides 205

At the larger scale, the space occupied by the ice laden shore is where the Kangiqsualujjuamiut consider the land and sea boundaries to dissolve. To illustrate this concept, and to describe some of the horizontal seascape units that are conceptualized by the Kangiqsualujjuamiut during wintertime, it might be helpful to chart the spaces that would be encountered by a hunter as he or she travelled from the shore onto the sea ice. Being introduced to the seascape units in this manner will allow the reader to gain a contextual appreciation for the way in which an Inuk hunter might perceive the spaces through which he or she moves. In my experiences with the Kangiqsualujjuamiut, they have explained to me that they perceive space as they travel through it; that is to say, space “unravels” as they move across the land and sea. The section below presents a description of the spatial components of the sea ice along a horizontal plane, as they would appear and unfold through the eyes of an Inuk hunter on an outing over the land and sea.

3.1

Horizontal units of the sea ice

Although we could begin our journey with the hunter from any point in the land or seascape, for the purposes of this exercise, we will begin on the nunak, moving seaward, during the winter months. The hunter begins by moving across the snow-covered land and approaches the coast. The act of passing from nunak to sea ice (tuvaq) is known as tuvalirtuk.17 It involves traveling from snow on the land across a gap onto the ice. The land side of this gap – what might be best described in English as a cliff – is exposed to the rising and falling water levels of the tides. As water levels rise and fall, ice forms on the “cliff ” face at various points. The most frequently used descriptor for the ice that develops vertically along the cliff edge is qainnguq. After traversing this ice phenomenon, the hunter might arrive at sitjak (a beach) or, in some instances, qangngurusirtuq (a frozen inlet). As illustrated in Figures 2–4 and described in Table 1, these mark the first of several seascape units that extend beyond the shore. Moving seaward, the hunter might confront an area of rough ice known as ittiniq. This is a finite ribbon of undulating ice that changes with the rhythms of the tides and runs parallel to the shore. In some places, it extends for several kilometers and rises over several meters. Canyons and bridges of ice form within the ittiniq zone (Figure 4). The transitional space between ittiniq and the smooth ice positioned beyond it, tuvaq, is referred to as ittinikitaa.

17. Note that tuvalirtuq can be parsed as tuva-lir-t-uq, which means “ice-towards”. This is a fairly generic formula which is not specialized to ice but can be used to mean “move toward” for any noun.

9

3

1

8

The shaded boxes indicate the terms used by lnuit hunters according to age and gender.

Legend:

Male Female

4

Aputik Inuktitut term Reference

Generation A age 55+ Generation B age 31-54 Generation C age 0-30 2

6

3 4

27 Niaqunnakallaq

2

1

26 Napaark

30 Nunalik

25 Naapaatuit

1

26

lkkatuq

lkaut ilua

Aputi sikuup qanganiituq Aputik

Table 2. Vertical Units 29 Nunak

5

30

6 ikpiapik

Table 1. Horizontal Units

2

27

29

9 lttisisuk

8 ltijuk

7 lrqaq

5 llumaniq

25

Figure 4.╇ Inuktitut nomenclature for horizontal and vertical units of the land and sea ice realm. Inset: an ice cave close to shore.

© Dr Scott Heyes, 2010 Graphics by Dr Heyes & Anna Sundman Hand drawn illustrations by Kangiqsualujjuamiut hunters

Inuktitut nomenclature for horizontal and vertical units of the land and sea ice realm.Inset:an ice cave close to shore

Figure 4

7

206 Scott A. Heyes



Between the trees and the tides 207

Once the hunter has reached tuvaq, which specifically refers to the area of flat or smooth sea ice on the seaward side of ittiniq, the length of the journey to the next seascape unit is entirely dependent upon the ice conditions. In the depths of winter, tuvaq may extend tens of kilometers; during the spring, it may extend only a few hundred meters. It is a seascape unit which expands and contracts, often at rapid speed. The outer edge of tuvaq is known as sinaa, the junction of land fast ice and open water. In English this is sometimes referred to as the floe edge. The sinaa is associated with a specific story, which was recounted by Kangiqsualujjuamiut elders who participated in this study. It is said that in ancestral times, the cries of “Lu-Lu-Lumauk” (my son) would be heard by hunters as they travelled toward the sinaa (Heyes 2007a:â•›220). These were the cries of Lumauk, a woman who was pulled into the water while harpooning a beluga whale and, as a consequence, separated from her blind son. The tether, which was wrapped around her body and which caused her to become entangled, is said to have eventually formed a vessel of life for the boy’s mother. It allowed her to transform into a half-beluga, half-human creature, which is referred to as Lumauk. It is said that she appears along the floe edge from time to time, lamenting her separation from her son (ibid.). Within the horizontal tuvaq zone, there are a number of other smaller seascape units that may be present and may be encountered by the hunter. As we have seen earlier in Figure 3, these include: allutinnni, a big hole in the ice where seals and walrus are often hunted; akluk, a seal breathing hole; aajuraq, an ice crack that opens and closes with the tides; and qunniq, a large crack that remains open. The exposed sea water in these cracks and holes is known as tariuk (sea water/salt water), and it is the same term used to describe the open water that meets the floe edge. During the evening, as temperatures drop, another layer of ice forms along the floe edge and the rims of seal holes. These dangerous areas, which disappear with the warmth of the day, are referred to as ivujuk with respect to the floe edge, and siniritsijuk with respect to seal holes. If the hunter were to travel by boat beyond the floe edge, he or she might pass another spatial unit known as parniq. Especially common in spring, parniq refers to lines or bands of ice with open water between them. Each band of ice is itself composed of smaller, broken pieces of ice (Figure 2). These bands of ice shift back and forth across tariuk in tandem with the tides; as the tide comes in, it pulls the bands with it; as the tide goes out, it takes the parniq with it. After travelling through parniq by boat, the hunter would then approach imarpik. This term is used to describe the offshore region and distant marine waters. Informants indicate that the imarpik region begins near the mouth of the George River estuary, or at the point at which the land disappears from view. The Kangiqsualujjuamiut informants recounted that imarpik is also a term used to describe the first tide

208 Scott A. Heyes

that reaches the shore after the sea ice has disappeared from the bay in springtime. In this sense, the hunter might say “Imarpik has come; it has arrived” (Heyes 2007b:â•›110). The hunter’s wintertime journey from nunak to the imarpik, as described above and shown in Table 1, might involve passing through up to 25 distinct spaces, depending on the ice conditions present. From a functional perspective, the attribution of conceptual units to sea ice formations has been generated by the Kangiqsualujjuamiut to communicate and derive information about travelling conditions, travel routes, the ferocity of the tides, the location of animals, and dangerous areas. At the time of the designation of these spatial formations, it would have been crucial for Kangiqsualujjuamiut to be aware of various geographical features if they were to become successful hunters, and if they were to avoid falling into the water.

3.2

Vertical units of the land-water interface

The vertical world concealed by sea ice is hidden, but it is certainly not imperceptible to the Kangiqsualujjuamiut. Whether frozen over or ice free, we see from Table 2 that they attribute at least 19 terms to the vertical dimensions of the land-water interface. A rich nomenclature of vertical dimensions of the sea likely developed over time to provide purposeful information about sea and ice conditions to seal hunters, net fishers and hand line fishers. The terms used to describe depth and features beneath the sea are illustrated in Figure 5. The sea floor is referred to as irqaq, and the earth beneath the sea floor, a realm that has never been seen or explored by the Kangiqsualujjuamiut, is known as ikaut ilua. In a sea environment that has a very dark bottom, where gauging depth is difficult, and for a people who do not swim or dive in the sea,18 it is noteworthy that the Kangiqsualujjuamiut have developed terms to describe depth. Shallow water is called ikkatuq, deep water is termed itijuk, and very deep water is described as ittisisuk.19 In order to describe some of the other vertical spatial components of the winter landscape perceived by Kangiqsualujjuamiut, the present section again adopts the perspective of an Inuk hunter. The hunter begins by approaching a seal breathing hole on the ice. Looking down the vertical column of the hole, like that illustrated in Figure 5, the hunter might see up to six layers of ice and snow, 18. The waters of Kangiqsualujjuaq are too cold to bathe in, even during the middle of summer. 19. Further investigation is required to establish what these notions of depth are based on.

7

Sea floor

aluq (Sch)

6

irqaq1 2 3 var. irqak2 3 5 6 alt. siuraup iqqaniq2; iqqana3; irqanga4

ilumaniq1

5

Uppermost layer of new freshwater ice

iiukittuq (Sch)

ikkatuq1 3

4

sikuliaq (Sch)

Shallow water

NR

ikaut ilua2

3

ipiutaq1 alt. kakiattusia2; siku sattuk2; sikuak3

Realm below the sea floor

NR

aputik4

2

An ice cave; an opening where one can go under  the ice to collect clams from the sea-floor1

Snowpack on ice

NR

aputi sikuup qanganiituq5

1

NR

Snow on ice

Closest Inuktitut orthography equivalent and/or term from Ungava region (not necessarily from Kangiqsualujjuaq)

English equivalent description (by Inuit participants)

Surface

Surface

Area

Area

Area

Surface

Surface

Frozen sea or ice free sea

Frozen sea

Frozen sea

Frozen sea or ice free sea

Frozen sea or ice free sea

Frozen sea

Frozen sea

Geographical/ Spatial element/unit seasonal association (referent)

Note: The terms in this table support the illustrations in Figures 4–6. Reference numbers have been provided to locate the terms on the figures. The term in bold typeface is either the term identified by more than one participant or the term provided by the most senior generation (where multiple terms and variations were provided by more than one generation). The original spelling of the terms used by the hunters has been retained. The author is aware that such spellings and definitions may not necessarily be consistent with those in Inuktitut dictionaries and those in common usage in communities other than Kangiqsualujjuaq.

Ref Inuktitut term # (by Inuit participants)

KEY 1 Gen A Male 55+ 2 Gen A Female 55+ 3 Gen B Male 31–54 4 Gen B Female 31–54 5 Gen C Male 0–30 6 Gen C Female 0–30 var Variant, where applicable alt Alternative term, where applicable Sch Equivalent term(s) from L. Schneider (1985) NR Not recorded

Table 2.╇ Spatially-laden Inuktitut terms for land, sea ice and ice free features along the vertical plane that appear in Figures 4–6

Between the trees and the tides 209

Surface

Area Area

Snow that forms atop sea ice; uppermost layer of snow

Old ice; lowest portion of freshwater ice; situated below ipiutaq

sikutsiarippuq; sikutsiarikpuq (fine, very transparent ice) (Sch)

qinnuaq (very thin ice when the sea Top layer of freshwater ice starts to freeze); sikuaq (the first very thin ice on water) (Sch)

To describe the thickness of ice

NR

sikuliaq (fresh or newly formed ice  Term used to describe the middle layer/level of  at sea) (Sch) sea-ice (general)

Term to describe the underneath portion/space/ edge of ice; the sea ice surface that meets the open sea

qaanniq (water that has rise above lake or river ice; water that can be heard trickling under snow or ice; water freed by the floating away of the loose parts of the pack ice) (Sch)

NR

NR

12 qarni sikumi2 alt. atsuilik1

13 qingngugutik1 alt. sikuliak3

14 qinnitarsiaq1

15 sikuliarq1 alt. sikuup qanga1 5

16 sikuu ataa1 alt. itjuniq1; tuvaup ataanga1 5; sikuup ataanga4

17 sikuup itjuninga1 2 alt: itjuninga sikuup2 Itjunik3

19 tuvaup qaanga1 5 var. tuvaup qanga1

NR

tuvaq (Sch) 18 tuvaq1 2 3 4 5 6 var. tuuvuq 3 alt. tariup sikunga1; sikuk tariuk2; tariuk sikunga 3

11

Upper surface of sea ice; top layer/level of sea-ice (general)

Sea ice; general term. Often used to describe ice on the seaward side of the ittiniq zone. Flat ocean,  when frozen over

Lowest portion of sea ice, the most bottom layer –  as seen through a crack or when a hole is made through the ice; often black and translucent

Area

Bottom layer of snow; snowpack base

pukak (Sch)

qaanik1

10 pukak2

Surface

Surface/ substance

Area

Surface

Surface

Surface

Area

Very deep water

ittinirsiutuq (looks for deeper  water) (Sch)

ittisisuk1 alt. irqanittuit3; itijualuk1

9

Area

Deep water

imaqpik (ocean); sitjarittuq (Sch)

itijuk1 3

8

Frozen sea

Frozen sea

Frozen sea; frozen lake

Frozen sea

Frozen sea; frozen lake

Frozen sea

Frozen sea

Frozen sea

Frozen sea

Frozen sea

Frozen sea or ice free sea

Frozen sea or ice free sea

210 Scott A. Heyes

8

Male Female

4

7 Irqaq Inuktitut term Reference

Generation A age 55+ Generation B age 31-54 Generation C age 0-30

13

18 Tuvaq 19 Tuvaup qaanga 14 Qinnitarsiaq 7 Irqaq

17 Sikuup itjuninga

16 Sikuu ataa 13 Qingngugutik

9 Ittisisuk

17

4 Ikkatuq

8 Itijuk

3 Ikaut ilua

19 14 16

1 Aputi sikuup qanganiituq

Table 2. Vertical Units

18

Figure 5.╇ Inuktitut nomenclature for vertical units of the sea ice realm showing water depths and seal breathing holes.

© Dr Scott Heyes, 2010 Graphics by Dr Heyes & Anna Sundman Hand drawn illustrations by Kangiqsualujjuamiut hunters

The shaded boxes indicate the terms used by lnuit hunters according to age and gender.

1

Legend:

3

Inuktitut nomenclature for vertical units of the sea ice realm showing water depths and seal breathing holes

9

Figure 5

7

Between the trees and the tides 211

9

3

1

8

The shaded boxes indicate the terms used by lnuit hunters according to age and gender.

Legend:

Male Female

4

10

32 Irqaq Inuktitut term Reference

Generation A age 55+ Generation B age 31-54 Generation C age 0-30

6

11 Qaanik 12 Quarni sikumi

7 Irqaq 8 Itijuk 9 Ittistsuk 10 Pukak

1 Aputi sikuup qangannituq 4 Ikkatuq 6 Iplutaq

12

3 Iksut ilua

11

Figure 6.╇ Inuktitut nomenclature for vertical components of freshwater ice and snow layers.

© Dr Scott Heyes, 2010 Graphics by Dr Heyes & Anna Sundman Hand drawn illustrations by Kangiqsualujjuamiut hunters

Figure 6 Inuktitut nomenclature for vertical components of freshwater ice and snow layers

7

212 Scott A. Heyes



Between the trees and the tides 213

depending on water clarity and the age of the ice, each bearing a different name. Looking downwards into the hole, the hunter would see: tuvaup qaanga (upper surfaces of tuvaq); qingngugutik (middle portion of the sea ice); and qinnitarsiaq (lowest layer of sea ice). The surface of the sea ice that meets the sea is known as sikuu ataa. Depending on wind and compaction, several layers of snow may have built up over tuvaq. Where snow has accrued in such situations, the hunter would describe the upper crust of snow that forms atop sea-ice as qaanik. The snow layer beneath the qaanik would be called pukak. This layer is the cleanest form of snow, and is the most suitable for making drinking and cooking water. In springtime, when freshwater ice (sikuk) breaks away from rivers inland, it is often thrown up against saltwater ice through tidal and wave action. This ice is distinguishable from other forms of ice by its vivid blue color. It occasionally appears crystalline in shape, with brittle vertical columns both above and below the water line. When it appears in this formation, it is known as illaujait. As with sea ice and snow, the Kangiqsualujjuamiut use certain terms to describe the dimensions of freshwater ice. The vertical components of freshwater ice, as well as some of the snow layers that accrue on freshwater ice when adrift, are shown in Figure 6. In this illustration, we observe that two layers of snow conceal ice formations that bear Inuktitut names. The uppermost layer of new freshwater ice is termed ipiutaq. The lower, older ice underneath is known as qarnisikumi. The hunter’s journey from the seal hole proceeds to a large crack in the ice that appears near the shore. In search of shellfish and seaweed, the hunter climbs down the crack to reach the sea bed beneath. This sea bed is only accessible at low tide. After passing through the crack, and scrambling down meters of ice, the hunter enters a dangerous cavity known as ilumaniq. The sea ice above these caves is supported by arches of ice, as appeared earlier in Figure 4. In this spectacular setting, the hunter is surrounded by a cathedral of ice that glows a tinge of blue as the light filters through the ice. Looking up from this vantage point, the hunter would observe the different layers of sea ice, which are defined by color, texture and density. The hunter must be aware of the incoming tide while gathering shellfish and seaweed in ilumaniq, as the open areas quickly fill with seawater. On returning to the surface, the hunter would pass through various layers of ice and snow that were also observed at the seal breathing hole. It is through this act of moving up and down the frozen platform of ice that a spatial consciousness of the vertical units of sea ice among the Kangiqsualujjuamiut becomes most apparent.

214 Scott A. Heyes

4.

Sea water, sea space

When the sea ice is swept away from the bay in early spring, the coastal environment is transformed into a new spatial regime. While conceptualizations of the sea ice environment are largely determined by the morphology of certain types of ice, the spatial regimes of the ice free coastal environment are largely determined by tidal action and surfaces. Resuming our journey with the Inuk hunter, we travel across the land-water interface in summer at low-tide. The hunter progressively passes over nunak, ikpiapik (small mountain), qaingngu (land near water), sitjak (beach) and then into tininnik, the tidal flats. The tinninik realm (Figure 7) is further broken down into a series of smaller spaces, which include: tinniniup sitjanga (tidal beach), qiquinaaluk (rocky areas on the shore), tuupauq (gravel areas), marqaq (muddy areas), iqqaauk (that which becomes an island when the tide recedes), and marqaulu siuraulu akunninga (sand bars). The tininnik region finishes at an edge called ingirranikisaup ungummuavinga, which marks the low tide mark. Since the tides are constantly changing, so too is the location of this edge and the breadth of the tininnik space. Like sinaa (floe edge), ingirranikisaup ungummuavinga is the point at which the area known as tariuk (sea) begins. The region known as imarpik, the furthermost space within the land-water interface spectrum, begins where the horizon disappears when looking landward. The conception of this space among the Kangiqsualujjuamiut is the same in summer and in winter. The existence of terms used by the Kangiqsualujjuamiut to describe edges of features within the coastal zone suggests that they have systematic and consistent ways of discerning sea space. In some cases, as with sinaa, these edges are obvious; but in other cases, as with imarpik, the edges are porous, blurred, or plastic, and are open to interpretation. As we see in Figures 7 and 8 for example, the Kangiqsualujjuamiut use several terms to describe the edges at which the tides rise and fall. The low tide edge is known as tininniup isua; the high tide edge is described as igirnikisuk; the water line of a tide that has partially receded is called tiniriartuk; and the line of tidal water in its transitional phase is known as kasak.20 Any of these edges might potentially define the point at which tinninik transforms into tariuk.

20. See Heyes (2007a) for more discussion of tidal edges.

38

15

46

17

Legend: The shaded boxes indicate the terms used by lnuit hunters Male according to Female age and gender.

19 59

31

62

24

32 Ivuniq Inuktitut term Reference

Generation A age 55+ Generation B age 31-54 Generation C age 0-30

9

10

5

10

61 Tininik 63 Tuupauq

45 Sikuk 19 Kasak

59 Tininntupisus 38 Qarqak

33 Qaingngu

15 Ivuniq 17 Kangisuk

53 Sitjak

46 Sikuk killirut

Iqqaauk at high tide

9

45

24 Marqaulu siuraulu Kunninga 31 Nuvuk

Iqqaauk at low tide

61

Horizontal Units 9 Ingirrarikissup Ungmmuavinga 10 Iqqaauk

Table 1.

63

High

5 Igirnikisuk 62 Tininiup Sitjanga

Low

Tidal range

Figure 7.╇ Inuktitut nomenclature for horizontal units of the near shore and tidal flat realms when ice free, with an emphasis  on the elements exposed by tidal fluctuations.

© Dr Scott Heyes, 2010 Graphics by Dr Heyes & Anna Sundman Hand drawn illustrations by Kangiqsualujjuamiut hunters

Figure 7 Inuktitut nomenclature for horizontal units of the near shore and tidal flat realms when ice-free,with an emphasis on the elements exposed by tidal fluctuations

33 53

Between the trees and the tides 215

58

21

56

23

The shaded boxes indicate the terms used by lnuit hunters Male according to age Female and gender.

Legend:

57

22

4

23 Marqaq Inuktitut term Reference

Generation A age 55+ Generation B age 31-54 Generation C age 0-30

5 42

60

41 Qikittak

21 Maniraqtuq

55 Tariuk

54 Siuraq

48 Siluk

28 Nunak qaningituk 35 Qairtuq

55

8 Imaup killinga 16 Kangirsuo qingua

28

44 Sanirajak

8

42 Qiauinaaluk

16

23 Marqaq

54

22 Marqaq kiligu

Aputik killiut

44

5 Igirnikisuk

4

Table 1. Horizontal Units

48

60 Tininiartuk

58 Tininip ujarangit

57 Tassiup sinaanga

56 Tasiq

41

Figure 8.╇ Inuktitut nomenclature for horizontal units of the near shore and tidal flat realms when ice free, with an emphasis  on the shoreline elements.

© Dr Scott Heyes, 2010 Graphics by Dr Heyes & Anna Sundman Hand drawn illustrations by Kangiqsualujjuamiut hunters

Figure 8 Inuktitut nomenclature for horizontal units of the near shore and tidal flat realm when ice-free, with an emphasis on the shoreline elements

35

216 Scott A. Heyes



5.

Between the trees and the tides 217

Generational disparities of spatial nomenclature

To address the knowledge of spatial nomenclature among the study participants with respect to age brackets and gender, we will continue with the vignette of the Inuk hunter’s journey across the land and onto the sea ice. For this part of the story, the reader might refer to the information pertaining to nomenclature knowledge, which is provided in Tables 1 and 2 and Figures 2–8. On this particular excursion, the hunter leads a party of three male hunters (an elder, his son and grandson) and three female hunters (an elder, her daughter, and granddaughter) to hunt seals at the sinaa. These are representative of the generations of Kangiqsualujjuamiut that participated in this study. While passing onto the sea ice, the hunter points out features of interest. All members of the travelling party observe and recognize these ten features, which include: nunak (land), sitjak (beach), tinninik (tidal flat), ittiniq (rough ice), aajuraq (crack), tuvaq (flat ice), sikuk (ice), marqaq (mud), tariuk (sea water) and irqaq (sea floor). Traveling seaward, the hunters confront a number of other features of the sea ice realm, but on this occasion the younger and middle-aged generations indicate that they are unfamiliar with these features. The male and female elders pause to explain the names of the features to them. These include: ittini sinaanga, ittinikitaa, sikuutsuni tuvaq, sinaa tariup, siniritsijuk, uttuaq, ikaut ilua, ilumaniq, qinnitarsiaq, pukak, qaanik, quarni sikumi, and ivuniq. Having discussed the meanings and characteristics of these features, the party follows the lead of the Inuk hunter to the sinna. Waiting at sinna for seals, the hunter occupies time by recounting a story of hunting that includes a description of a number of spatial features of the land and sea. While appreciating the story, the youngest male and female hunters remain confused at various points throughout the telling, for they were only familiar with the ten terms described earlier by the hunter at the shore and the following four terms: akluk, ivujuk, siku ataa, tuvaup qaanga. The youngest male hunter joins in the conversation by talking about his knowledge of aputi sikuup qanganiituaq. The hunter and other members of the party discuss this term, but indicate they have not heard of it before. A greater discussion then arises amongst the party about how spatial knowledge is shared and conveyed among and across generations of hunters, with a real concern that the younger generations may not have, or may never attain, a strong command of nomenclature relating to spatially-laden terms. The hunters agree that this disparity in Inuktitut nomenclature knowledge between generations might be an outcome of the following factors: the infrequency of interaction and exchange of dialogue between elders and youth within the home or other social venues; the prohibitive costs of hunting and travelling, which effectively reduce the ability to engage with the environment; the difficulty

218 Scott A. Heyes

of joining and being accepted within well-established hunting groups; the proliferation of English in the community; and the social challenges that young members face in learning and communicating in Inuktitut in forums outside of the classroom. On their journey back to the village, the hunters discuss the extent to which the nomenclature of spatial terms might remain robust and relevant for future generations. There is consensus that this ultimately depends on addressing the factors outlined above, in addition to encouraging the development of new forums in which younger members of the community may learn and retain information about nomenclature.

6.

Elastic spaces and dissolved edges

By following the hunter’s journey across the sea when frozen and ice free, as discussed in the previous sections, we come to realize that the edges of the geographical units that make up these spaces are not necessarily obvious or clearly demarcated in all instances. The fluid and dynamic characteristics of the frozen and ice free settings, from decaying and accruing ice to rapid tidal fluctuations, essentially imply that many geographical units are shifting platforms. The extent and edges of the maritime spaces recognized and named by the KangiqsualujÂ� juamiut are continually being reconfigured and thereby negotiated. The spaces and edges of geographical units on the frozen and ice free realms are more elastic than they are fixed.21 The edges of one space dissolve into the edges of another, resembling what might be regarded as a patchwork of spaces that are broadly knitted together. By exploring how the Kangiqsualujjuamiut perceive spaces and edges at a regional scale – beyond the spaces that constitute the land and sea interface – we find that their ideas of edges are also elastic and negotiated. Influenced by notions of time, modes of mobility, senses of movement, seasonality, temperature and olfactory senses, the study participants offered descriptions and illustrations to explain how they conceptualized the edge of what they believed marked the extent of the Kangiqsualujjuamiut region proper. One participant explained that “It’s kind of … across the [George] River, down the hill or if I’m not a far distance 21. There are competing understandings of the point at which the sea and the land converge in the Arctic, depending on one’s perspective. Hydrographers argue that the boundary between the two realms is defined by the mean low water mark. To oceanographers, the boundary of the land and sea extends to the rim of the continental shelf (Cowen 1960). The UN Convention on the Law of the Sea (1982:â•›27) stipulates that, with respect to territorial claims to the sea, the inner boundary of the sea begins at the low-water line and extends seaward for 12 nautical miles.



Between the trees and the tides 219

from the town” (Heyes 2007a:â•›200–207). Another stated: “It feels like you have left George River when you go towards Nachvak Fiord, because the weather changes; it’s cold” (ibid.). In another account, the bounds of the Kangiqsualujjuamiut region were determined by viewsheds and hydrology: “The point of departure … is near the Tunnuliq River. It’s here that you start to see the ocean and there are fewer trees in that area. But when you go [south of here] you start to feel that there are more trees and fewer small mountains. That’s when you know you are in another area …. And, towards Kuujjuaq there are … big rivers. When you pass by these rivers, you always know that you are in the Kuujjuaq region” (ibid.). These examples suggest that ways of conceptualizing geographical units and edges are highly personalized and are based on one’s experience, knowledge, and level of interaction with the environment. They also suggest that the location of edges, and the spaces they contain, may be distinguished differently between even experienced hunters who are each well versed in spatially grounded Inuktitut terms. The interview excerpts illustrating Kangiqsualujjuamiut notions of edges on a broad scale, like those occurring at micro-level scales, highlight the variations in discriminating the extent of spaces among residents. The extent of geographical units, and the edges which demarcate them, might be commonly recognized by the Kangiqsualujjuamiut in a broad sense rather than in an absolute sense because it is not necessarily critical for them to maintain a consensus on the composition and configuration of geographical units, as do other indigenous groups. The precision with which geographical units are conceptualized, when compared to many coast-dwelling indigenous societies from Oceania, does not demand the same level of consensus.22 This is largely because the lands and waters of the Kangiqsualujjuamiut are not contested amongst themselves; the land and waters, and their bounty, are shared amongst the Kangiqsualujjuamiut people and fellow Inuit from elsewhere. This way of life has continued for generations and has been protected through land rights that were a product of the 1975 James Bay and Northern Quebec Agreement. In the absence of Arctic examples23 to draw upon, studies of Oceanic societies’ conceptualizations of sea space in the southern hemisphere provide avenues for 22. Ways of knowing and seeing space through metric distances and points of the compass, while certainly used more frequently among the Kangiqsualujjuamiut in the modern era of GPS travel, are still not the principle means used by this group for the conceptualization and communication of knowledge of their surrounds. As the use of GPS expands over time to replace traditional forms of navigation, and if the use of Inuktitut continues to diminish, Kangiqsualujjuamiut ways of conceptualizing space may change considerably. 23. The findings of the Sea Ice Knowledge and Use project, an International Polar Year initiative, may provide some insight. They are soon to be released.

220 Scott A. Heyes

inquiry about how sea space might be constructed amongst indigenous peoples. Their usefulness for comparing how the Kangiqsualujjuamiut conceptualize the coast is limited, though, because of ontological and epistemological differences in the ways that sea space is configured amongst these different groups. With this in mind, we see that the Solomon Islanders of Morovo Lagoon have strict sea tenure and management properties that are associated with parcels of sea space (Hviding 1996). Similar studies have been undertaken amongst the Torres Strait Islanders, which have shown that they possess a rich set of terms to describe parcels of land and sea for both tenure and utilitarian purposes (Nietschmann 1989; Scott & Mulrennan 1999). Among the Yanyuwa people of the Gulf of Carpentaria, it has been shown that sea space is described and classified by them in terms of relations to spirit ancestors, animals, tides and geography (Bradley et al. 2006). And in a study of Rossel Islander conceptions of terrestrial and seascape regimes in Papua New Guinea, Levinson (2008) has shown that they have a rich vocabulary to describe the submerged seascape, such as reefs and sea marks. The underwater seascape is a significant geographical region that has often been overlooked in studies of human cognition of the environment.24 In these examples from Oceania, the coastal environment is broken into spatial units that are largely arranged and managed according to ancestral, family and spiritual ties to the environment. Apart from the earlier discussions of Kangiqsualujjuamiut cosmological beliefs regarding the occupation of non-�human entities within distinct realms, the Kangiqsualujjuamiut do not appear to have tenurebased ways of thinking about geographical units and appropriating space. Unlike the examples from the southern hemisphere, the Kangiqsualujjuamiut do not have any cultural or social restrictions that prohibit the passage through, or the entering of, the coastal environment or any other environmental setting. The coastal environment, or parts thereof, are not owned or managed by any one person or family.

7.

Conclusion

Through the narrative of a hunter’s journey across the sea ice and open sea, and accounts from Kangiqsualujjuamiut informants about their notions and use of coastal space, we can see that the Kangiqsualujjuamiut conceptualize the coast as a series of spaces. These spaces, whether in the horizontal or vertical plane, have been conceptualized by the Kangiqsualujjuamiut over time as a way to describe

24. See Toupal et al. (2001) and Fletcher et al. (2001) for exploratory insights.



Between the trees and the tides 221

and transmit purposeful information about the coastal setting between hunters and fishers. In addition to developing ways of knowing and discriminating between spaces for safety and hunting efficacy, records from oral histories suggest that the Kangiqsualujjuamiut once had a sense of space of the coast that was based on spiritual and mythological beliefs. Their cosmological conceptions of sea space, while no longer observed today amongst the Kangiqsualujjuamiut as they were historically, suggest that there have always been ways to discern and discriminate the expansive Arctic coast amongst Inuit from the Kangiqsualujjuaq region. Although it has yet to be established whether the Kangiqsualujjuamiut have a taxonomy of sea space or other spatial regimes, their terminology for spaces and edges of the coastal environment suggests that they may have systematic and structured ways of organizing space. Whether sea space is conceived of as being structured or organic, a better understanding of how the Kangiqsualujjuamiut describe and discern sea space will undoubtedly help to inform discussions of sea rights, sea claims, global warming, hunting boundaries, and ways in which the coastal environment can be managed in the Arctic. Given that the KangiqsualujÂ� juamiut conceptualize the seascape as a series of spaces, it might be worth investigating whether or not other Inuit groups across the Canadian Arctic understand space in a similar manner. If so, new ways of mapping and describing the Arctic coast might become apparent to those who rely on non-Inuit ways of knowing and mapping the land. As inquiries of spatial cognition in the Arctic expand, nonInuit may come to appreciate that the “Great White North” is not an “empty room”, but a complex configuration of snow and ice spaces embedded with meaning.

Acknowledgments Thank you to the elders and hunters of Kangiqsualujjuaq for sharing their knowledge with me. I am grateful for the support that Prof. Peter Jacobs of the University of Montreal has provided in relation to the production of this chapter and associated fieldwork. I thank Christine Heyes LaBond, anonymous reviewers, and the editors of this book for their valuable comments and feedback on this chapter. For the production of the illustrations, I thank Anna Sundman for her kind assistance. Funding or in-kind support was generously provided by the following agencies and institutions at various stages of the study: John Crampton Scholarship, McGill University SSHRC, ArcticNet (2002–2006); Melbourne University Early Career Researcher Grant and Joint Research Grant (2007–2009); Faculty Research Program Grant, Canadian High Commission, Australia (2008); Arctic Studies Center, National Museum of Natural History, Smithsonian Institution;

222 Scott A. Heyes

Trent University SSHRC (2010), and the Roberta Bondar Postdoctoral Fellowship for Indigenous Studies and Canadian Studies, Frost Centre, Trent University€(2010).

References Arbess, Saul E. 1966. Social Change and the Eskimo Cooperative at George River, Quebec. Ottawa ON: Department of Northern Affairs and National Resources. Anonymous. 1833. The Moravians in Labrador. (20 May 2006). Edinburgh: J. Ritchie. Bradley, John, Holmes, Miles, Marngawi, Dinah N. & Karrakayn, Annie I. 2006. Yumbulyumbulmantha ki-Awarawu. All Kinds of Things from Country. Yanyuwa Ethnobiological Classification. The University of Queensland QLD: Research Report Series, Aboriginal and Torres Strait Islander Studies Unit. Brice-Bennett, Carol (ed.). 1977. Our Footprints are Everywhere: Inuit Land use and Occupancy in Labrador. Nain LB: Labrador Inuit Association. Cowen, Robert C. 1960. Frontiers of the Sea: The Story of Oceanographic Exploration. Garden City NY: Doubleday & Company. Fletcher, Christopher, Breeze, Heather, Cunningham, Stacey & Gregory, Shawn. 2001. Community Seascapes Project: Project Report to the Royal Canadian Geographic Society Research Grants Program. Saint Mary’s University NS: Gorsebrook Research Institute. Hallendy, Norman. 2000. Inuksuit: Silent Messengers of the Arctic. Vancouver BC: Douglas & McIntyre. Hantzsch, Bernhard. 1932. Contributions to the knowledge of extreme north-eastern Labrador. Canadian Field Naturalist (46): 153–163. Hawkes, Ernest W. 1916, The Labrador Eskimo. Ottawa ON: Government Printing Bureau. Heyes, Scott A. 2002. Protecting the authenticity and integrity of inuksuit within the arctic milieu. Inuit Studies 26(1–2): 133–156. Heyes, Scott A. 2007a. Inuit Knowledge and Perceptions of the Land-Water Interface. PhD dissertation, McGill University. Heyes, Scott A. 2007b. Inuit Knowledge and Perceptions of the Land-Water Interface: Supporting documentation to the PhD dissertation, Montreal QC: McGill University. Heyes, Scott A. & Jacobs, Peter. 2008. Losing place: Diminishing traditional knowledge of the Arctic coastal landscape. In Making Sense of Place: Exploring Concepts and Expressions of Place through Different Senses and Lenses, Frank Vanclay, Matthew Higgins & Adam Blackshaw (eds), 135–154. Canberra: National Museum of Australia. Heyes, Scott A., Labond, Christine & Annanack, Tummasi. 2003. The Social History of Kangiqsualujjuaq, Nunavik. Montreal QC: Avataq Cultural Institute. Hubbard, Leonidas Jr. 1906. Labrador, from Lake Melville to Ungava Bay. Bulletin of the American Geographical Society 38(9): 529–539. Hviding, Edvard. 1996. Guardians of Marovo Lagoon: Practice, Place, and Politics in Maritime Melanesia [Pacific Islands Monograph Series 14]. Honolulu HI: University of Hawaii Press.



Between the trees and the tides 223

Kohlmeister, Benjamin & Kmoch, George. 1814. Journal of a Voyage from Okkak, on the Coast of Labrador, to Ungava Bay, Westward of Cape Chudleigh. Undertaken to Explore the Coast, and Visit the Esquimaux in That Unknown Region. London: W. McDowall. Levinson, Stephen C. 2008. Landscape, seascape and the ontology of places on Rossel Island, Papua New Guinea. Language Sciences 30(2–3): 256–290. Lewis, Brian W. 1966. Inukshuks and Inunguaks on Foxe Peninsula and the North Quebec Coast. Canadian Geographic 73: 84–87. Lynch, Kevin. 1998[1960]. The Image of the City, 26th edn. Cambridge MA: The MIT Press. MacDonald, John. 2000. The Arctic Sky: Inuit Astronomy, Star Lore, and Legend. Toronto ON: Royal Ontario Museum. Makivik Corporation. 1985. Nunavik Inuit Land Use and Ecological Mapping Project. Montreal QC: Makivik Corporation. McGhee, Robert. 1984. Thule prehistory of Canada. In Handbook of North American Indians, Arctic, Vol. 5, David Damas (ed.), 369–376. Washington DC: Smithsonian Institution. McLean, John. 1849. Notes of a Twenty-Five Years’ Service in the Hudson’s Bay Territory: Volume€II. London: Richard Bentley. Müller-Wille, Ludger. 1987. Inuttutut Nunait Atingitta Katirsutauningit Nunavimmi (Kupaimmi, Kanatami): Gazetteer of Inuit Place Names in Nunavik (Quebec, Canada). Inukjuak QC: Avataq Cultural Institute. Müller-Wille, Ludger & Weber, Linna. 1983. Inuit Place Name Inventory of Northeastern Quebec-Labrador. Marburg: Avataq Cultural Institute. National Geospatial-Intelligence Agency (NGIA). 2005. Sailing Directions (Enroute): Newfoundland, Labrador and Hudson Bay [Pub. 146: 9th Ed]. Bethesda MD: United States Government, Dept. of Defense. Nietschmann, Bernard. 1989. Traditional sea territories, resources and rights in Torres Strait. In A Sea of Small Boats. Cultural Survival Report (Twenty Six), John Cordell (ed.), 61–93. Cambridge MA: Cultural Survival. Pereira, Irene Rice. 1956. The Nature of Space: A Metaphysical and Aesthetic Inquiry. New York NY: Privately Published. Saladin D’Anglure, Bernard. 1964. L’organisation sociale traditionnelle des Esquimaux de Kangirsujuak (Nouveau-Québec). PhD dissertation, Université de Montréal. Schneider, Lucien. 1985. Ulirnaisigutiit: An Inuktitut-English Dictionary of Northern Quebec, Labrador and Eastern Arctic Dialects. Quebec City QC: Les Presses de l’Universite Laval. Schultz, J. C. 1895. The Innuits of Our Arctic Coast. Proceedings and Transactions of the Royal Society of Canada for the Year 1894 12: 113–134. Scott, Colin & Mulrennan, Monica. 1999. Land and sea tenure at Erub, Torres Strait: property, sovereignty and the adjudication of cultural continuity. Oceania 70(2): 146–176. Toupal, Rebbeca S., Zedeno, Nives M., Stoffle, Richard W. & Barabe, Patrick. 2001. Cultural landscapes and ethnographic cartographies: Scandinavian-American and American Indian knowledge of the land. Environmental Science & Policy 4: 171–184. Tuan, Yi-Fu. 1974. Topophilia: A Study of Environmental Perception, Attitudes, and Values. Englewood Cliffs, NJ: Prentice-Hall. Turner, Lucien M. 2001 (1894). Ethnology of the Ungava District: Hudson Bay Territory. Montreal QC: McGill-Queen’s University Press. United Nations. 1982. United Nations Convention on the Law of the Sea. (25 June 2006). New York. Division for Ocean Affairs and Law of the Sea.

chapter 10

Differing conceptualizations of the same landscape The Athabaskan and Eskimo language  boundary in Alaska Gary Holton

This paper further explores the non-universality of landscape terms by focusing on one particular landscape, the Yukon Intermontane Plateau of western Alaska. This region serves as the boundary between two great language families of North America, Athabaskan and Eskimo, and thus offers a unique laboratory in which to examine the extent to which cultural factors in two genetically unrelated languages influence the categorization of a single, fixed landscape. Drawing on published lexical sources, unpublished place name documentation, and firsthand interviews with Native speakers, the results presented here demonstrate that, while Athabaskan and Eskimo speakers may occupy the same landscape, their respective languages conceptualize that landscape in different ways.

1.

Landscape as a linguistic domain

Recent research into the ontology of places in human language suggests that cultural ideas and practices are the major force driving the categorization of landscape (cf. Levinson 2008). As noted by Mark et al. (2007), “different language groups/cultures have different ways of conceptualizing landscape, as evidenced by different terminology and ways of talking about and naming landscape features”. As is evidenced by the papers in this book, the cross-linguistic study of these different landscape terminologies reveals different ways of conceptualizing landscape in language. To a certain extent this result is not unexpected. Landscapes vary significantly across the earth, so human experiences of landscape must necessarily vary as well. Our experience of taiga woodlands at the Arctic Circle is necessarily different from our experience of coral reefs on a Pacific atoll. Indigenous and non-indigenous landscape terminologies may differ greatly. But what

226 Gary Holton

happens when two different groups come to experience the same landscape? In this brief essay I explore that possibility by considering the conceptualization of landscape at the boundary of Athabaskan and Eskimo languages in Alaska. Alaska is the meeting ground of two of the New World’s largest language families: Athabaskan-Eyak-Tlingit (hereafter Athabaskan) and Eskimo-Aleut (hereafter Eskimo). Athabaskan has the largest territory of any indigenous language family in North America, reaching from western Alaska east nearly to Hudson Bay and south to the California coast and the desert Southwest. Pairwise mutual intelligibility varies little throughout the entire complex of some forty or so languages. A relatively small population ranging over a huge territory over thousands of years has resulted in a rich array of linguistic tools for categorizing the landscape of that territory. Within Alaska the Athabaskan languages occupy one contiguous region of territory, stretching without interruption from the middle Kuskokwim River in the west to the Alaska-Yukon border and beyond in the east, and from Cook Inlet in the south all the way to the Brooks Range in the north. Throughout Alaska the Athabaskan region shares a border with just one other language family: Eskimo. Eskimo is a clearly defined language family consisting of some five languages, three of which border Athabaskan: Inupiaq (Inuit), Central Alaskan Yup’ik, and Sugpiaq (Alutiiq). But this border is extensive, stretching in an arc nearly 2000 km long from Cook Inlet in the Gulf of Alaska, paralleling the coast of Alaska all the way to the Alaska-Yukon border (see Figure 1). In spite of the length of this border there is relatively little evidence of contactinduced influence on either of the two language families. Loanwords are relatively few (especially in the direction of Athabaskan from Eskimo), and there is little shared grammatical structure. In fact, it is difficult to imagine two more different language types. Eskimo is exclusively suffixing, while Athabaskan is almost exclusively prefixing. Eskimo morphology is agglutinative, while Athabaskan is the type example of synthetic, templatic morphology. Eskimo exhibits ergative case marking, while Athabaskan shows no traces of ergativity. The following examples from Tanacross Athabaskan and Central Yup’ik Eskimo demonstrate these differences. (1)

Tanacross Athabaskan  na-n-t-n-eg-’iił again-2sg-fut-see-1sg.cl-see.fut ‘I will see you again’

(Arnold et al. 2009)

. Although the term ‘Eskimo’ is considered derogatory in some parts of the world, in Alaska it is the preferred cover term to refer to both the Inuit (including Alaskan Inupiaq) and Yupik (including Central Alaskan Yup’ik) branches of the family (Kaplan 1999).



Differing conceptualizations of the same landscape 227

q

pia

Inu

Athabascan

Yu p

ik

Figure 1.╇ Zone of contact between Athabaskan and Eskimo in Alaska

(2)

Yup’ik Eskimo  ataam tanger-ciqa-mken again see-fut-1sg.subj/2sg.obj ‘I will see you again’

(Jacobson 1984)

The Tanacross example demonstrates a common property of Athabaskan verb stems, namely the phenomenon of distributed morphemes. The verb stem ‘see’ consists of two parts in the linear string: the rightmost form â•‚’ iił and a “prefix” n- occurring in a defined slot in the prefix string. The meaning ‘to see’ is distributed across both of these forms. In contrast, the Yup’ik morpheme ‘to see’ occurs on the left edge of the verb word. In Yup’ik the morphemes combine in an agglutinative manner with minimal phonetic accommodation. Tanacross exhibits a high degree of synthesis in the verb complex. For example, the prefix eg- in (1) results from a phonological merger of the first person singular subject prefix ih- and the “classifier” l-. Finally, note that in the Tanacross form the adverbial ‘again’ is incorporated as part of the verb complex, whereas in Yup’ik it occurs as a separate word. Yet while there may not be much that is shared between the two language families in linguistic terms, there is no denying the great extent of shared landscape along their shared border. The western portion of this contact zone lies largely within the Yukon Intermontane Plateau, a physiographic region of taiga woodlands characterized by broad valleys and low rolling hills ranging in elevation from 100€m to 500€m. This region is bordered by the high peaks of the Alaska

228 Gary Holton

Range to the south and east, including the highest peaks in North America. To the west are found the lowland river basins and deltas of large rivers, including Alaska’s longest river, the Yukon. Vegetation is characterized by spruce, aspen, and birch forest at mid elevations; willow and alder thickets on floodplains; and open tundra at higher elevations. Soils are poorly drained and wetlands occupy more than half of the region; permafrost is widespread (McNab & Avers 1996). Especially in the north, riparian features prevail, with numerous oxbow, thaw, and morainal lakes. The climate is continental with an average annual temperature of just below freezing and typical extremes ranging from –45 to +30°C. The landscape described in the preceding paragraph differs from the barren coastal plain and endless sea ice which are typically imagined to characterize the Eskimo region. This planar stereotype conceals the presence in Alaska of significant populations of Eskimo speakers in inland regions, in direct contact with Athabaskan speakers. The Yukon Intermontane Plateau has a long history as a zone of contact between Athabaskan and Eskimo peoples. Even today, after decades of indigenous language attrition under the influence of English, bilingual speakers of Eskimo and Athabaskan can still be found in border regions, including Stony River on the Kuskokwim River, Holy Cross on the Yukon River, and Alatna on the Koyukon River. Far from being seaward-looking coastal settlements, these villages are located well inland – hundreds of kilometers from the Bering Sea – in areas of taiga forest and topographic relief. Bilingualism between Koyukon Athabaskan and Inupiaq Eskimo facilitated a late-19th century cultural and linguistic shift in the Kobuk River valley, in which Athabaskan speakers adopted Inupiaq as their first language and eventually abandoned Koyukon (Burch et al. 1999). Similarly, bilingual Yup’ik Eskimo and Dena’ina Athabaskan place names attest to an area of overlapping Eskimo and Athabaskan territory on the middle Kuskokwim River. For example, Barometer Mountain (elevation 757 m) is known in Central Yup’ik Eskimo as Ekleq ‘burned place’ and in Deg Hit’an Athabaskan as Digheq’un Deloy ‘burned mountain’ (Kari 1980). The two names share a specific meaning ‘burned’, but only the Athabaskan name contains the generic ‘mountain’. In spite of this common landscape, preliminary investigation of landscape terminology in Athabaskan and Eskimo reveals some striking differences. Given the relative wealth of documentation for these two great language families as compared to other indigenous languages of North America, this is an area ripe for further research. In this brief essay I will focus on just two aspects of landscape categorization: elevations and streamscapes.



2.

Differing conceptualizations of the same landscape 229

Elevation terms

Both Athabaskan and Eskimo have a well-defined category of mountain. In this sense Athabaskan and Eskimo are similar to each other (and to English) and different from languages such as Yélî Dnye, which Levinson (2008) has convincingly argued lacks such a category. However, Athabaskan and Eskimo each have different systems for categorizing elevations. In particular, Athabaskan has two distinct generic elevation terms for which scale is a crucial defining feature, while Eskimo languages only have one such term. The Dena’ina Topical Dictionary (Kari 2007), the most complete source of Alaska Athabaskan landscape terminology, lists dozens of Dena’ina Athabaskan elevation terms but just three primary generic terms: dghili ‘mountain,’ tex ‘hill,’ and ses ‘ridge.’ These terms can be reconstructed for Proto-Athabaskan, and though there is quite a bit of variation as to their semantics in the individual languages, reflexes of each of these terms can be found in all of the Alaska Athabaskan languages. What is particularly interesting here is that Athabaskan languages distinguish two convex landform terms roughly equivalent to English ‘mountain’ and ‘hill’. As in English, scale is a defining criterial feature distinguishing these terms. This distinction holds true across the Athabaskan languages. In Koyukon Athabaskan, spoken just to the north of Dena’ina, the term teyh refers to a ‘small hill, hillock’ and dleł to a ‘mountain, large hill, range of mountains or high hills.’ The latter term is “also said of hill, but generally restricted to larger ones” (Jones & Jetté 2000:â•›156). Scale is an important defining factor for both dghili and tex in Dena’ina. Not only is dghili generally larger than tex, but both are large as viewed by humans. The term dghili often refers to high, glaciated mountains which are relatively inaccessible to humans. For example, the Central Alaska Range is Dghili Teh ‘among the mountains’, and Mt. McKinley (Denali) is Dghili Ka’a ‘big mountain’. The term tex may refer to more accessible places – for example, hills which can be easily traversed on foot – but tex still refers to elevation features which are larger than humans. One cannot simply step over or jump off of tex. Scale is crucial to the meaning of Athabaskan elevation terms. Eskimo languages also distinguish two primary elevation terms corresponding roughly to English ‘mountain’ and ‘hill’. The Yup’ik Eskimo Dictionary (Jacobson 1984) lists ingriq ‘mountain’ and penguq ‘hill’. The term ingriq refers to a prominent peak and is common in place names. These include both names of mountain peaks such as Ing’errlak ‘Roberts Mountain’ (lit. ‘major mountain’) on Nunivak Island and Ingri’urluq ‘Three Step Mountain’ (lit. ‘poor dear mountain’) near the Kwethluk River, as well as village names such as Ingricuar ‘Twin Hills village’ (lit. ‘little mountain’) or Ingrirrlleq ‘Holy Cross Village’ (lit. ‘shabby

230 Gary Holton

bit of mountain’). This root does not occur in words which do not refer to a mountain or place name. Scale is an important factor in the semantics of ingriq; this term refers only to large peaks. In contrast, the Yup’ik term penguq is quite different in that it may refer to a range of scales. Jacobson (1984) lists nine terms derived from the root form peng-, namely pengurpall’er ‘great big hill,’ pengurpak ‘big hill,’ penguq ‘hill’, penugucuar ‘small hill’, pengurraq ‘little hill’, penuyaaq ‘tiny hill,’ penguruaq ‘imitation hill’, penguguayaaq ‘baby hill’, penguquiner ‘little bit of a hill’. The crucial semantic feature of penguq is its convex shape. Other derivations of the root peng- include the terms pengunquq ‘mound’ and pengulkuq ‘tussock’, both of which may refer to rather small convex protrusions. In some dialects (e.g., Hooper Bay, Chevak, and Nunivak) the corresponding form pengu means not ‘hill’ but rather ‘(sand) dune’ of any size. The cognate Inupiaq Eskimo term piŋuq means not only ‘individual round hill’ but also ‘swelling’. The English geologic term ‘pingo’ is borrowed from Eskimo and denotes a mound formed by an expanding ice lens as the result of rapid draining of a thermokarst lake. While the most famous examples of pingos in the MacKenzie Delta of northwestern Canada may reach 50 m in height, pingos in western Alaska are typically much smaller, many no taller than a human. In English as in Eskimo, the defining feature of a pingo is its convex swollen shape, not its height. In fact, ‘pingo’ is probably a better translation of Inupiaq piŋuq than is ‘hill’, in the sense that the English technical term ‘pingo’ is scale-independent (though of course the Inupiaq term is not restricted to ice lens features). This is in sharp contrast to English ‘hill’ or Dena’ina tex, for which scale is a defining feature. Further evidence for the primacy of shape rather than size in the meaning of penguq comes from the metaphorical use of the root in derived word forms. The verb pengiga’rte- ‘to break out in hives’ means literally ‘to suddenly be full of penguq,’ implying penguq as a kind of swelling. In the Yukon dialect the word pengulkuq refers to a tussock of grass on the tundra, which could also be viewed as a kind of swelling. It is also interesting to note that, while all the other Yup’ik elevation terms occur frequently as roots in place names, the root penguq does not. The term penguq describes the shape of the landscape but fails to isolate a landform. The presence of two different scale-dependent elevation terms in Athabaskan where Eskimo has just one reflects a greater emphasis on the scale and extent of elevations in Athabaskan. In contrast, Eskimo elevation terminology focuses more on the convex shape or quality of the landscape than on its vertical extent. Further

. The only potential exception that I am aware of is Pengui, denoting an area of sand dunes on southwestern Nunivak Island.



Differing conceptualizations of the same landscape 231

evidence of the greater Athabaskan focus on topographic relief can be found by comparing the conceptualization of streamscapes in the two language€families.

3.

Streamscapes

Taken in their entirety, Athabaskan and Eskimo landscapes differ most in their relation to the ocean. The greater part of Eskimo territory lies along the coast, stretching inward along estuaries and river deltas; no Eskimo language community is without a saltwater coast. In contrast, Athabaskan territory lies for the most part inland, and only one Athabaskan language, Dena’ina, has an ocean coastline€– and even then Dena’ina can be shown to have annexed coastal territory relatively recently (Kari 1996b). So it is not surprising that Eskimo marine and coastal terminology is much more elaborate and developed than it is in Athabaskan. Oozeva et al. (2004) document nearly one hundred terms relating to sea ice alone. See also Heyes, this volume. In contrast, in Athabaskan the ocean is simply yaatuu chox ‘distant big water’ (Tanacross) or nuti ‘salt’ (Dena’ina). In fact, the Dena’ina Athabaskan place name for Cook Inlet, the major body of saltwater within Athabaskan territory, extending inland some 300 km from the Gulf of Alaska, draws an analogy with a river: Tikahtnu ‘big water river’. But the zone of contact with which we are concerned here lies far from the ocean shore. Indeed, it is appropriate to focus here on the concept river, for the zone of contact between Eskimo and Athabaskan – the landscape with which this paper is concerned – lies not on the coast but inland among broad river valleys and rolling hills of the subarctic taiga. While both Athabaskan and Eskimo languages have extensive inventories of river terminology, several differences can be discerned. First, while there is quite a bit of variation in the realization of stream terms among Athabaskan languages (Kari 1996a), all Alaska Athabaskan languages distinguish major and minor streams with distinct roots. Examples are Koyukon no’ ‘river, creek, stream, course of stream’ versus hen ‘large river’. In contrast, Yup’ik stream terms are either based on the root kuik ‘river’ or derived from noun or verb roots which themselves have more general meanings. There are no distinct Yup’ik roots distinguishing sizes of stream. Associated riverine terminology is much more developed in Athabaskan than Eskimo. One place where this is apparent is in the concept of river mouth. The Dena’ina stem â•‚kaq’ ‘river mouth’ occurs frequently in suites of place names sharing the same specific term (see below). It refers only to river mouths and is distinguished from roots such as â•‚du ‘orifice’ and â•‚zaq’ ‘human mouth’. In contrast, Yup’ik (Nunivak) and Inupiaq paa- refers more generally to ‘entry, door, opening

232 Gary Holton

(river mouth)’ (Amos & Amos 2003; Webster & Zibell 1970). In order to refer unambiguously to a river mouth, the term paa must occur in absolutive case as the possessed noun in a possessive construction: kuigem paanga ‘river’s mouth’. Other possessor nouns yield other types of openings, for example, qayam paanga ‘kayak’s hatch.’ In contrast, Dena’ina â•‚kaq’ is unambiguously a river mouth. Only with additional derivation can its meaning be extended, as in hâ•‚daâ•‚kaq’ ‘doorway’ (area-opening-mouth). Toponymy provides additional evidence of the relatively greater importance attached to the concept of river mouth in Athabaskan as compared with Eskimo. The root â•‚kaq’ is a common component of Dena’ina Athabaskan place names, usually referring to villages located at the mouth of a river. Examples include K’qizaghetnu Hdakaq’ ‘Stony River Village’, Hek’dichen Hdakaq’ ‘Lime Village’, and Tsayehtnu Hdakaq’ ‘Pile Bay Village’. In contrast, there are few if any toponyms in Yup’ik based on the root paa-. A possible candidate would be Paamiut, literally ‘people of the river mouth’, referring to the people of the southeast coast of Nunivak Island – and by extension the site at which those people reside in this region. Another candidate is Paaluyar, referring to the site at the mouth of the Mekoryak River, though this word is perhaps better viewed as a general term for ‘river mouth’. The Dena’ina Athabaskan river mouth term â•‚kaq’ differs from Yup’ik paa- in another important way. Rather than being an isolated term, the Athabaskan river mouth term is part of a larger structural system of riverine categories, reflecting Proto-Athabaskan *-kæq’e ‘stream mouth’, *â•‚tł’at ‘stream headwaters’, *â•‚wen ‘lake,’ and *-na’ / *-niq’e ‘stream.’ These four landscape categories are the most frequent landscape terms in Athabaskan languages and occur repeatedly as generic terms in suites of toponyms sharing a common specific name (Kari 1996c). It is possible to identify semantically equivalent Eskimo terms for at least some of these categories, but they do not function in the same systematic way that the Athabaskan terms do. Yup’ik nanwar ‘lake’ and kuik ‘river’ have roughly the same semantic range as Dena’ina ven and k’etnu, respectively. But only in Dena’ina are these terms clearly connected via a term meaning ‘river mouth’. The Dena’ina root â•‚kaq’ regularly denotes the mouth of a stream – and by metynomic extension any settlement or site located there. As discussed above the Yup’ik root paa- has similar semantics but does not occur regularly. The fourth common Athabaskan riverine term, Dena’ina â•‚tł’ugh, doesn’t seem to have any corresponding term in Eskimo. This reflects another fundamental difference in landscape categorization. In Dena’ina all streams have headwaters . The distinction between PA *â•‚na’ and *â•‚niq’e and their role in distinguishing hydronymic districts are discussed in Kari (1996c).



Differing conceptualizations of the same landscape 233

which can be directly designated with the term tł’ugh. In Yup’ik the meaning ‘headwaters’ is a metaphorical extension of the term kangiq, whose basic sense is ‘meaning, principle, source’, as in the derived verb stem kanginge- ‘to discover’, literally ‘to acquire meaning.’ It is also possible to refer to upstream areas using directional terms or postural roots, though Yup’ik lacks the full inventory of directional terms available in Athabaskan. And postural roots have a much wider semantic range than Dena’ina tł’ugh. For example, the Yup’ik postural root aci‘area below, under’ can refer to the base of a mountain, the area under a table, or the area under a river bank. While Athabaskan and Eskimo riverine terms may be used to refer to the same landscape, they delineate distinct prototype landscapes. The grid of Athabaskan terms denoting stream, mouth, and headwaters reflects a prototype streamscape consisting of a V- or U-shaped valley which rises significantly in elevation from mouth to head. The headwaters of these valleys may terminate in a headwall, steep pass, or even a glacier. There is no distinct term meaning ‘valley’; in a real sense the suite of stream terms is the valley. Jones and Jetté (2000) even list ‘valley’ as one of the meanings of the cognate Koyukon term â•‚tł’ot. Kari (2007) lists a Dena’ina term ch’ahdaniłtunt ‘valley, gully, ravine’, but this term actually derives from the verb h-d-ł-tun ‘ravine extends’, based on an abstract root tun referring to linear objects. So this Dena’ina term is really making reference to the linear feature of a valley rather than to the valley itself. Even more striking evidence for the primacy of the valley in Athabaskan conceptualization of the landscape can be found from cases where valleys suddenly empty out onto large floodplains. In these regions the stream mouth term (reflecting PA *-kæq’e) may be used to refer to the point of transition from steep valley to flat floodplain, even if there is no actual stream mouth. An example of such a situation occurs in Tanana Athabaskan territory. Figure 2 shows an oblique view looking northeast up the valley of the Chatanika River. In the center of the image the river suddenly leaves the valley and emerges onto the Minto Flats, a region of interconnected lakes, streams, and marsh covering some 300,000 ha. The line marking the boundary between the base of the mountains and the start of the flats runs roughly from the upper left to the lower right corner of the figure. The English name Chatanika derives from the Tanana Athabaskan name Dradlaya Nik’a, literally ‘whitefish (Prosopium cyÂ� lindraceum) river’. The place where the river emerges from the valley is Dradlaya Chaget, literally ‘whitefish mouth’, following the standard Athabaskan generative naming pattern whereby a single specific term (in this case dradlaya) is combined with a generic geographic term (in this case chaget). Thus, we also find Dradlaya Sedha’ ‘whitefish hill’, etc. Viewed from the perspective of English physiography, it is possible to trace the course of the Chatanika River as it meanders across the

234 Gary Holton

Dradlaya Nik’a

Dradlaya Chaget

Figure 2.╇ Minto Flats and the “mouth” of the Chatanika River; see text for further details

vast Minto Flats (referred to in Athabaskan as Men Teh ‘among the lakes’), nearly merging with Washington Creek, until it finally meets the Tolovana River, a more major stream which flows across the western side of the Minto Flats. Indeed, the official English name assigned to the river extends along this route. But from the point of view of Tanana Athabaskan the Chatanika River ends at Dradlaya Chaget. Beyond that point the river is no longer identified as an individual stream course but rather by a variety of names referring to interconnected lakes and sloughs. When it finally does straighten out and begins to look like a river again, just prior to joining the Tolovana, it is called not Dradlaya Nik’a but Nonilen No’, literally ‘water flows creek’. It is not the same river but a different river. In contrast, the 100 km of river above Dradlaya Chaget are continuously referred to as Dradlaya Nik’a. The valley is the river. Returning to Eskimo we find a variety of synonymous terms used to denote river valleys. In Yup’ik the most prominent of these is naqa, which Jacobson (1984) defines as ‘valley’, but which comparative evidence suggests might better be glossed as ‘pass,’ as in a passage between two mountains. The term does not occur in other Yupik languages or even in all dialects of Central Alaskan Yup’ik (Fortescue et al. 1994), and the corresponding Inupiaq term aaġiaq is glossed ‘valley, pass’. More common are terms based on the root kuik ‘river’, as in Nunivak dialect kuigyaner, literally ‘thing that results from a river’. In the Yukon dialect we find the term ilutak ‘valley, dip, bay’, based on the root ilu- ‘interior’. These terms suggest a conceptualization of the valley as a depression such as those found on the lower courses of a river delta rather than a montane valley.



4.

Differing conceptualizations of the same landscape 235

Conclusions

Due to the preliminary nature of this study, the above discussion has been limited to elevation and streamscape terminology. Yet even with this limitation we can make some clear generalizations about Athabaskan and Eskimo conceptualizations of their shared landscape. Most significantly, Athabaskan languages emphasize vertical features and mountain valleys, while Eskimo languages are less concerned with vertical scale and the notion of valley. (This generalization is further supported by comparison of toponymic structure and directional systems, though those issues must remain beyond the scope of this brief essay.) Ultimately, landscape categories tell us something about how humans relate to the landscape. “Landscape features are more likely to be driven by their ‘affordances’, by what they are good for in human activities and purposes” (Levinson 2008). In light of this statement we might surmise that Athabaskan and Eskimo speakers experience their shared landscape differently, perhaps because those conceptualizations are based on significantly different archetypal landscapes. While Athabaskan and Eskimo may share a long boundary region in Alaska, the archetypal Athabaskan landscape remains the rugged mountainous Alaskan interior region, while the archetypal Eskimo landscape lies along the relatively flat Bering Sea and Arctic Ocean coasts. Of course, such speculation must of necessity be viewed with caution. The study of landscape categorization relies ultimately on a comprehensive documentation of the landscape lexicon, and while Alaskan languages are generally considered to be well-documented, the coverage of landscape vocabulary is rather uneven. For the Athabaskan languages, reasonably comprehensive lexical coverage is available for only two languages: Ahtna and Koyukon. In addition, comprehensive coverage of landscape vocabulary for Dena’ina can be found in Kari (2007). For the remaining languages the extent of coverage of landscape vocabulary must be inferred indirectly by seeking cognates from the better documented languages. For the Eskimo languages the situation is even more challenging. Technically comprehensive dictionaries exist for two Alaskan Eskimo languages: Central Alaskan Yup’ik (Jacobson 1984) and St. Lawrence Island Yupik (Jacobson 2009), but while these dictionaries provide meticulous coverage of morphological issues, their coverage of semantics is less detailed. For the study of landscape categories such cursory treatment limits our ability to understand fine distinctions between terms. We saw an example of this when we attempted to understand the distinction between Yup’ik naqa ‘valley’ and ilutak ‘valley, dip, bay’. We can infer something about the semantics of naqa by examining cognates in Eskimo languages outside the Yup’ik subfamily, but such methods are not wholly reliable for understanding synchronic meanings.

236 Gary Holton

These difficulties point to the need for more first-hand field work with speakers of Athabaskan and Eskimo languages. Unfortunately, for this short paper I was able to do no more than make brief inquiries of speakers – sometimes relying on email and telephone. The tentative conclusions drawn here thus hide interesting variation across languages within each of the two families. Clearly, much more work needs to be done to understand the full extent of landscape categorization in Athabaskan and Eskimo languages.

References Amos, Muriel M. & Amos, Howard T. 2003. Cup’ig Eskimo Dictionary. Steven Jacobson (ed.). Fairbanks AK: Alaska Native Language Center. Arnold, Irene, Thoman, Richard & Holton, Gary. 2009. Tanacross Learners’ Dictionary: Dihtâad Xt’een Iin Anděg Dínahtlaˇa’. Fairbanks AK: Alaska Native Language Center. Burch, Ernest S., Jr., Jones, Eliza, Loon, Hannah P. & Kaplan, Lawrence. 1999. The ethnogenesis of the Kuuvaum Kaŋiaġmiut. Ethnohistory 46(2): 291–327. Fortescue, Michael D., Jacobson, Steven & Kaplan, Lawrence. 1994. Comparative Eskimo Dictionary: with Aleut Cognates [Alaska Native Language Center Research Paper 9]. Fairbanks AK: Alaska Native Language Center. Jacobson, Steven. 1984. Yup’ik Eskimo Dictionary. Fairbanks AK: Alaska Native Language Center. Jacobson, Steven. 2009. St. Lawrence Island Yupik Eskimo Dictionary. Fairbanks AK: Alaska Native Language Center. Jones, Eliza & Jetté, Jules. 2000. Koyukon Athabaskan Dictionary. Fairbanks AK: Alaska Native Language Center. Kaplan, Lawrence. 1999. Inuit or Eskimo: Which names to use? Alaska Native Language Center, (30 March 2010). Kari, James. 1980. Kuskokwim river placenames by Gusty Mikhail [Item IK974K1979f]. Ms, Alaska Native Language Archive, Fairbanks AK. Kari, James. 1996a. A preliminary view of hydronymic districts in Northern Athabaskan prehistory. Names 44(4): 253–271. Kari, James. 1996b. Linguistic traces of Dena’ina strategy at the archaic periphery. In Adventures through Time: Readings in the Anthropology of Cook Inlet, Alaska, Nancy Y. Davis & William E. Davis (eds), 51–63. Anchorage AK: Cook Inlet Historical Society. Kari, James. 1996c. Names as signs: The distribution of ‘stream’ and ‘mountain’ in Alaskan Athabaskan languages. In Athabaskan Language Studies: Essays in Honor of Robert W. Young, Eloise Jelinek, Sally Midgette, Keren Rice & Leslie Saxon (eds), 443–68. Albuquerque NM: University of New Mexico Press. Kari, James. 2007. Dena’ina Topical Dictionary. Fairbanks AK: Alaska Native Language Center. Levinson, Stephen C. 2008. Landscape, seascape and the ontology of places on Rossel Island, Papua New Guinea. Language Sciences 30(3): 256–90.



Differing conceptualizations of the same landscape 237

Mark, David M., Turk, Andrew G. & Stea, David. 2007. Progress on Yindjibarndi ethnophysiography. In Proceedings of the 8th International Conference on Spatial Information Theory, Stephan Winter, Matt Duckham, Lars Kulik & Ben Kuipers (eds), 1–19. Melbourne: Springer. McNab, W. Henry & Avers, Peter E. 1996. Ecological Subregions of the United States [Administrative Publication WO-WSA-5]. Washington DC: U.S. Department of Agriculture, Forest Service. Oozeva, Conrad, Noongwook, Chester, Noongwook, George, Alowa, Christina & Krupnik, Igor. 2004. Watching Ice and Weather Our Way; Sikumengllu Eslamengllu Esghapalleghput. Washington DC: Arctic Studies Center, Smithsonian Institution. Webster, Donald & Zibell, Wilfried. 1970. Iñupiat Eskimo Dictionary. Fairbanks AK: Summer Institute of Linguistics.

chapter 11

A case study in Ahtna Athabascan geographic knowledge James Kari

Ahtna is an Athabascan language of south-central Alaska centered mainly on the Copper River. Drainage-based files of over 2,200 Ahtna place names have been maintained for over thirty years, making Ahtna one of the best researched geographies for an Alaska Native language (Kari 2008, 2010). There are transparent principles that govern the content, structure, and distribution of Ahtna place names. The Ahtna geographic system is framed in a riverine absolute landmark orientation system (Levinson 2003) with nine directional roots that occur in over sixty derived forms. There is a distinct generative geographic capacity to this system, whereby a specific “sign” combines with an array of generic terms and directionals to facilitate memorization of the geography. The ways in which place names intersect with the directionals are complex and invite further study. Furthermore, the Ahtna system is representative of the Northern Athabascan languages.

1.

Introduction

There is increasing interest in cross-linguistic research on spatial cognition that has been fostered by Levinson’s 2003 book, the case studies on landscape and language in Language Sciences (Burenhult 2008 and Levinson 2008), as well as the papers presented at the October 2008 conference “Landscape in Language”. The Ahtna language area is predominantly the drainage of the Copper River, a 250-mile long stream that heads on the north slopes of the Wrangell Mountains and then flows into the Gulf of Alaska (Figure 1). Ahtna is a member of the

. I would like to thank Adeline Kari, Gary Holton, Andrea Berez and Siri Tuttle for comments made on earlier drafts of this paper. I also would like to thank participants in the Oct. 26–31, 2008 Landscape in Landscape workshop, in particular Steve Jett, Niclas Burenhult, Tom Thornton, and David Mark.

240 James Kari

Athabascan language family, the largest indigenous language family in territory in North America. It is part of the Na-Dene (or Athabascan-Eyak-Tlingit) language stock. The Ahtna people have a strong sense of identity and territory. Today there are about 1000 persons who call themselves Ahtna, and there are about 50 speakers of the language. The Ahtna live at or near traditional village sites. Ahtna Inc., the regional Native corporation founded in 1974, owns or manages 1.77 million acres of land within the boundaries of the traditional Ahtna language area. Drainage-based files of over 2200 Ahtna place names have been maintained for over thirty years, making Ahtna one of the best researched geographies for an Alaska Native language (Kari 2008). The most famous and striking themes about the Na-Dene language stock are: (1) the sheer size of the language area in western North America, about 53 languages with a total area of over 1,500,000 square miles at the times of first historic contact, divided into three regions with groups of contiguous languages – the northern subarctic, the Pacific coast (coastal Oregon and California), and the desert Southwest; and (2) the unique typological profile of the languages that features elaborate and rather similar verb complexes with strings of rigidly ordered prefixes before a verb root. Of recent note, (3) there is rather substantial evidence that the Yeniseian language family of Central Siberia is related to Athabascan and to Na-Dene (Vajda 2010). This hypothesis is prompting wide-ranging questions about how and when Dene-Yeniseian may have been configured in Siberia, Beringia, and Alaska. In this paper I summarize in Section 2 the key features of Ahtna landscape terminology and geographic names. The place names are highly structured and analyzable and can be readily memorized. These features are presented in fuller detail in the 36-page introduction to Kari (2008). In Section 3 I give an overview of the riverine directional system that has a pervasive role in Ahtna and in other Athabascan languages and a brief discussion of some of the ways in which Ahtna place names intersect with the riverine directionals in a collection of Ahtna travel narratives (Kari 2010). We draw the Ahtna language and dialect boundaries as in Figure 1 at the onset of historic contact in the 1880s. The language area comprises approximately 90,000 square kilometers (35,000 square miles). The 2008 report includes many Ahtna names beyond the language boundaries in an area well over 130,000 square kilometers (50,000 square miles). Ahtna ethnogeography has been well documented with place names lists, geographic narratives, and many historic records. The result is a shared, verbally transmitted, and strongly confirmed geographic names corpus. The portrait of Ahtna territory in Copper River Basin and adjacent areas reflects human foot travel on trail networks without much use of boats and without major dependence on motorized vehicles. Aboriginally the Ahtna



A case study in Ahtna Athabascan geographic knowledge 241

Figure 1.╇ Ahtna language area

had four or five types of watercraft. There is only one area where birchbark canoes were useable in the warmest months (the upper Gulkana River lakes and the Tyone Lake area). The mutually known geographic system that is conveyed via the compiled Ahtna place names lists precedes all historic modes of transportation, land ownership patterns, and various land management jurisdictions.

2.

Ahtna geographic names

There are many sources on Ahtna language and culture. Of importance is the highly readable description of the Ahtna people at the onset of historic contact in 1885 by Henry T. Allen (1887). Ahtna had a very small aboriginal population (perhaps never more than 1000 persons), and large band territories (twelve or so bands, each with band territories of 8,000 to 13,000 square kilometers (3000 to 5000 square miles) in area, with comprehensive and ancient trail systems that were in use year-round in this harsh subarctic climate. The fact that such a comprehensive network of Ahtna place names extends throughout and beyond the

242 James Kari

large language area is a significant demonstration of the prodigious travel prowess and geographic expertise of the small Ahtna or Athabascan populations that have occupied the Copper River Basin for a very long time. We can identify an array of principles and mnemonic devices that combine to facilitate place recognition and efficient foot-travel through and beyond Ahtna territory. The extensive sources on Ahtna geography are summarized in Kari (2008:â•›7– 12). About 150 Ahtna speakers have contributed names and locations in the historic period, beginning with twenty Ahtna words and place names recorded in 1797 by Russian navy man Dmitri Tarkhanov (Black 2008). The oral transmission of Ahtna and Athabascan place names is conservative; people report only the names they have learned, and they do not coin names. There is strong reiteration of name sequences over time by different speakers or by the same expert speaker. Probably 60% of the 2208 names have been confirmed by several speakers. The recent book of travel narratives (Kari 2010) provides context and exemplification on different patterns of place name redundancy. A few Ahtna names, such as Tezdlen Na’ for Tazlina River, have been reported consistently by Ahtna speakers in 1797, the 1830s, 1848, 1885, 1898 and throughout the 20th century. Thanks to experts such as Jake Tansy, Jim McKinley, Katie John, Fred John, Adam Sanford and others, Ahtna has the most comprehensive geographic name data set for an Alaska Native language. The following subsections are summaries of the essential features of Ahtna geographic names. (The introduction to the 2008 report has a more extensive presentation.) Ahtna place names can be summarized in terms of structural patterns, information content, distribution, reiteration, use in overland navigation, occurrence in narrative, and other features. Based upon several Athabascan place name

NAME STRUCTURE

NAME CONTENT

NAME DISTRIBUTION

NAME NETWORKS

Reinforcing memorization & conservatism

Figure 2.╇ Integrated elements in Ahtna and Athabascan geographic names systems



A case study in Ahtna Athabascan geographic knowledge 243

projects over the past decade or so, I have identified four elements in Athabascan geographic names that are similar. These elements, as shown in Figure 2 – place name content, structure, distribution and networks – are mutually reinforcing and contribute to the learning and memorization of the names.

2.1

Place name structure

Ahtna and other Alaska Athabascan languages have a very large battery of lexical and grammatical features to portray space and orientation. Athabascan languages have a grammatical dichotomy between simplex nouns and complex verbs. The Ahtna geographic names are a mixture of the simplex and the complex. There is extensive use of basic nouns and postpositions, many of which are monosyllabic and are not subject to much derivation and modification. The verb contrasts sharply with the nouns and postpositions. Ahtna (and more generally Athabascan) verb structure consists of ten to fifteen rigidly ordered prefix positions or zones of similar prefixes before a stem, which is followed by sets of suffixes. The key concept for organizing and understanding Ahtna verbs is the verb theme. Many verb themes have hundreds of derived forms that offer subtle distinctions for space, path and orientation (Kari 1989, 1990:â•›38–58, 1992). The verb morphology is templatic, with layered interdigitation of strings of morphemes. In the conservative Ahtna verb complex, virtually every prefix, suffix and layered derivation can be discerned. Athabascan dictionaries, such as for Ahtna (Kari 1990) and Koyukon (Jetté & Jones 2000), list out verb themes in their underlying structures, which can consist of zero to several prefixes before the underlying root as well as a notation for transitivity and gender marking. About 30% of Ahtna place names are nominalized verbs; some of the most common in geographic names are extension verb themes that can be derived into many hundreds of words. Many of these same extension verb themes are common in place names in other Athabascan languages.

ti#0+taan 0+len, d+ł+len G+0+’aa⁰ G+0+’aa

‘trail extends’ ‘current flows’ ‘linear feature extends’ ‘sg. compact object is in position’

. In the Ahtna practical orthography, c::k, c’::k’, and g::gg are the front vs. back velar series. The e- is the schwa vowel, which has the quality of [ε] epsilon. Other symbols for glottalized consonants and affricates are typical of other Athabascan practical orthographies such as Navajo. Also note that in Ahtna plain stops are written as d, dl, dz, g, gg, but when in word-final position, they are written as t, tl, ts, c, k. Underlyingly these remain d, dl, dz, g, gg.

244 James Kari

G+0+laa⁰ (lae) ‘plural objects are in position’ G+0+taan ‘elongated object is in position’

Each of these verb themes can be derived into hundreds of words by derivations that occur in strings that interdigitate with the basic verb theme as well as with other strings. For example, two Ahtna derivational strings are na# gh- momentaneous/neuter. ‘downward’ and ts’i# n- momentaneous/neuter. ‘out from.’ Each string has three prefixes and a set of four suffixes. When these apply to the verb theme, they interdigitate with preceding layers of morphemes. These strings apply like multi-pronged clamps that grip around and within several pieces of wood. Many strings that apply to verb themes are the same in other Athabascan languages (Kari 2010). Consider the striking congruity in three place names: the well-known Navajo-origin name Chinle, and two different Ahtna or Dena’ina place names. Navajo: Ts’ínílí˛ (Chinle) ‘one that flows out’ Ahtna: Łuu T’aa Ts’iniłeni (upper Klutina River) ‘one that flows out from beneath glacier’ Dena’ina: Ch’aniltnu; Ahtna Ts’anilna’ (Chunilna Creek) ‘flows out stream’

Each of these names has the same verb theme ‘current flows,’ and the same three prefixes and a perfective suffix apply. (In the Dena’ina/Ahtna name for Chunilna Creek, the verb stem len ‘current flows’ reduces to -l- by ellipsis). Ahtna has a large array of landscape and geographic terminology. Many of the geographic terms are ‘areal nouns,’ words that end in a schwa suffix -e. There are many commonly used mono-syllabic nouns and compound nouns with specific meanings: lake, flat, ridge, hill and so forth. Anatomical terms, a great pan-Athabascan specialty, are not prominent in the landscape lexicon. There are the well-established place name conventions that block the occurrence of some landscape terms in place names. Some common landscape terms that are rare or non-existent in the Ahtna geographic names corpus include areal nouns: -baaghe ‘shore,’ hwnene ‘slope,’ -ghak’aay ‘accessible flank on mountain,’ kayax ‘village’ and hwnax ‘house.’ These words can compound with many other terms on an ad hoc basis: c’ena’ baaghe ‘shore of stream,’ dghilaay nene ‘mountain slope.’ Although such terms are commonly used, they do not occur much in the extensive lists of Ahtna place names. Such conventions contribute to the economical distribution of Ahtna place names. There are three broad structural patterns in Ahtna place names: a. About 30% of the Ahtna names are nominalized verbs with the suffixes (y)i (‘the one that’) or den (‘place of ’).



A case study in Ahtna Athabascan geographic knowledge 245

b. About 7% of the Ahtna names are binomial postpositional phrases, involving a noun and the common postpositions, ‘on,’ ‘by’ and so forth. Also, there are binomials NOUN + ADJECTIVE (usually ‘big’ or ‘small’). c. Over 60% of the names in the Ahtna lists are binomials or trinomials where the head of the phrase is the last term – one of the commonly used generic terms ‘river,’ ‘river mouth,’ ‘hill’ and so forth. In Kari (2008:â•›19–22) there is a table of about 50 structural patterns and generic terms that are found in the list of 2208 Ahtna place names. In Table 1 we list the most frequently occurring patterns (those in 20 or more place names). Observe how the Athabascan simplex-complex grammatical dichotomy is reflected throughout the various place names. Note also that two dimensions are represented in every Ahtna place name: (1) the specific name is a “sign” that is analyzable and informative, and (2) an array of generic terms and structural patterns classify the Ahtna landscape. Levinson (2003:â•›69) makes the point that “there are possibilities of generative systems of place names which, as far as we know, are never exploited in societies of simple technology.” However, there is a clear generative geography capacity to the Ahtna geographic system. Specific place names often group into sets or clusters with the generic terms or with directional terms to offer an informative classification to landscapes. The Klutina River is a distinctive drainage, with a stream coming from a lake, which heads in a glacier that leads to a pass over the Chugach Range to Valdez Arm. The Ahtna term for Klutina River, Tl’atina’, has the specific term, or “sign,” tl’ati- ‘rear water.’ For the Klutina drainage there is a set of nine generated place names with generic terms ‘river’, ‘mouth’, ‘lake’, ‘river bank’ (2), ‘headwaters’, ‘glacier’, ‘pass’ and ‘entire drainage’. This is the default pattern for a complex drainage. The generative capacity of Ahtna and Northern Athabascan geography also employs the riverine directional system (discussed in Section 3). For example: Daniits’en Tezdlen Bese’ (upper Tazlina Bluff) ‘upstream side swift-current bank’ Dadaats’en Tezdlen Bese’ (lower Tazlina Bluff) ‘downstream side swift-current bank.’

At this time there are two ways to gauge the generative geographic capacity. (1)€We can count the number of generated names per drainage; for the Klutina River place name list, there are 65 specific names that generate into the list of 98 names; for the adjacent Tazlina River drainage, there are 73 specific names that generate into 139 attested names. Or (2) we can track how sets of names are used in the detailed travel narratives by Ahtna experts. In Kari (2010) the 310 distinct place names are

Ahtna term

Example

-den, de

/174

ca./250 200 for hills, mts., ridges

Ni’aani, Mount O’Neil ‘that which protrudes’

various water or landforms ce’e ‘large’, ggaay ‘small’,  ts’aek’e ‘narrow’, others

Dghelaay Ce’e Mt. McKinley ‘big mountain’

/41

/28

Tez’aani K’ae near mouth of East Fork Chistochina R ‘fishtrap place’

k’ae, k’aet

cavity, hole, depression, place

/24

K’ey Tsaaygha Hogan Hill ‘by the small birch’

gha

by, at, near

/19

Tes K’et hill by Copper Center airfield ‘on the hill’

k’e, k’et, k’edi

on, at a place

C. NOUN + ADJ

Total

Taghaelden, Taral ‘barrier in water place’

B. Common areal nouns or postpositions in place names: NOUN + PP or VERB + PP

landforms, some water -(y)i features: inanimate relative suffix, ‘that which is VERB’

sites, some streams, lakes, hills: ‘at, where, specific place, place of VERB’

A. Common verb suffixes or enclitics in place names: VERB + ENC (without generic)

Geographic term

Symbols: - hyphen indicates possessed form or compound form of noun / left of / = number of “specific” place names; right of / = number of “generic” place names Total number of Ahtna place names: 2208; ** more such names could be generated, increasing the total

Table 1.╇ The most common structures in Ahtna toponyms

246 James Kari

28/33**

Gguus Kulaen Tl’aa upper Kuskalina River ‘celery exists headwaters’

-tl’aa, tl’a-

22/23 47/10 8/19

Saas Dzeł mt. on Boulder Ck ‘sand mountain’ Nay’tsen’staan Tese’ hills at mouth of Raft Ck ‘chunk of meat is there hill’ Nen’ Yese’ ridge between Charley Lake and Lake Louise ‘land ridge’

tes, -tese’

ses, -yese’, -yedze’

hill

ridge

23/2 5/25 /28**

Tsaani ’Aeł Bese’ bluff N of Kotsina mouth ‘bear trap bank’ Sdaa Cii point on E side of middle Tyone L ‘tip of the peninsula’ T’aghes Nuu island below Tasnuna River mouth ‘cottonwood island’ Stl’aa Na’ Ngge’ upper Slana River area ‘rear river uplands’

sdaa, sda-

nuu, -nuu’

riverbank

point, peninsula

island

tene, teni

tinitaan (vb.)

-tates (CLW), łates (U)

trail

trail extends

pass, portage

1/42 /32 /20**

Hwtsiił K’ae Tene trail to weir site at mi. 149 ‘trail to bridge place’ Uk’ese Natinitaan Bene’ Monsoon Lake ‘trail-crosses its-outlet lake’ Tatitl’aa Tates Thompson Pass ‘back water pass’

D3. Cultural or man-utilized features: Common geographic noun roots or nominalized verbs

uplands, drainage of stream ngge’

14/12 34/22

Cetakolyaes Cene’ Monahan Flat ‘things (meat) are brought down flat’

cen, -cene’

bes, -bese’

plain, clearing, flat

mainly U

Dghateni Dghilaay Nenana Mountain ‘stumbling trail mountain’

dghelaay, dghilaay,  -dghelaaye’, -ggalaaye’ dzeł, -dzele’

mountain CLW

5/54

7/63

Tsedi Na’ Łuu’ Chitina Glacier ‘copper river glacier’

D2. Land features: Common geographic nouns or nominalized verbs

93/277

Kaggos Bene’ lake 4 mi. E of Sourdough ‘swan lake’

ben, -bene’

łuu, -luu’, -łuu’

lake

glacier

headwaters

3/116**

Tsedi Cae’e mouth of Chitina River ‘copper mouth’

-caek’e

stream mouth, confluence

/611

-na’

stream (primary generic)

Tsedi Na’ Chitina River ‘copper river’

D1. Water features: Common geographic nouns or nominalized verbs

A case study in Ahtna Athabascan geographic knowledge 247

248 James Kari

generated from 237 specific names (or 23% of the distinct places). Thus, in Ahtna, streams and landforms are treated as natural fractals that vary in shape and can be subdivided into parts. The specific Ahtna names function as signs and can occur in sets. In this way a young Ahtna can learn by the oral tradition a group of specific Ahtna names for streams or hills. When he or she applies the generic terms to these names, more names can be generated. Thus with a core set a basic names for streams and landforms one can memorize and travel around a good section of country. As we discuss in Section 2.4, the Ahtna geographic principles are very similar in adjacent languages such as Upper Tanana and Dena’ina.

2.2

Place names content

Athabascan place names serve as signs. Name content is nearly always analyzable and informative. Over 89% of the Ahtna place names have straightforward meanings, and an astounding 98% of them are fully or partially analyzable. The largest portion of Ahtna names (about 75%) are related to phenomena of natural history such as hydrology, landforms and rocks, and various biota (vegetation, fauna). The portion is similar to that of Dena’ina (Kari & Fall 2003:â•›40). A smaller portion of the names (about 15%) specify human activities (subsistence places, material culture, and human-built structures, trails or human events). Also noticeable is the paucity of reference to weather phenomena (ice, low water, wind). There are only a few modern loan words in Ahtna place names. Moreover, only a few Ahtna names refer to specific events (ten or so). Names like Sc’aen C’anizetde ‘where a child drowned’ (for an incident that occurred in the 1930s) are uncommon. Also uncommon are obviously post-contact names (i.e., after the 1880s). There are a few personal names in the Ahtna place names and only two or three of these appear to be commemorative. Cultural traits such as avoidance behavior and aversion toward self-aggrandizement or cultural grandiosity have affected the content of the Ahtna place names. Furthermore, there is virtually no evidence of any nonAhtna or non-Athabascan presence or influence in the Ahtna place name inventory. Significantly, there are no non-canonical, aberrant-looking names, and no evidence of a substratum of names from another language. Ahtna and Alaska Athabascan cosmography and sense of the sacred deserve more attention and discussion. Overt and covert sacred gestures are embedded throughout Athabascan beliefs and customs; however, gods or supernatural beings are not designated in place names. There are few overtly ritualistic or religious Ahtna place names, and very few overtly commemorative Ahtna place names. It is also noteworthy that there is very little mythic association in the Ahtna place



A case study in Ahtna Athabascan geographic knowledge 249

names (see Section 2.7). Two names do appear to be mythic, but we find that these two places lack known myths. There is routine repetition of names such as ‘water lily lake’, ‘big lake’, ‘round lake,’ ‘narrow lake,’ but there is some patterned repetition that sometimes is overt and distinctive. For example, Klutina Lake and Tazlina Lake are two large Lshaped glacial lakes with streams that drain from the west into the Copper River. It is noticeable that the first sizeable north-side bluffs on each river have the same name (Ts’inahwnet’aaden ‘object that protrudes out’), and that the two ridges at the south-side right-angled points on both lakes have the same name (Tahwghi’aayi ‘one that extends into water’, see Kari 2010:â•›49). Repeated names such as these imply that they were intentionally or strategically chosen as such. A summary of the analyzability of the literal translations of the Ahtna place names is shown in Table 2. For this tabulation I tried to use consistent criteria for grouping the names, looking for opacity, ellipsis, homophony, ambiguity and so forth. Most of the place names are easy to translate and group within the Ahtna lexicon, as presented in Kari (1990), Ahtna Athabaskan Dictionary. Treated as Table 2.╇ Degrees of analyzability: Analysis of literal translations Total (of 2208) & percent Fully analyzable names opaque names  (part or all of name uncertain)

1970/89.2% C’iidze’ Na’, MacLaren River C’ezaeni hill by Crosswind L ’Atna’ Copper River

folk etymologies  Keghiil Na’ Gilahina River (often from one speaker)

no other use or meaning, stem does not match with others

31/1.5%

‘“steps against place” stream’

13/.05% 102/4.6%

speculative (with ‘?..’)

Łet’aes island on Ewan Lake

‘?larch’

ellipsis in names  (truncated, meanings  inferred)

Bendiil Bene’ Tazlina Lake

-diil- not mean34/1.56% ingful < ben dilen bene’ ‘current flows lake’ -ti- similar to tuu ‘water’

Tl’atina’ Klutina Lake homophonous, ambigu- K’ełt’aeni Mt. Wrangell ous (multiple meanings) K’aasi Na’ Oshetna River

Total fully, partially analyzable

? morphemes clear, 43/2% but meaning is not ‘quiver river’ or ‘jagged peak river’ 98%

250 James Kari

opaque are names with stems that do not match with any others, or for which Ahtna speakers do not offer specific meanings. The Ahtna names that are opaque and unanalyzable still look regular and conform to canonical structures. Thus, as we whittle down the corpus, it is striking that 98% of the Ahtna names are analyzable to some degree.

2.3

Place name distribution

Many interesting issues in Ahtna and Athabascan place name distribution remain unexplored. For convenience, in the phases of our GIS mapping, all places have been treated as points. However, if the names are geometrically plotted as points vs. lines, or as large and small polygons, we can see how remarkably comprehensive the naming system is. While there has been some loss of the names (for example in the Chitina River drainage, which has been depopulated by the Ahtna since about 1900), there is thorough coverage of features along the main stem of the Copper River. The Ahtna band territories show a recurrent pattern: (a) a segment of the main stem of the Copper River; and (b) upland lake districts that are mostly west and north of the Copper River (see Reckord 1983 for further discussion). Furthermore, there is information embedded in the Ahtna name system about distances between winter settlements, summer fish camps, game lookouts, fall hunting camps, fall and spring fisheries. By our count there are about 445 named features along the main stem of the Copper River – streams, stream mouths, clearings and flats, and nearby bluffs and hills. Every side-stream of any size above Bremner River has a documented Ahtna place name. All the larger tributaries have full nomenclature for their side streams. The order of these tributaries is the key to the organization of the geography. Also, areas of hindrance that are difficult to access have fewer names, such as sections of the Copper River on the Wrangell Mountains side of the river. The area of the densest coverage of Ahtna names is the region west of the Copper River between the Tazlina River, the lower Gulkana River and the West Fork of the Gulkana River, and the Lake Louise-Tyone Lake area. In this region there are about 505 recorded place names, about 23% of the corpus. This is the logistical hub of the language area and the only subdistrict of the language area that was navigable in summer with the traditional birchbark canoe. There are principles that work against excessive proliferation of names. Streams have a single name for the whole drainage, with main features such as lakes usually grouped with the main name. As noted in Section 2.1, some common landscape terms are not used to form proper place names. The Ahtna do not have high-density, large-scale naming for rocks, sloughs or specific eddies on the



A case study in Ahtna Athabascan geographic knowledge 251

Copper River. Where the environment may have changed due to glacial outwash at a river mouth, the name system would have simply stayed the same. As noted above, Northern Athabascan band territories were large, averaging about 8,000 to 13,000 square kilometers, and the active men typically knew two or three band territories fairly well. Given that the Alaska Athabascan population was small and the band territories so large, an even, low-density distribution of place names is maximally practical and functional.

2.4

Geographic networks among Athabascans

The fundamental features of Ahtna geography are further objectified in the context of the geography being employed by their Athabascan neighbors. As we address the commonalities and regional differences in comparative Athabascan geographic knowledge and territoriality, we can recognize various retentions and variations in generic terms and other structural patterns. Northern Athabascan geography has been a web of interconnected place names networks. Names with similar structural, semantic and distributional properties are interlinked from language area to language area across huge and continuous bioregions. For the Athabascan languages that are connected by trail systems and for which there is good place names documentation, such as Ahtna, Upper Tanana, Lower Tanana, and Dena’ina, we can see dominant patterns and subtle differences in the Athabascan geographic data. Here is a recap of the most salient patterns that have been found. There are many concrete indications that extensive regional networking and travel prowess are reflected in features of Athabascan geographic names. 1. The same set of place names extends across Northern Athabascan language boundaries (with a few rare but interesting exceptions). Speakers who are not acquainted share the same name for mutually known features. A good published example of bilingual naming is the vast interface of Dena’ina and Ahtna in the Talkeetna Mountains, where there are about fifty bilingual place names: Talkeetna River is I’delcuut Na’ in Ahtna and K’dalkitnu in Dena’ina, both meaning ‘food is stored river’ (Kari & Fall 2003:â•›Chapters 10, 11 & 14). 2. There are numerous ways we can demonstrate that Athabascan interregional travel was routine and the trail systems in Interior Alaska are ancient. Corroboration about many aboriginal Athabascan trails can be pieced together from lists of place names, travel narratives and the many historic, cartographic or ethnographic sources related to Copper River. The reiteration of sequences of place names by different speakers as well as by the same speaker strongly confirms specific trails and routes and the overt functionality of

252 James Kari

the memorized names. For the Tazlina Lake-Matanuska River Trail (much like the route of the Glenn Highway) we can cite at least 20 sources that have overlapping confirmation of sequences of names (Kari 2008:â•›11–12 and additions). The Wrangell map of 1839 (eastern portion, discussed in Kari & Fall 2003:â•›85–87) has about twenty recognizable place names and several northerly trail routes that emanate from Upper Cook Inlet for distances of 300 or more miles. This one map demonstrates the scale at which Athabascan experts knew south-central Alaska in the early 19th century. There was general awareness of distant Athabascan territories in aboriginal times. Ethnonyms have served as a small-scale interregional index system. There is repetition among numerous ethnonyms such as ‘the dwellers of the lakes’, ‘the headwaters people’, ‘the downriver people’, ‘the dwellers of the uplands’, ‘the mountain people’. Foot travel and trail knowledge has been facilitated by the memorization of sequences of place names. 3. Athabascan place name content is nearly always analyzable and informative. This is due to the recurrent use of high frequency nouns (e.g. water, rock, soil, colors, common flora) and the common postpositions. There are similar grammatical patterns in place names in Athabascan languages. The patterns reflect the dichotomy between simplex nouns (noun + generic and noun + postposition) and complex verbs (nominalized verbs with common verb themes such as ‘current flows’, ‘linear object extends’, and the common stative-classificatory verbs). The content of the place names often refers to natural history or to traditional activities. Even in distant Athabascan languages in very different environments, such as Hupa in Northern California and Navajo in the Southwest, the content and structure of geographic names are rather similar. There are few non-Athabascan elements in large corpora of place names, other than a few loan word place names. This strong aversion to using other languages or preexisting place names and this high degree of analyzability or purity is indicative of a very robust territorial ethos. Stephen Jett’s (2001:â•›71) comparisons of Navajo and Alaska Athabascan place naming (quoted elsewhere in this volume) underscore the strong similarities in place name meanings and structure between Navajo and Ahtna. 4. When comparing sets of names in adjacent Athabascan languages, we find some noticeable region-marking traits. There are seven mutually exclusive hydronyms in Northern Athbabascan (Kari 1996a). The hydronymic districts are the most overt and salient pattern we have detected in Northern Athabascan geographic naming. Although we need to revisit the hydronymic data, this general distribution of Northern Athabascan hydronyms is valid. For example, in eight western Alaska Athabascan languages (including Ahtna) the



A case study in Ahtna Athabascan geographic knowledge 253

term *-na’ is used for streams, whereas in four eastern Alaska Athabascan languages *-niq’ә (lit. ‘on the upstream’) is used. Several Alaska languages even have primary vs. secondary hydronyms which give a ‘street’ vs. ‘avenue’ effect to stream names (Kari 1996a, 1996b). In addition to the striking patterns for ‘stream,’ other regional patterning in the generic terms deserves closer examination. There is a patterning in generic terms for ‘mountain’ in Central and south-central Alaska (Kari 1996b) that is conspicuous. The Ahtna term dghilaayi ‘mountain’ is shared with Dena’ina, Fig. 7: Hydronymic districts in Northern Athabascan languages (Kari 1996c)

Figure 3.╇ Hydronymic districts in Northern Athabascan languages (Kari 1996a) . Niclas Burenhult comments that the ‘street’ vs. ‘avenue’ principle applies also in Jahai (Austroasiatic, Malay Peninsula) hydrological terminology (Burenhult 2008).

254 James Kari

Upper Kuskokwim and Kuskokwim Ingalik. This term has been extended to mean ‘mountain’ from a Tanana Valley Athabascan term for ‘Alaska Range’. Proto-Athabascan *yәs ‘ridge’ vs. *dzәł ‘mountain’ may have significant patterning in some Alaska languages. There is a marked contrast in the use of *-tu’. In five western Canadian languages (such as Southern Tutchone and Kaska) it means ‘river’, but in six Canadian languages east of the Mackenzie River (such as Hare, Slave, Dogrib and Chipewyan) *-tu’ applies to the larger lakes. On the other hand, 21 other Northern Athabascan languages use *wәn for ‘lake’.

3.

Riverine directionals

It is not possible to discuss here several other Ahtna grammatical features and word categories that mark space, directions and orientation. These include (a) a deictic/demonstrative system; (b) a large set of postpositions; and (c) verbal prefixes and suffixes of space and orientation and direction. However, it is important to introduce the Ahtna (and Athabascan) riverine directional system. The Northern Athabascan language groups are oriented to the major rivers. The riverine directionals are pervasive and are the intersection between the geography, the lexicon, and the grammar. The major rivers often have totally different geographic axes, such as the Copper River (which flows in an arc north to south) versus the Tanana River (which flows east to west). The Athabascan riverine directionals are what Levinson (2003:â•›90) terms an “intermediate absolute landmark” frame of reference. Northern Athabascans do not make use of north-south-east-west for orientation, and terms for ‘left’ and ‘right’ are not used at all for spatial orientation or in geographic terminology. The riverine directionals system in Ahtna and in the other Northern Athabascan languages is so pervasive that it constitutes an organizing “semplate”, a semantic template, in terms of the semantic theory of Levinson and Burenhult (2009). Riverine elements are used in all speech registers in indoor and outdoor settings, and they appear in distinct word categories such as the outer (disjunct) verb prefixes, the noun lexicon (e.g. parts of houses or boats and especially place names), and postpositions. They are indicated by gestures and body movement (a great topic for further study). In Ahtna and other Northern Athabascan languages the directionals are a special word category with prefix+root+suffix structure. This structure resembles the verb complex in miniature and typically occurs in over sixty derived forms. Table 3 displays the morphology of the Ahtna riverine directionals with the prefixes and suffixes that combine with the set of nine roots. Also see Table 4 below. With the plain adverbials (without prefix or suffix) and the combinations of prefixes, roots and suffixes, there are 69 possible directional terms. Expert



A case study in Ahtna Athabascan geographic knowledge 255

Table 3.╇ Ahtna riverine directional structure Meanings

Plain ADV

5 prefixes

Bound 5 suffixes: -Ø, -xu, -t, -dze’, roots -ts’en

across

naane

ts’i- ‘straight’

-naane

-naane, -naaxe, -naat, -naadze’

downstream

daa’

’u- ‘far’

-daa’a

-daa’a, -daaxe, -daat, -daadze’

upstream

nae’

da- ‘close’

-n’e

-n’e, -nuuxe, -niit, -niidze’

lowlands, toward  water

tsene

ka-  ‘adjacent, next’

-tsene

-tsen, -tsuughe, -tsiit, -tsiidze’

uplands, upland area, from water

ngge’

na-  ‘intermediate’

-ngge’

-ngge’, -nggu, nggat, -nggadze’

up

tgge’

P+gha  ‘in relation to’

-tgge’

-tgge’, -tggu, -tggat, -tggadze’

down

yax

-ygge’

-ygge’, -yggu, -yggat, -yggadze

outside, beyond,  other side

’ane’

-’ane’

-’ane’, -’aaxe, -’aat, -’aadze’

forward, ahead,  front, toward fire

nse’

-nse’

-nse, nse’, -nsghu, -nset, -nsedze’

Totals

9 + 9 × 6 × 5 = 69 plain or derived directional forms

speakers can tap into all of these fine distinctions. It is challenging to translate the precise meanings of derivative directionals, short of having a morphemic gloss for every word: natsiit ‘close + downland + at specific place’ ’unuuxe ‘far + upstream + in general area’

Suffixes can reverse the path of directionals: kanggadze’ ‘from the next place upland’ contrasts with katsene ‘towards the next place downland’. Also, there are morphophonemic alternations typical of the verb stem variants, for example tsen€+ t → tsiit, ’an + xu → ’aaxe, or n’e + t → niit. Aside from their morphological composition (for Ahtna, Kari 1985, 1990:â•›633; for Koyukon, Jetté & Jones 2000:â•›808–810), there has been very little directed elicitation or detailed research on riverine directionals in any Northern Athabascan language. The Athabascans, by any comparison, are among the world’s foremost pedestrian foragers. In Ahtna and other Northern Athabascan languages, the directionals provide a remarkably comprehensive frame of reference, one that must contribute to every aspect of orientation, way-finding, or travel. In Figure 4 we summarize the semantic oppositions and applications of the directionals.

daa’ “downstream”

-tgge’ “up”

Figure 4.╇ Ahtna directionals, composite view

ross

-tsene “downland”

nii “rear, back”

5. anatomy, intrinsic shapes

-ngge’ “upland”

daa ” “downstream

-nse’ “ in front, out in local range”

-ngge’ “upland, to shore”

nse’ “out in laket”

4. lake, large stream

nae, niit “upstream”

e ac naan

3. in view, in the landscape

house drawings by Alan Boraas, photo by James Kari of upper Slana River, looking south toward Indian Pass

‘ane’, -aaxe’ “outside”

nse’ “to fire, in front”

nae’, niit “upstream (wall)”

nae, niit “upstream” -tsene “downland”

naan “across” -tsene “download, front wall” ygge “down”

-ngge’ “upland, back wall”

(wall)”

daa’ “downstream

2. house interior

yax, -ygge’ “down”

-ngge’ “upland”

dak, -tgge’ “up”

1. house (entrance toward river)

nse’ “front”

‘ane, -’aaxe’ “beyond, to other side”

256 James Kari



A case study in Ahtna Athabascan geographic knowledge 257

As shown in Figure 4, interestingly, the nine roots are mainly in orthogonal oppositions; that is, they are mutually perpendicular. Seven of the nine roots have the same orthogonal oppositions, both indoors and outdoors (in the local landscape), as in settings 1, 2 and 3 in Figure 4: across::across, upstream::downstream, upland::downland, and up::down (vertically). The entrance to the traditional Ahtna house faced the local stream, so the front wall is ‘downland’ and the back wall is ‘upland.’ The root nse’ is the most semantically complex directional root. In Figure€4, nse’ occurs in three pairs of oppositions. The hearth was in the center of the house, and persons face the fire in both indoor and outdoor settings and at three distinct scales. Both inside and outside, the position of the hearth is the semantic thread (settings 1, 2 and 3) for nse’ opposed by ’ane.’ In Figure 4 for intrinsic shapes (boats, any shaped object) and anatomy, nse’ is ‘front’ and n’e, nii is ‘back’, giving the dual meaning ‘upstream; back’ for n’e, nii. Table 4 is a summary of the pervasive riverine directionals in Ahtna lexicon and grammar, evidence that the directionals are a powerful semantic template or “semplate” in Ahtna (and other Northern Athabascan languages). While numerous derivational prefix strings have meanings that parallel those of the riverine directional system (as shown in Table 3), note that most of these strings contain morphemes that are unrelated to the nine basic directional roots. Directionals are commonly used in all registers and genres of Ahtna speech. I use the term elite travel narrative to refer to connected text language that describes with precision sequences of places or travel in a region that the speaker knows well. In detailed travel narratives we see the orchestration of all of the grammatical or lexical spatial and orientational features of the language. The five travel narratives by five Ahtna experts (Kari 2010) are “walking tours” of traditional Ahtna lands. The places that are named and the routes that are traced are mapped fairly closely and translated with consistency. Texts about places and travel are an important complement to place names research. The interaction between the geographic names and the directionals is highly complicated. All nine of the directional roots are attested. In the future it may be possible to achieve a better understanding of how directionals frame the landscape for speakers of Ahtna and Athabascan. Having a good corpus of mapped Ahtna travel narratives is an essential step toward a more formal study of the use of geographic names and directionals in the language. Consider the raw facts about this selection of Ahtna travel narratives. In the five chapters and in just over 102 minutes of speech, nearly 1200 miles of trails and routes are described; Ahtna names for 310 different places are mentioned 698 times by the same narrator. The precise riverine directional terms are used 569 times in two segments of 45 minutes. Jim McKinley, the narrator of one of

258 James Kari

Table 4.╇ Summary of uses of riverine directionals 1. Directionals used pragmatically to specify direction and space in all settings and genres a. Directionals in indoor settings, and in yenida’a legends NB. Ahtna yenida’a myths with human-animal interaction always lack place names or any local geographic references. The collection of yenida’a stories by Jake Tansy (1982) can be considered as pure fiction; the stories contain no place names. b. Directionals used in the landscape description, in conjunction with place names  (see Section 4). 2. Lexicalized directionals in place names (several types) a. DIR + i ‘the one in DIR’: Natsiidi ‘the lowland one’ Devils Mountain b. DIR (specific name) + GENERIC: Hwdaadi Na’ ‘downriver river’ Dadina River c. DIR + place name (to create sets of names): Hwtsuugh Naknelyaayi ‘lower ridge-that-extends-across’mt 4716’ S of Denali Highway Henggu Naknelyaayi ‘upper ridge-that-extends-across’ mt above 4716’ d. place name + DIR: Tsedi Na’ Ngge’ ‘copper river uplands’ Chitina River drainage 3. Nouns, adverbs formed with directionals nsedze’ niidze hwt’aene dasts’eni ’unsedi ke’ c’eghanggaden c’eghantsiiden

toward spring ‘to the front, ahead’ Nabesna-Chisana people ‘people from upriver’ southwest wind ‘the one on the forward side’ front legs, lit. ‘front feet’ first (older) wife ‘the one in the uplands’ second (younger) wife ‘the one in the lowlands’

4. Derivational verb prefix strings with meanings that parallel the directionals ko+ce#d+i (0 mom) s# (gh mom) ko+tsi#d+i (i mom) ta# (ss mom) dak/tgge’ n# (s rev)

‘up river, extending, going upriver’ ‘downriver’ ‘descending to water, to lowland’ ‘up from shore, upland’ ‘up vertically, standing up’

the segments, mentions 116 different place names 375 times for about 280 miles (450 km) of trail routes, and he uses 231 directional terms. In one 3:08 minute segment, Jake Tansy, the second narrator, gives a detailed set of alternate routes of about 125 miles (200 km) between the Nenana River and Yanert Fork and back to the Upper Sustina River, mentioning 23 place names 36 times along with 46 directionals. Redundancy and the reconfirmation of geographic particularism are hallmarks of Athabascan geographic knowledge. Several types of repetition or



A case study in Ahtna Athabascan geographic knowledge 259

redundancy can be defined and examined in a corpus such as the one in Kari (2010): repetition of place names (the same name for different features); corroboration by different speakers of names for the same feature; or the relative salience vs. the obscurity of places.

4.

Concluding remarks

For over thirty years I have been collecting lists of Athabascan place names and making many of the same generalizations about Athabascan geographic knowledge. I continue to be amazed at the fact that this is a shared, verbally transmitted system that is congruent across language and dialect boundaries. For Ahtna we can marvel at the strict purity, orderliness, symmetry, functionality, and memorizability of the geography. When we compare Athabascan geographies, some of the more salient patterns are the congruence in names across language boundaries, the overt region-marking traits such as the hydronymic districts, occasional boundary-marking place names, and some striking patterned repetitions in place names. The riverine directional system interacts with the place names to provide a very precise frame of reference. Since antiquity, the principles and mnemonic devices embedded in place names and directionals have facilitated geographic perception and efficient foot-travel throughout and beyond Ahtna territory.

References Allen, Henry T. 1887. Report of an Expedition to the Copper, Tanana, and Koyukuk Rivers in the Territory of Alaska in the Year 1885. Washington DC: U.S. Government Printing Office. Black, Lydia. 2008. Commentary on journal of 1796 by Dmitrii Tarkhanov. In Anooshi Lingit Aani Ka, Russians in Tlingit America, The Battles of Sitka, 1802 and 1804, Nora Marks Dauenhauer, Richard Dauenhauer & Lydia Black (eds), 67–90. Seattle WA: University of Washington Press. Burenhult, Niclas. 2008. Streams of words: hydrological lexicon in Jahai. Language Sciences 30: 182–199. Jett, Stephen C. 2001. Navajo Placenames and Trails of the Canyon de Chelly System, Arizona. Frankfurt: Peter Lang. Jetté, Jules & Jones, Eliza. 2000. Koyukon Athabaskan Dictionary. Fairbanks AK: Alaska Native Language Center. Kari, James. 1985. A note on Athabaskan directionals. International Journal of American Linguistics 55: 424–455. Kari, James. 1989. Affix positions and zones in the Athabaskan verb complex: Ahtna and Navajo. International Journal of American Linguistics 55: 424–455.

260 James Kari

Kari, James. 1990. Ahtna Athabaskan Dictionary. Fairbanks AK: Alaska Native Language Center. Kari, James. 1992. Some concepts in Ahtna Athabaskan word formation. In Morphology Now [SUNY Series in Linguistics], Mark Aronoff (ed.), 107–133. Albany NY: SUNY Press. Kari, James. 1996a. A preliminary view of hydronymic districts in Northern Athabaskan prehistory. Names 44: 253–271. Kari, James. 1996b. Names as signs: the distribution of ‘stream’ and ‘mountain’ in Alaskan Athabaskan. In Athabaskan Language Studies, Essays in Honor of Robert W. Young, Eloise Jelinek, Sally Midgette, Keren Rice & Leslie Saxon (eds), 443–475. Albuquerque NM: The University of New Mexico Press. Kari, James. 2008. Ahtna Place Names Lists, 2nd edn., revised. Fairbanks AK: Alaska Native Language Center. Kari, James. 2010. Ahtna Travel Narratives, a Demonstration of Shared Geographic Knowledge among Alaska Athabascans. Fairbanks AK: Alaska Native Language Center. Kari, James & Fall, James A. 2003. Shem Pete’s Alaska: The Territory of the Upper Cook Inlet Dena’ina. Fairbanks AK: University of Alaska Press. Levinson, Stephen C. 2003. Space in Language and Cognition. Cambridge: CUP. Levinson, Stephen C. 2008. Landscape, seascape and the ontology of places on Rossel Island, Papua New Guinea. Language Sciences 30: 256–290. Levinson, Stephen C. & Burenhult, Niclas. 2009. Semplates: A new concept in lexical semantics? Language 85(1): 153–174. Reckord, Holly. 1983. Where Raven Stood: Cultural Resources of the Ahtna Region [Occasional Paper No. 35]. Fairbanks AK: Cooperative Park Studies Unit. Tansy, Jake. 1982. Indian Stories: Hwtsaay Hwt’aene Yenida’a, Legends of the Small Timber People. Anchorage AK: NBMDC. Reprinted in 1997 by Ahtna Heritage Foundation. Vajda, Edward J. 2010. A Siberian link with Na-Dene languages. In The Dene-Yeniseian Connection [Anthropological Papers of the University of Alaska 5], James Kari & Ben Potter (eds), 33–99. Fairbanks AK: University of Alaska.

chapter 12

Revitalizing place names through stories and songs Susan Paskvan

Native place names research conducted near Kaltag, Alaska, reveals that the Koyukon Athabascan have an intimate relationship with the land. The land, its resources and its inhabitants are examined through the etymology of the place names. Most Koyukon Athabascan place names are descriptive, are binomial, and can be classified into three semantic categories. Oral histories about place names reveal beliefs, stories, and songs of the people. Challenges and opportunities in passing on this knowledge to the younger generation are explored. Several Athabascan language revitalization efforts are underway that combine indigenous teaching strategies and modern technology.

1.

The land and people of the Kaltag area

Thirty-five Athabascan languages were spoken in an area that extends from interior Alaska through the northwest coast and into the desert southwest. In Alaska, denaakkenaage’ ‘our language,’ commonly known as Koyukon Athabascan, is spoken in Interior Alaska along the Koyukuk River, the middle Yukon River, and Tanana River. It is comprised of three dialects: Upper, Central and Lower Koyukon. The area of this study is situated in the Lower Koyukon village of Kaltag. Kaltag is located on the middle Yukon River, with a current population of approximately 175 people. Prior to the settlement of the village in the early 1900s, people were nomadic, moving from camp to camp following the food sources. These speakers of the Lower Koyukon dialect occupied the land in three regions: a section of the middle Yukon River, the Kkaayeh flats region and the Kaltag Portage. They referred to themselves by the area they lived in, i.e. Tlaamaas Hut’aane ‘people of the Tlaamaas’ (tlaamaas, a semi-circular knife, is commonly known in Alaska by the Eskimo word ulu); Yookkene Hut’aane ‘people of the Yukon’; and Kkaayeh Hut’aane ‘people of the Kkaayeh’ (kkaayeh means village)

262 Susan Paskvan

(Michael 1967:â•›242; Jetté 1910:â•›77; Osgood 1936:â•›14; Loyens 1966:â•›17). In general, the people referred to themselves as tleeyegge hut’aane ‘local people’ or denaa ‘man, human’. Kaltag is located along the Yukon River, named Yookkene ‘big river’ in Koyukon Athabascan. The Yukon, one of the largest rivers in North America, is approximately 1 1/2 miles (2.4 km) wide near Kaltag. On the north side, the area is primarily hills that rise up 1,500 feet (460 m); in the valleys are tributary rivers, creeks and lakes. Large islands divide the Yukon. The greatest resources of the Yukon are the Chinook, Chum, and Coho salmon, which begin to arrive in early€June. The boundary between the Lower and Central Koyukon dialects is midway between the villages of Koyukuk and Nulato. The boundary between Lower Koyukon and Holikachuk Athabascan is at Sołtoł Denh (etymology unknown), commonly known as Blackburn, a fish camp approximately 91 miles (146 km) below Kaltag. The Kkaayeh flats region, commonly known as Kaiyuh Flats, is on the south bank of the Yukon with the headwaters of Kaiyuh Slough near Pilot Mountain (a small hill located between Galena and Koyukuk), flowing westerly where it connects with the Hotolno’ River. The mouth of the Hotolno’, Khotol River, is 22 miles (35 km) below present-day Kaltag (Sullivan 1942:â•›40). During the winter, bands of people lived throughout this vast area of lakes, rivers, and sloughs hunting and trapping. The first explorer in the interior of Alaska was the Creole Russian explorer Andrei Glazunov (Glasunoff, Glazanof), who traveled from the St. Michael (Mikhailovskiy Redoubt, Fort St. Michael) over the portage to Anvik River and was the first undisputed European to reach the middle Yukon. He told Zagoskin (Michael 1967) he traveled as far as Ttutago, which made him the first explorer to travel to the Koyukon territory. Following Glazunov was Creole Petr Vasilevich Malakov (Malakhof, Malakhov), who traveled overland from the Russian trading post Unalakleet (Unalaklik, Unalachleet) on the Norton Sound to Hudokkaakk’et (Kakhokgokhakat) on the Yukon River, then continued upriver to the village of Nulato (Nulagito), then upriver to the Koyukuk (Kuyukak, Yunnaka, Co-Yukuk, Koyukuk) river in 1838. His exploration is deemed as entry into one of the last unexplored regions of North America. Malakov traveled back downriver the following spring, becoming the first explorer to travel the stretch of river from Nulato to the mouth of the Yukon. The Kaltag Portage is located in the Nulato Hills, on the north bank of the Yukon, with streams on the east side flowing to the Yukon and streams on the west side flowing to the Norton Sound coast (Wahrhaftig 1965:â•›28). The boundary between the Koyukon Indians and the Eskimos of Unalakleet was at the small



Revitalizing place names through stories and songs 263

Map 1.╇ Area location of place names near Kaltag, Alaska

village of Edemełek Denh, Chiroskey River, 70 miles (113 km) from Kaltag and 20 miles (32 km) from Unalakleet (Jetté 1910:â•›77). The original inhabitants of the coastal area near present-day Unalakleet were Unalit (Degnan 1999:â•›55). After a smallpox epidemic in 1836, the Malymiut Eskimos, Inupiaqs from the north, moved south. The dialect of Inupiaq spoken today in Unalakleet has traces of Yup’ik prosody. The Koyukon place names extend to the coast and a few places along the beach. Trading systems were well established, especially between the Eskimos of Norton Sound and the Athabascan bands living along the KaltagUnalakleet portage (Michael 1967:â•›136).

2.

Place names

“Living off of the land,” “subsistence,” “our culture” are all phrases that Alaskan Natives use to describe their relationship with the land. In the summer Kaltag families move to fish camps along the Yukon; in the fall they travel to the Kaiyuh for moose and to nearby hills for berries; in the winter they travel to trapping camps and in the spring to ratting (muskrat) camps. A common safety practice is to let someone know when and where you are planning to travel. They tell each other in Denaakk’e (‘like us,’ the word used for the Koyukon language) where they are going and which trail they are traveling on. The Denaakk’e place names are interspersed throughout the conversation. These place names are mnemonic devices that “are symbolic and evocative of the relationship which we have had

264 Susan Paskvan

with the land for generations and generations” (Jones 1985). Through oral history, place names in the Kaltag region have been passed down for generations as people traveled through areas, told stories, sang songs, and shared their family and community histories. Early anthropologists recognized that place names were an important part of a culture’s identity. The father of American anthropology, Franz Boas, shifted from hard sciences to the more humanistic field of ethnology as he discovered that indigenous place names held the key to cultures’ geographical knowledge and cosmography (Thornton 1997:â•›3). Edward Sapir, Alfred Kroeber, Samuel Barrett, John Harrington and Thomas Waterman, students of Boas, contributed significantly to the field. Sapir, who developed the proto-Athabascan theory of Na-Dene languages, promoted a humanistic approach to studies of the environment through studies of the indigenous nomenclature, believing that these references held clues to how the world is perceived (Thornton 1997:â•›4; Basso 1996:â•›43). Thomas Waterman developed a cognitive theory of how place names can be classified into semantic categories, such as myths, animals, food supply, and human activity. Documentation of indigenous place names in the Kaltag area began with the early Russian explorer Lt. Zagoskin in 1843. His journal was initially published in Russian, with a translated English version in 1967. Zagoskin avidly wrote down information on the geography, the people, language and culture, the camps/villages, climate, and travel conditions. His journals provide rare insight into the ethnography of the first decade of European contact with the indigenous people. The next exploration team in 1866–1867, the Western Union Telegraph Company, included Frederick Whymper and William Healey Dall. During this trip, they learned that Russia was sold to the United States in 1867. The Americans re-established the trading post at Nulato, but didn’t keep it running continuously. In the period of 1867–1887, increased exploration into the interior began. Lt. Allen explored the Copper River, Tanana and upper Koyukuk Rivers in 1885. The gold exploration expanded into this new territory in the mid-1880s. Missionaries also began traveling on an intermittent basis along the rivers. A church was built at Nulato 1888, a year after a Roman Catholic priest, Fr.€Tosi began living there (Loyens 1966:â•›117). Renowned priest and ethnographer, Fr. Jules Jetté, traveled extensively throughout this region. He learned the language and documented all aspects of the culture. His work can be divided into four phases over a 29-year period. Jetté’s academic documentation of place names (Notes on Geographical Names of the Ten’a) throughout the interior provided the groundwork for this paper. He documented the location, etymology, literal meaning, and any pertinent cultural information on each place name. My mother, Koyukon linguist



Revitalizing place names through stories and songs 265

and fluent speaker, Eliza Jones, became fascinated with Jette’s manuscripts and began her own documentation of place names throughout the interior. In 1986 she interviewed speakers at Kaltag, reviewing Jetté’s list of place names in this area, updating the orthography, documenting the locations on maps, and recording the stories. My own work began in 1999, with field visits to Kaltag beginning in 2003 and subsequent interviews of elders in 2004 and 2005. This corpus of data will eventually be published as part of my Masters of Arts thesis. Kari and Fall (2003:â•›40) divide the semantic typology of Dena’ina place names into three different categories: Natural History, Human Activities, and Other. This analysis has been useful to my examination of Koyukon place names. Some examples of the classifications include: 1. Natural History: resources such as animals, fish, birds and geographical features such as terrain, geology and weather. Examples: Ggaagge Kk’aateyet ‘animal habitat’ for animals and Tlaa Tso Tsuh ‘big space behind the rock island’ for terrain and geology. 2. Human Activity: things of material culture, food, places of customary activity and historical events. Examples: Tso Negge ‘behind (inland) of the cache’ is an example of material culture. Daa’oghe Yeh ‘place where people traditionally used to get flat rocks for tanning’ is an example of customary activities. 3. Other: mythological events, unanalyzable names and unclear class. Examples: Nełe K’ehudegheetotl Hu ‘place where the ground has been trampled over,’ creating a fairly good trail or travel area through the portage. The trail was created by Yuh K’etseeyh ‘you guys’ grandpa,’ which refers to the Great Raven, the creator of the universe in Koyukon beliefs. Kari and Fall (2003:â•›37) found that Athabascan place names follow a binomial pattern: ‘specific + generic,’ with the generic part classifying the geography and the specific part revealing information about their environmental or cultural resources. Some examples of generic words include tlaa ‘bluff’; le’on ‘rock’; yeł ‘mountain’ and noo ‘island’. Specific words could be nouns, such as ‘rabbit,’ or a description, such as size or a directional. In the following example, specific directionals are used to distinguish two bluffs: Tlaat’ot K’uhuneets’en ‘beneath the bluff on upstream side’ and Tlaat’ot K’uhudots’en ‘beneath the bluff on the most downriver side’. The descriptive quality of place names can yield a mental picture of the actual site. The Koyukon verb structure is polysynthetic, in that one word is composed of many morphemes in a structured order and a complete sentence may be contained in a single verb (Thompson 1977:â•›14). The verb is composed of a stem, classifier and prefixes. A slight change in one of the prefixes can alter the meaning to evoke a different quality. In the following example, a nominalized

266 Susan Paskvan

Neełts’aadodelenh Denh

Figure 1.╇ Neełts’aadodelenh Denh. Source: Google Earth 2010

verb evokes such a place. In 2004, consultant Mary Rose Agnes (2004) described a section of the Kkaayeh Slough “just like it forks”.

Neełts’aadodelenh Denh Neeł - ts’aa # de – ghe D – lenh Denh forking apart out # (extension theme) forms a current place where (aspect derivational string) ‘place where creek separates’

The slough is located at the cutoff point of a soon-to-be oxbow lake where it is almost completely merged upon itself. In addition to the descriptive place names, the use of directional verbs provides an insight into the cognitive mapping of the Athabascans. Living on the river, they orient themselves based on the flow of the river in reference to their location, and noting the distance. This riverine directional system is also used for describing location inside the home: back away from the river, upriver side, downriver side etc. (Jetté & Jones 2000:â•›808).



3.

Revitalizing place names through stories and songs 267

Creating a sense of place

Place names as mnemonic devices illustrate the rich relationship the people have with the land. Noted ethnographer Keith Basso was encouraged by his consultants to “learn the names” on the Western Apache reservation at Cibecue, Arizona. Basso discovered that storytelling with place names has a direct impact on the extent to which an individual follows the Apache moral values (Basso 1990). Inevitably, something bad happens to someone at a site if they did not abide by the moral values. When someone commits a violation or similar behavior, a caring relative relates the story of what happened at that place. In this indirect fashion, a moral lesson is conveyed. The impact of just saying the place name evokes the “particular physical setting in which listeners can imaginatively situate everything that happens” (Basso 1990:â•›117). The Koyukon Athabascans also use storytelling as a tool for behavior modification. Koyukon Athabascans have an intricate relationship with the spirit world, the land and the resources. The relationship with the environment is aligned on two interconnected levels: first, an empirical knowledge that results from living and relying on the resources, and second, a spiritual knowledge that guides one’s behavior through “an array of supernaturally based rules that ensure the well-being of both humans and the environment” (Nelson 1983:â•›15). Activities that cause one to lose one’s luck are avoided. If someone does something that offends the proper behavior, an elder would say “hutlaanee” to let him or her know that their action was against the ways and beliefs. Hutlaanee roughly translates as ‘taboo’ or ‘spiritual violation.’ These beliefs originated during the huloyh ‘story’ era (or kk’edonts’ednee ‘a time of long ago’ in Central Koyukon dialect), which describes a primordial time when animals, birds, fish, etc. were humanlike, able to carry on conversations and relationships like people. An animal’s personality or physical characteristics today reflect the type of a person it was during the huloyh era. For example, the tl’eghes ‘burbot’ (lota lota), locally known as loche, was a thief in the huloyh times. Some of the bones in his body are shaped like the articles of items he stole such as a comb, a caribou antler, etc. The beliefs are passed down through oral history stories that are primarily told during the darkening winter months. The stories tell the origin of the world and people, much as Genesis in the Christian Bible (Attla 1996:â•›v). “Our Native beliefs are inside those stories,” Eliza Jones is quoted in a newspaper article. “It is like gospel to us. It is very much a part of my belief in living in harmony with nature, with the land, trees, water, animal and bird spirits” (Smetzer 2000:â•›A-1, A-7). In the following huloyh, the creator, Yuh K’etseeyh ‘you guys’ grandfather’, made physical changes in the landscape on the Kaltag Portage. The story takes

268 Susan Paskvan

places at Nełe K’uhudegheetotl Hu ‘place where someone has trampled over the area’. Consultant Franklin Madros, Sr. (1986) told this story in a tone of voice that reflected his awe of the Raven’s powers to shape the land. Go doogh eesee Nełe K’uhudegheetotl Hu. This is the area where hemsteek’aal ‘grandfather’ trampled out a trail. Hugho’eenaaltsonh eey go yet. He got very tired at that time. Eey nonłe gheholee eey go hemtseek’aale. The one who was walking out to the coast, their old grandfather. Dleł te ehoye ts’uh daats’e de’eneeyh. He goes into the mountains and does this.

The Great Raven separated the hills and made a smooth portage trail. Consultant Mary Rose Agnes (2004) said that Yuh K’etseeyh was looking at the trail, like it was all mixed up. He said: “Gen nedaaz dohoot’aa hu daatlen neełtehunaaldukk kk’e dehoot’aa hu?” What’s going on down (or out) this way that it’s all tangled up?

Agnes gestured by holding her upraised arms together, then blowing while opening her arms. The verb neełtehunaaldukk describes the trees or brush all tangled up (Jetté & Jones 2000:â•›152; Jones, personal communication, 2004). Jones explained that the use of nonverbal communication is an important part of storytelling, especially when describing major events – such as the Great Raven separating the hills and forest and creating the trail (Jones, personal communication, 2004). In addition to huloyh two other types of stories are told: yoogh done ‘long ago’ stories and kk’odeet ts’ednee ‘modern stories’ (Jetté & Jones:â•›2000). It is the yoogh done stories that are directly associated with physical places. Basso (1990) emphasized the importance of the individual relationship each person has with places. This trait is most prominent in travel narratives, which reflect intimate knowledge of the land as a result of slow travel by foot, canoe, and/or dog team. The following yoogh done story is similar to those Basso found in that there is a moral. The story takes place at Taamaa Ts’e’oyh Denh, which possibly means ‘beach where we customarily set a fish trap.’ A young couple started to live together in her parents’ home. Suspecting that the young man was lazy, the father told his daughter to have the young man go to this place: . It should be noted that all of the consultants participating in this research were the last generation to live year-round in seasonal camps.



Revitalizing place names through stories and songs 269

“Nugh nodegge, Taamaa Ts’e’oyh Denh, taa’aan noneenle’aanh medeenee nugh nugh ledonenh,” deden’aa’ ełnee. “Dad (the person near you) said to tell you to check the fish traps up there at Taamaa Ts’e’oyh Denh,” she apparently told him. Eey go keel haadeeyo. Naadegge ghehoł, naadegge ghehoł. Go Taamaa Ts’e’oyh Denh neeyo. The boy went out. He walked upland. He walked upland. He arrived at Taamaa Ts’e’oyh Denh.

When he returned to the house, he complained that it was a long way up there. The father got mad and told his daughter to tell the man to leave. The young man had to leave. Stories of conflict within families like the one of the lazy man are noteworthy because the moral of the stories teach the Athabascan values, such as the value of hard work and self-sufficiency. The late Charlie Brush, a respected storyteller, tells how his dad and uncle began telling him stories when he was a child, but he did not appreciate the lessons until later in life. They told him “We don’t say this for ourselves. This is for you, for your own good that we talk this way to you” (Brush 1984). He learned valuable lessons from the stories as he matured. Knowing the physical locations where the stories take place and the associated place names connects people to the land and provides a cognitive map of their history. In addition to the stories, the Koyukon Athabascans have a rich singing tradition that reflects a close relationship with the land. There are several types of songs: animal and bird songs; lullaby songs; love songs; mourning songs; and dancing songs. Some of the animal and bird songs are a part of the huloyh stories. The lullaby and love songs are personal between two people. The mourning and dancing songs are memorial songs that are composed in memory of a person who died and are sung at a ‘Stickdance’, which is a memorial ceremony held a few years after the death of a loved one. The Stickdance is only held in the two lower Koyukon villages of Kaltag and Nulato. In a Stickdance ceremony, place names are indirectly referenced in the introduction of the song and in the order that the songs are sung. The singer introduces the song by naming the composer, where he/she was from and who the song was made for. Consultant Franklin Madros (personal communication, 2004) sang the songs based on where the composer lived, starting with Kaltag, upriver to Nulato and Koyukuk. So many times when we were working together, Madros would begin singing a song, then he would say “Gee whiz, I better tell you a story about that”. He always mentioned the place where the song came from. . Notice how emphasis is placed on the distance by repetition of the sentence.

270 Susan Paskvan

The following excerpt is from the first song that Franklin Madros used to sing. It is by Alexie Pitka for Grandpa Oldman Stickman, one of three brothers who moved to Kaltag from Dishkaket on the Innoko River (Madros, personal communication, 2004). It has been translated by Eliza Jones, who stated that “it is probably not perfect but I tried” (Jones, personal communication, 2006). The lyrics composed by singers are often poetic, like riddles, especially those composed by fluent singers who knew the “high” language, which is similar to college-level English (Jones, personal communication, 2007). Hodee do’ots’e k’edenh ne nen’ kk’ets’en’ huts’enh hots’eghelneeyh kk’e dets’eeloh ghee denh Recalling (memories) coming from over a way, a different place, land, we gather our treasures together like we did when he was living.

In this lyric the composer refers to Grandpa Oldman Stickman as coming from Dishkaket, which is part of the Holikachuk Athabascan territory. The treasures refer to the furs that he trapped, the tl’eeyegge baabe ‘our Native food’ that he caught or gathered, and the stories and songs he shared. In Kaltag, these memorial songs are sung at a semi-annual Stickdance, a week-long ceremony comprised of food potlatches, traditional dancing, and gift giving. When everyone is singing and dancing together in unison the spirit of the community is powerfully lifted. An elder from Minto, Neal Charlie, said that when he teaches he starts with a song, tells a story, sings another song, gives some advice, then sings another song (Charlie, personal communication, 2010). This is an example of indigenous teaching methods that Charlie learned from elders a long time ago. In addition to memorial songs, there are songs or stories for animals that are sung or told when one is out hunting or trapping so as to bring luck (Sullivan 1942:â•›80).

4.

Education and technology

Modern transportation and fixed settlements have changed the relationship between the people and the land. It used to take two days by dog team in the winter to travel between camps. Now that distance can be traversed in a few hours by snowmachine or boat. The change in lifestyle from nomadic to the more sedentary year-round village life has reduced the amount of time people spend out on the land. During the subsistence seasons, some families still go to fish camps along the Yukon, most go to fall moose camps in the Kkaayeh, and some people go out trapping. Even though some people still have intimate knowledge of the land, there is a marked language shift to English place names or to corrupted



Revitalizing place names through stories and songs 271

Denaakk’e names. Documenting and teaching the language, stories and songs is crucially important for the Koyukon as well as other Alaskan languages. Learning any language is a serious undertaking that takes motivation, commitment, and patience. Language revitalization methods, such as the MentorApprentice model (Hinton 1997), Total Physical Response (TPR) (Asher 1982), Total Proficiency through Reading and Storytelling (TPRS) (Ray & Seely 1998) and the Natural Approach (Krashen & Terell 1983), in an immersion setting in which the target language is the sole language of instruction are very effective. Lessons are designed so that students are provided comprehensible input through the use of physical movement, visual aides, repetition and engaging the five senses. Directly experiencing events that generate a strong emotion is an effective way of creating long-term memory. Learning the place names and ethnographical information by physically visiting the sites will help learners recall the lay of the land as they traversed to the site, the smells, sounds, and feelings it evokes. One curriculum that utilizes students’ personal knowledge in learning mapping skills is MapTEACH (Mapping Technology Experiences with Alaska’s Cultural Heritage), an educational curriculum that fosters community mapping, GIS and GPS (Stevens et al. 2008:â•›1). In the first lesson students draw a map of their favorite place. Using the map as a guide, each person shares his or her own oral history, in effect telling a story of his or her life. This lesson expands to include local elders who share their own mental maps of where they were raised. The addition of storytelling, following Charlie’s practices of stories alternating with song and advice, should help reinforce long-term memory. Oral storytelling is an art that takes practice. Eliza Jones recalls the advice given to her when she was young: “I’m going to tell a story, and I want you to listen real good, because tomorrow night you are going to tell it” (DeMartino 1986). If the learner made a mistake, the storyteller repeated that part of the story over again until the learner was able to tell the whole story. The kk’edonts’ednee ‘a time of long ago’ stories had to be retold verbatim with no alterations, each section told in full, and ended with the phrase “Et’egh ł huydo hutaaldlet yeenslenh de huyh naatlgus, ts’ednee” – ‘now I chewed off part of the winter,’ an adage intended to ward away a long winter. Another method to remember stories is to draw pictorial representations of events with a yaaruin ‘storyknife,’ a method used by female Yup’ik Eskimos in Southwest Alaska to tell stories and riddles or remember patterns (National Museum of Natural History, Smithsonian Institution 2002). In a modern classroom adaptation, both methods can be used to help students hear the story, draw pictures as mnemonic devices and retell it.

272 Susan Paskvan

Elders and community members strongly support language revitalization efforts, especially those that include the teaching of memorial songs. Interactive language lessons that engage participants are effective in teaching vocabulary and phrases. The use of pictures, props, and gestures can be used to provide students with comprehensible input, so translation into the student’s first language is not necessary. Once students learn the vocabulary and phrases for the lyrics, singers begin teaching the songs. When a singer is not available, students can use recordings of the songs. Replaying one lyric at a time is possible if the song is in a digital format. This method has been used in recent years, generating a strong cohort of young singers at potlatches. By tying the stories and songs to the traditional land use, trails and places can be taught by sharing maps of the area. Using current technology, such as Global Positioning Systems (GPS), Geographical Information Systems (GIS), Google Earth and a “sense of place” curriculum, such as that designed by MapTEACH, teachers in the Yukon-Koyukuk School District learn to create holistic maps. These skills are then taught to the students. In spring 2010, elders and students in five villages worked together on engaging map exercises that included Denaakk’e place names, traditional use areas, and map comparisons. As part of the process, sensitive information such as sacred sites or knowledge is restricted at the elder’s or community’s request. Documenting place names with the aid of video footage can be another stimulating tool to engage conversation amongst elders, especially if they are unable to travel to the land. In the spring of 2009 a group of elders were reviewing place name maps near Minto. When elders watched video footage of the Minto flats landscape it stimulated many more stories than were stimulated by looking at a flat paper map. They became very excited as they recalled historical stories, especially places where the whole community participated in events. Technology can help document and revitalize our indigenous language and knowledge, but there is no substitute for actually spending time on the land with elders.

5.

Conclusion

For the Koyukon people, learning the stories and songs of their ancestors through place names can be an enriching experience. Just understanding the verb’s structure, its morphemes, and the use of directional verbs is an educational journey in itself in Athabascan linguistics. Different semantics of nouns, specific or general, can teach one about the natural history and environment. The ethnographical information of a place name teaches one about the relationship of the land and



Revitalizing place names through stories and songs 273

people. Stories about place names can be of three types: origin, historical, and modern. In addition, the lyrics in memorial songs may indirectly refer to places that the deceased used for subsistence purposes such as trapping, fishing, or camping. Passing on this knowledge to the younger generation is a challenge, but using the indigenous teaching methods and technology, this can lead a learner one step closer towards the goal of language revitalization.

References Asher, James. 1982. Learning Another Language through Actions: The Complete Teachers’ Guidebook. Los Gatos CA: Sky Oaks. Attla, Catherine. 1996. Bekk’aatugh Ts’uhuney: Stories we Live By, Traditional Koyukon Athabascan Stories. Fairbanks AK: Yukon Koyukuk School District and Alaska Native Language Center. Basso, Keith H. 1990. Stalking with stories: names, places and moral narratives among the Western Apache. In Western Apache Language and Culture, Keith H. Basso (ed.), 99–137. Tucson AZ: The University of Arizona Press. Basso, Keith H. 1996. Wisdom Sits in Places: Landscape and Language Among the Western Apache. Albuquerque NM: University of New Mexico Press. Brush, Charlie. 1984. Interview by Eliza Jones, June 19, 1984. Unpublished ms. Fairbanks AK: Alaska Native Language Center. DeMartino, Marjorie. 1986. Storytelling: The preservation of an art. Alaska Today 13: 50–52. Degnan, Frances Ann. 1999. Under the Arctic Sun: The Life and Times of Frank and Ada Degnan. Unalakleet AK: Cottonwood Bark. Hinton, Leanne. 1997. Survival of endangered languages: the California Master-Apprentice Program. International Journal of the Sociology of Language 123: 177–191. Jetté, Jules. 1910. On the Geographical Names of the Ten’a. Microfilm AM 34:688-701. Spokane WA: Jesuit Oregon Province Archives, Foley Library, Gonzaga University. Jetté, Jules & Jones, Eliza. 2000. Koyukon Athabaskan Dictionary. Fairbanks AK: Alaska Native Language Center. Jones, Eliza. 1985. Native place names: our heritage. Paper presented at the Symposium on Indigenous Names in the North, March 22, 1985, at McGill University, Montreal. Kari, James & Fall, James A. 2003. Shem Pete’s Alaska: The Territory of the Upper Cook Inlet Dena’ina. Fairbanks AK: University of Alaska Press. Krashen, Stephen D. & Terrell, Tracy. 1983. The Natural Approach: Language Acquisition in the Classroom. Oxford: Pergamon. Loyens, William J. 1966. The Changing Culture of the Nulato Koyukon Indians. PhD dissertation, University of Wisconsin, Madison. Michael, Henry (ed.). 1967. Lieutenant Zagoskin’s Travels in Russian America 1842–1844. Toronto: University of Toronto Press. National Museum of Natural History, Smithsonian Institution. 2002. Alaska Native Collections: Sharing Knowledge. Arctic Studies, Smithsonian Institution, (28 January 2010). Nelson, Richard K. 1983. Make Prayers to the Raven. Chicago IL: University of Chicago Press.

274 Susan Paskvan

Osgood, Cornelius. 1936. The distribution of Northern Athapaskan Indians. Yale University Publications in Anthropology VII: 1–23. Ray, Blaine & Seely, Contee. 1998. Fluency through TPR Storytelling: Achieving Real Language Acquisition in School, 2nd edn. Berkeley CA: Command Performance Language Institute. Smetzer, Mary Beth. 2000. A meeting of the minds: Work of elder, priest culminates in Koyukon dictionary. Fairbanks Daily News-Miner, December 31: A-1, A-7. Stevens, De Anne, Stephens, Sidney, Burns, Patty, Batzli, Sam & Olsen, Timothy. 2008. MapTEACH: Place-based Geospatial Learning and Applications in Alaska. Fairbanks AK: University of Alaska Fairbanks. Sullivan, Robert J. 1942. The Ten’a Food Quest. Washington DC: The Catholic University of America Press. Thompson, Chad. 1977. Koyukon Verb Prefixes. MA thesis, University of Alaska Fairbanks. Thornton, Thomas F. 1997. Anthropological studies of Native American place naming. American Indian Quarterly 21: 1–23. Wahrhaftig, Clyde. 1965. Physiographic Divisions of Alaska [Geological Survey Professional Paper 482]. Washington DC: U.S. Government Printing Office.

chapter 13

Language and landscape among the Tlingit Thomas F. Thornton

Processes of perception, accommodation, and cultivation of places are critical to understanding the nature of language and landscape among coastal and river peoples. I present a processual model of landscape conceptualization in order to analyze how specific landscape classification schemes operate among the Tlingit of Southeast Alaska. Inhabiting some of the most dynamic and productive coastal and riparian ecosystems anywhere in the world, the Tlingit provide an especially rich case for examining these connections between land, river, and sea. Although the language is considered endangered, Tlingit toponyms and geographical nomenclature are well documented and many groups continue to occupy and use their traditional territory in ways that support traditional conceptualizations of the lands and waters of their living space.

1.

Introduction

The Tlingit of Southeast Alaska orient themselves fundamentally to the coastline that divides land from sea. In between lie the dynamic intertidal zones, or marinescapes, that are their predominant dwelling spaces within the broader cultural landscape that encompasses the uplands and sea. Arrayed along this coastal ribbon are settlements, subsistence sites, canoe trails, refuges and other landmarks of significance marked by thousands of richly descriptive place names revealing their perceptual and cultural salience (cf. de Laguna 1960, 1972; Nyman & Leer 1993; Thornton 1997a, 1997b, 1999, 2008, 2011). Here dwell the salmon, halibut, herring, shellfish, marine and terrestrial mammals and plants upon which these richly-endowed hunting-gathering people have depended for their living during various seasons of the year. Even today, with modernization (e.g., wage jobs and power boats replacing dugout canoes), language loss (fewer than 400 fluent speakers remaining in a population of over 20,000, rendering Tlingit endangered), and increased sedentism (in what used to be only their winter settlements), Southeast Alaskan Tlingits remain predominately oriented to

276 Thomas F. Thornton

the maritime coast. And despite participation in the modern cash economy, men and women of all ages still travel ancestral pathways for many miles to hallowed sites to fish, hunt, gather and trade subsistence resources. Tlingit is an isolate with the Na-Dene language family and consists of four mutually intelligible dialects: Gulf Coast, Inland, Northern, and Southern (de€Laguna 1972:â•›15–16). It is characterized by its agglutinative nature and grammatical emphasis on the verb complex. Typically the verb stem sits toward the end of a word and may be augmented by up to twelve prefixes and three suffixes, including relational nouns or theme prefixes, directionals, classifiers, and possessives, all of which figure in ethnogeographical reckoning. That the land and sea divide is foundational is evidenced by two of the most common Tlingit directionals: daak, translated as “inland from shore”, and daak, or “seaward from shore”, which permeate Tlingit landscape terminology and discourse. This distinction is also reflected in the culture’s metaphysics and cosmology, in which Raven gave the world light by stealing the sun, causing the animals to divide into creatures of the land and of the sea (Emmons 1991). Even the spirit world abides this division, distinguishing those that dwell inland (dáagi yéigi) from those that dwell out to sea (deikée yéigi). The horizontal distinction is complemented by vertical divide between those spirits above the earth ([di]kée yéigi) and those below water (heiya kwaaní) (see Veniaminov 1984:â•›397–398; Emmons 1991:â•›289, 291). These fundamental orientations remain salient for many Natives despite shifts from Tlingit to English and from animism to Christianity. Further, they are not unique to the Tlingit and may be grounded in a broader conceptual model that emphasizes four critical processes of landscape cognition among humans (Figure 1): perception, affordance, practice, and controlling force. Through an analysis of these four processes – how a place is perceived, what it affords, how it is accommodated in practice, and what constitutes its controlling forces (physiographic and biospiritual) – cultural landscapes may come to be understood not simply as named sites but as a composition of meaningful places within a socioecological system. Further, by understanding such places as ecotopes – generally “the smallest ecologically-distinct landscape features” (Hunn & Meilleur 2010:â•›15ff.) – within a landscape classification system, we may move toward building a theory of landscape ethnoecological classification, of kinds of places, across cultures. In doing so it should become clear what, if any, unity underlies the seemingly stunning level of diversity in biological, cultural, and landscape forms among the world’s cultures.



Language and landscape among the Tlingit 277

Perceptual Salience

Affordance (prospect, possibility, etc.)

Cultural Interest

Landscape

Perspective (vantage, orientation, etc.)

Ecotope (smallest ecologicallydistinct landscape features)

Practice (techniques, behavior, etc.)

Place

Physiography

Biospiritual Controlling processes (physical, metaphysical)

Figure 1.╇ Landscape and cognition conceptual model

2.

Perception

The first process in defining ecotopes, illustrated on the left side of Figure 1, is perception. Perception of any landscape or landscape feature is, of course, based on a type of interaction. Therefore, as Jackendoff (1987:â•›83) emphasizes, “One cannot take for granted the ‘real world’ as the domain of entities to which language refers. Rather … one’s construal [of the world] is the result of an interaction between external input and the means available to internally represent it.” Foundational to this is the reciprocity between our sensing bodies and particular landscape features. As scholars from Merleau-Ponty (1962) to Ingold (2000, 2007) have argued, there is a phenomenological aspect to perception that turns upon the human body’s own being and participation in perception. Summarizing Merleau-Ponty’s major findings, Abram (1996:â•›90) notes: (1) The event of perception, experientially considered, is an inherently interactive, participatory event, a reciprocal interplay between the perceiver and the perceived. (2) Perceived things are encountered by the perceiving body as animate, living powers that actively draw us into relations … (3) The perceptual reciprocity between our sensing bodies and the animate expressive landscape both engenders and supports our more conscious linguistic reciprocity with others… (4) Human languages, then, are informed not only by these structures of the human body and the human community, but by the evocative shapes and patterns of the more-than-human terrain.

278 Thomas F. Thornton

In other words the mind itself is embodied in a particular “technology” and “landscape” (see Thornton 2008:â•›143–147) and thus cognition itself is product of having “a body with various sensorimotor capacities which are embedded in a more encompassing biological, psychological, and cultural context” (Varela et al. 1991:â•›173; see also Lakoff & Johnson 1999). These capacities are ultimately constrained by the particular physiographic context that frames the perceptual field, though there is ample space beyond this field in what might be termed geographies of the imagination or “geographic imaginations” (Gregory 1994). Given this logic and the primacy of the body as instrument of landscape perception, we would expect to find widespread metaphorization and anatomical referencing of the body in landscape terms. Indeed this is the case in Tlingit (see Thornton 2008:â•›24, 85) and many other languages, including English where we speak of the “head” and “mouth” of a river, or the “arm” of a bay, for example. Yet although common, the patterns for bodily referencing in geographic terms are by no means universal. Tlingit, perhaps again influenced by the dynamic flows and exchanges that occur along its dramatic coastline, is partial to orificial references, to mouths (x’é used for mouths of hot springs, rivers, bays, etc., as in T’aay X’é, “Hot springs Mouth”), noses and nostrils (lu, lutu, used often for points and features inside of points, as in Ltu.áa, “Lake Inside the Nostril”), anuses (tukx’e, for outlets of lakes or lagoons, as in Tuktayee, “Below the Anus” [outlet of lake]), and so on. As an agglutinative, verb-centered language, Tlingit can incorporate anatomical references not only as nouns, but also as verbs. An example of this is a place near Sitka, Yoo Luklihashgi X’aa, often translated as “Floating Point”. When unpacked more literally, this means “Point that Moved Up and Down in a Perching/Squatting Posture”, suggestive of an animate (human-like) body. Another point is known as Yoo Lititgi X’aa (“Undulating Point” [as if rocked by waves]). From a canoe voyager’s or beach dweller’s perspective such bodily references neatly capture the varying qualities of this physiographic feature in different atmospheric conditions. As Tlingit Elder, Herman Kitka, Sr. explains Yoo Luklihashgi X’aa: That’s where they saw that lost chief that went out to sea, Kaax’achgóok, coming back. They called it “The Floating Point” because they said it lifts if it’s going to change weather. From the camp, it looks like it’s floating in the air (you see things that way sometimes) and that’s where the name came from. In real good weather you see that point and it looks like it’s floating. So they had different things that they kept track of, and that’s one of the things. That’s why they named that point – they know when it’s going to rain or be good weather. (in Thornton 1995a:â•›240)

Mr. Kitka’s point about geographical names “keeping track of things” is an important one, as perception is not only a process of classifying landscape features



Language and landscape among the Tlingit 279

but distinguishing their varieties of appearance and influence in space, time, and particular conditions. This underscores another point about ethnophysiography, namely that geographic features, be they points or mountains or rivers, “are not bona fide objects in the usual sense, but are more properly considered to be parts of the Earth’s surface” (Mark et al. 2010:â•›30). Changes in the earth’s surface or atmosphere or in the orientation of the observer may, in turn, change a feature’s qualities and thus how it is perceived, just as a historical event may mark a place for remembrance. Tlingit is precise in relating these changes, in part because it is essential to navigation and reckoning. Thus, Mount Saint Elias, the region’s tallest mountain at more than 5489 meters (but named for its relation to an important bay, Waas’eit’ashaa, “Mountain inside Icy Bay”), might look like a “seagull” or a “rabbit” from a distance, according to visibility and conditions, and gradually change its perceptual form as the traveler moves closer. Distances are routinely related as part of object classification through the following affixes: hé (marking an object near and present); yaat (indicating something more distant but still in one’s scope); yut (denoting an object remote and impersonal); wé (so distant as to be hardly perceptible); and, finally, wuyík (beyond recognition). Beyond this, the dynamism and flows of the earth itself, from its daily cycles (e.g., tides), seasonal changes (e.g., spring growth), long-term climatic or geomorphological processes (e.g., glacial retreat and uplift), and momentous change events (earthquakes and tsunamis) were reckoned in the geographic lexicon. As an example, Glacier Bay, now a National Park and World Heritage site, became famously known in Tlingit as Sít’ Eeti Geeyí, or “Bay Taking the Place of the Glacier,” a precise description of what has happened geomorphologically in the wake of the rapid retreat of the Bay’s major tidewater glaciers (Thornton 1995b).

3.

Affordance

Beyond perception there is affordance, or the “action possibilities” presented by an object or landscape feature in relation to the actor perceiving it. Affordance theory was developed in ecological psychology by Gibson (1979) as a means of understanding how particular objects or environments “invite” certain action possibilities and constrain others. In ecological terms this means accepting that the landscape is not wholly a human construction or one born strictly of physiographic processes, but rather a dialogue between perception, embodied engagement, and environmental possibility. Gibson emphasized visual perception in developing his theory of affordances, but not in isolation. Other senses and the phenomenon of synaesthesia, or multisensory engagement, come into play as an

280 Thomas F. Thornton

agent pursues a course of activity in a particular environment, assessing affordances along the way. As Ingold (2000:â•›168) points out, [F]rom a Gibsonian perspective, it is apparent that the world becomes a meaningful place for people through being lived in, rather than through having been constructed along the lines of some formal design. Meanings are not attached by the mind to objects in the world, rather these objects take on their significance€– or in Gibson’s terms, they afford what they do – by virtue of their characteristic pattern of day-to-day activities. In short, far from being inscribed upon the bedrock of physical reality, meaning is immanent in the relational contexts of people’s practical engagement with their lived-in environment.

I would argue that affordance is relative both to peoples’ “practical engagement” and cultural interests in their environment and the technologies of engagement€– including language and knowledge systems – that they possess. It may be, as Abram (1996:â•›130) suggests, that “the animistic discourse of indigenous, oral peoples is an inevitable counterpart to their immediate, synaesthetic engagement with the land that they inhabit … Direct prereflective perception is inherently synaesthetic, participatory, and animistic, disclosing the things and elements that surround us not as inert objects but as expressive subjects, entities, powers, potencies”. But these entities need not be conceptualized or classified the same way across cultures. As Julie Cruikshank (2005) has shown in detail, Tlingit and Athabaskan participatory relations with glaciers, which they classify as possessing animate spirits that listen and respond to human actions from a moral-ecological perspective, are distinctly different from those of the early explorers and modern scientists who have encountered these spectacular landforms. This is true despite the fact that glaciers may present similar affordances across cultures. Raven, the prototypical namer and shaper of landscapes in Tlingit cosmology, has an encounter with place called Skanáx (unanalyzable; Saginaw Bay in English, a major historical site in Southeast Alaska) that is particularly revealing about the nature of synaesthetic engagement, affordances, and what might be termed “action impossibilities”. As recounted in Being and Place among the Tlingit (Thornton 2008:â•›111), Raven talked to it in order to make it into Nass [River] … but, when the tide was out great numbers of clams on the flats made so much noise shooting up at him that his voice was drowned, and he could not succeed. He tried to put all kinds of berries there but in vain. After many attempts, he gave it up and went away saying, “I tried to make you into Nass, but you would not let me. So you can be called [Skanáx].”  (in Swanton 1909:â•›15)



Language and landscape among the Tlingit 281

Though it resembles his beloved and prototypical Nass River, the affordances of Skanáx are not the same as those at Nass. Berries are perhaps not as abundant as at Nass, whereas clams are more bountiful, indeed so plentiful that their cacophonous “shooting up at him” overwhelms his senses, literally preventing him from redundantly creating another Nass purely out of want, longing, or nostalgia. In this episode Raven shows that place naming is an organic process involving a dialogue between the human senses, cultural models of experience, and key characteristics and affordances of the environment. Raven cannot simply impose a prototypical cultural model onto Skanáx. At a fundamental level the place must speak for itself, and the namer should listen. Raven’s adventures on the land also point to the importance of practical interests in shaping ecotopes in language and thought.

4.

Practice

While affordances imply or invite certain possibilities, practice is the process of enacting these possibilities. An ecotope’s utility is a reflection of its practical significance to those who make a living from it. As Hunn and Meilleur (2010:â•›3) stress, “Folk ecotopes highlight features of the landscape useful for people making a living off the land”. [emphasis in the original article]. In Tlingit, practice is reflected both in basic landform terms and descriptive activity names that dot the landscape. Significantly, however, the ecotopes do not only highlight features of the land useful for people making a living, but also for other communities of beings that dwell on the land. Thus, Tlingit ecotopes sometimes embrace a perspectivism extending beyond human orientation and affordance to those of non-human persons, including animals and powerful spiritual beings that also constitute and share the landscape. Examples of generic landscape terms indicative of practice are noow (fort or refuge) and náx (harbor or protected bay). In theory, nearly every defensible promontory or refuge site might be considered a potential noow, and every protected bay a harbor, or náx. However, in practice, a noow is defined by a number of practical considerations beyond defensibility. It may be a stronghold of a certain community – human or animal – as in Xutsnoowú (“Brown Bear Fort”), the Tlingit name for Admiralty Island, or Jánwu Noowú (“Mountain Goat Fort”) for the (otherwise rather perilous) cliffs above Yakutat Bay where mountain goats dwell. Alternatively, a noow might be a refuge only under certain conditions, be they seasonal (Táak’w Noow, “Winter Fort”, at Chaik Bay), prevailing (Kanasnoow, “Windbreak Fort”, at Killisnoo Island), physiological (Wéitadi Noow, “Fort of the Young Woman in Seclusion” [in menarche], in Excursion Inlet), or sociopolitical

282 Thomas F. Thornton

(Xunaa Káawu Noowú, “Hoonah People’s Fort”, in Glacier Bay National Park). Whether a refuge or stronghold becomes known as a noow, as opposed to a mere settlement (aan) or area of possessed land (aaní), ultimately depends on how a range of physiographic, ecological, and cultural-historic factors and contingencies are negotiated and realized in practice. Similarly, whether a bay (geeyí) or cove becomes designated as a harbor is a matter of cultural interest and practice and not merely affordance or physiography. In fact, some landscapes may be avoided despite obvious advantages of affordance because they are dominated by biospiritual forces that are potentially dangerous. Shamanic landscapes and the dwelling places of Land Otter Men are examples of ecotopes controlled by strong spiritual forces which may trump mundane cultural interests in the land, such as the pursuit of food and shelter. Generally, such landscapes are to be avoided, save for exceptional individuals like shamans. Of land otters, Swanton (1908:â•›456) remarks, Although apparently so harmless, the land otter was dreaded more than any other creature. This was on account of his supposed supernatural powers, fondness for stealing people away, depriving them of their senses, and turning them into land-otter men (ku’cta-qa). [Kooshdakaa]. As they lived at various points along shore, these land-otter men were called q!a’tu-qa [X’aatu Káa] (“men-inside-ofpoints”). Naturally enough the land otters were closely associated with shamanism, which, in fact, is said to have come from them. Years ago the Tlingit would not use their fur.

Place names often mark potent spiritual sites by referencing their constituent agency in conjunction with a generic landform term, as in íxt’ X’aayí (“Shaman Point”), or Kooshdaa Xágu (“Land Otter Sandbar”). In other cases, it was important to negotiate with them in order to secure one’s interest, be it safe passage or success in the hunt. As Swanton (1908:â•›453) relates, “The first big mountain at Cape Edwards is called Xās [Xaas], and when a person went by it he always said, ‘Spread out your legs (so that it will not be rough). Do not harm me’. There was another high rock to which people always talked as they passed. From Sitka around to Huna [Hoonah] there were various places to be spoken to”. An example is the “Volcano Woman” called Shee, for whom Baranof Island is named, who is said to have expected gifts from those who hunted or gathered from her island (Kruzof Island), lest she fume (literally) or cause them bad luck. Some Tlingit hunters reportedly still oblige her expectations by leaving a shiny coin or other offering on the island. Another critical dimension of ecotope conceptualization was cultural keystone species (Garibaldi & Turner 2004). Important keystone species include salmon, yellow cedar, certain berries, harbor seals, and a variety of medicines



Language and landscape among the Tlingit 283

and specialty foods. I have commented elsewhere on the high cultural value of red (sockeye) salmon (Oncorhynchus nerka) streams, which tend to be named for this species even when others are present (Thornton 2008:â•›94). Beyond their cultural importance, however, a red salmon stream represents a very different kind of ecotope, for it is nearly always a stream with a freshwater lake system, which other salmon do not require. Moreover, red salmon streams typically support 2–4 other complementary species of Pacific salmon, giving them especially high value as harvest sites. Similarly, high diversity value yellow cedar groves, strawberry patches, halibut banks, bird egging grounds, and other keystone cultural species harvest sites tended to be marked as ecotopes.

5.

Biospiritual forces

I have already alluded to the importance of spiritual forces in the Tlingit biosphere and cosmography. Particularly in animistic societies, powerful spirits and non-human persons often act as controlling forces in the ethnophysiography. So it is in Tlingit country, where the ocean, rivers, mountains, glaciers, and a variety of other landforms were considered alive or animated by spirits to whom one could appeal (Swanton 1908; de Laguna 1972). In some cases, potent biospiritual sites were avoided, as with shamanic landscapes, but in other cases gaining influence over their controlling force was possible, typically by means of cultivation in a reciprocal or moral-ecological context. Spiritual forces associated with animated landscapes might be appealed to for wealth, such as Tlanaxeedakw, the “Wealthbringing Woman” associated with Áak’w (“Little Lake,” Auke Lake near Juneau). As wellsprings of critical resources (e.g., salmon) and dynamic exchanges (e.g., between saltwater and freshwater ecosystems), coastal lakes, rivers, and lagoons are especially potent landscapes, and are often associated with particular controlling forces. Swanton relates, “At the head of every creek was an old woman whom the salmon tried to reach. The small sized ones wore out their noses trying to get up to her. In the Auk country is a creek to which one had to use good words. Otherwise, if he fell down near the stream, it would cut his hands to pieces, although the rocks are not sharp ordinarily” (1908:â•›453). Significantly, in Tlingit there is no clear distinction between a freshwater lake and a saltwater lagoon or enclosed cove (all may be termed áa, or áak’w if small), perhaps in part due to the dynamic shifts in coastline that have transformed lagoons into lakes (e.g., through isostatic rebound and uplift) and lakes into lagoons (e.g., through erosion, sea level rise, etc.). The dynamism of particular sea currents and tidal rips was also associated with spiritual forces, particularly “The Old Woman Underneath” (Haiyé Sháanak’w) who controlled the tides and “had charge of a post made from

284 Thomas F. Thornton

a beaver’s foreleg, on which the world rested. When Raven tried to drive her away from this post the earth quaked” (Swanton 1908:â•›452). Biospiritual aspects of landscapes often derive from mythic time, when the world was formed or shaped by Raven and other metaphysical agents. These aspects remain strong by virtue of the fact that ancestral spirits continue to dwell on the land. Their traces are etched and inscribed in places (Yéil Áa Ludaawdligoowu Yé, “Place Where Raven Wiped his Beak,” or Tsaa Takdi X’áak “Between which Seals Were Harpooned”) and their presences may still be palpable to those who frequent the same haunts. Sensing them engenders powerful sentiments, which can be strongly intuitional and emotional, creating powerful feelings of attachment, belonging, awe, and, in some cases, fear.

6.

Islands and refuges in Tlingit landscape and cognition

Having characterized the key processes involved in conceptualizing and classifying landscape, we can apply the model to two conventional features in the coastal Tlingit landscape: islands and canoe roads. Islands present a good case for study because they are widely recognized across cultures (save perhaps for desert societies), often with generic terms. But, as in many coastal populations, some groups may be more island-oriented than others depending on the physiography where they dwell, the islands’ perceptual salience, affordances, cultural values, and controlling forces. In Tlingit country, the distinction between island and mainland Tlingits is significant in terms of their landscape perception. De€Laguna (1972:â•›797) found, [Yakutat Tlingits] differ from the Tlingit of southeastern Alaska, at least from those of Angoon (de Laguna 1960:â•›19–30), in the extent to which the world in which they actually live includes more than the narrow ribbon along the shore, or the canoe routes across bays and inlets between landfalls … Thus, the country behind the ocean beach or the shores of the bay are visited and traversed much more freely and confidently by hunting parties than was the interior of Admiralty Island by those who lived at Angoon … Formerly they even ventured on arduous and dangerous journeys up the swift rivers and across the great plains of ice into the homelands of the Gunana [Interior Athabaskans] behind the mountains. The Yakutat people orient themselves [in addition to the seaward-inland distinction] primarily with reference to known, named, landmarks. Of these the most important are the two great peaks, Mounts Fairweather and Saint Elias [conceptualized as husband and wife], although every striking rock formation, reef, island, point of land, bay, stream, or lake serves as a guide to location.



Language and landscape among the Tlingit 285

Because orientation was reckoned according to seaward versus inland features and flows (especially those of rivers and winds), there were no cardinal points, but rather relational directionals based on the contours of the coastal physiography. Islands dominate the geography of the Alexander Archipelago, which defines Southeast Alaska below Yakutat country. There are thousands of islands/islets of various shapes and sizes. Not all constitute ecotopes of significance and many are not named. Large islands typically lacked generic references and could be conceptualized as consisting of multiple ecotopes, depending on their physiography and ecological diversity, as is true in many maritime cultures on the Northwest Coast and beyond (Thornton 2008; Waterman 1922). In Tlingit large islands were often named metonymically according to their contents or distinguishing features (e.g., Xutsnoowú, “Brown Bear Fort”, for Admiralty Island or Taan, “Sea Lion”, for Prince of Wales Island, or portions of these landforms). For other islands named in Tlingit, as Figure 2 illustrates, there are two common generic terms deployed, X’aat’ and Daa, and separate terms for reefs (eejí) and a variety of rocks and boulders which might constitute small islets conceptually. X’aat’ is by far the most common generic and may be rendered diminutive for an islet (x’aak’w). The term may be linguistically related to the word for point (x’aa), although the precise connection is not clear. As noted above, a point may also be referenced anatomically, as a nose (lú or luká), whether on a point of the mainland or an island.

Perceptual Salience

Affordance: food, navigation, shelter, etc.

Perspective

X’áat’ (Island)

– Large islands often lack generic term – Small islands – diminutive (X’áat’ak’u) – Daa islands used mainly by Island Tlingits; applies to smaller islands defined by what surrounds them

Cultural Interest Practice: – Refuge and subsistence areas – Fort sites, lookouts, ancestral domains

X’aa (Point) Yakwdeiyí (Canoe road) Eejí (Reef )

Daa (Surrounded)

Physiography

– Terrestrial and marine resources

Biospiritual Processes: tides and currents, microclimates, glaciers, etc.

Figure 2.╇ Landscape and cognition model for island marinescapes in Tlingit

286 Thomas F. Thornton

Significantly, x’áat’ is a basic landform term, while daa is not. Daa is a relational term meaning “around”, or more precisely, “surrounding”. The distinction can be understood in terms of the processes outlined above. Perceptually, x’áat’ is salient because of the terrestrial aspects of the island, while daa is distinguished by what surrounds the island, or its marine qualities. Thus Guwakaan X’áat’i (“Deer Island”) is a metonymic name that identifies an island as possessing deer, an affordance of its terrestrial environment of cultural interest to the Tlingit. (In addition to being a food source, deer are the symbol for peace). In contrast Tandaa, “[Fish] Jumping Around (Island)”, invokes the salient marine environmental aspects of an island that is most noteworthy for the boiling masses of salmon and herring that jump around it at certain times of the year. Another island sitting in the middle of the perilous Icy Straits-Cross Sound passage, which connects the northern inside passage of Southeast Alaska with the outer Pacific coast, is called Taas’ Daa, or “Double-Headed Tide Around It”, an apt named for a feature that affords no easy passage during heavy stages of the tide. The island (Lemesurier Island) also contains terrestrial features marked by names, including various points and a fort site, but these are clearly subordinate to the controlling forces that surround it. Yakwdeiyí, or canoe roads, are classic marine affordances, offering protected shoreline passage or ocean-to-shore routes to refuges. In contrast to islands, they are defined not by what is on and around them but where they lead. Yakwdeiyí are marine analogs of terrestrial trails (deiyí) but the term is also synonymous with refuge. As Figure 2 suggests, they occur naturally between islands, but may also be cultivated by supernatural beings or even humans, for example by removing boulders from rocky shorelines to create safe landings. Such refuges come to gain status as ecotopes. The ribbon of sandy shoreline that separated the river deltas of the lower Yakutat forelands from the open Pacific ocean formed one such canoe trail, affording a protected avenue of travel from Yakutat Bay to Dry Bay, some sixty miles to the south, which otherwise would not have been navigable during rough weather. Dry Bay itself was also yakwdeiyí, refuge, because the ocean is calm there and canoes would enter its mouth to come safely ashore after crossing the treacherous Gulf of Alaska (known as Yéil T’ooch’, “Raven Black”). In fact the canoe road there was made by Raven himself, who pulled in his famous “food canoe” at a site called Yéil Áx Daak Akaawajiyi Yé (“Place Where Raven Dragged in the Salmon Canoe Moving his Legs Funny” [i.e., taking a long step and then scooting and pulling]). This movement, in turn, created Yéil Áx Daak Uwanugu Yé (“Place Where Raven Scooted Back”), a set of sand dunes by Akwe River formed by the displacement of sand from Raven’s scooting tracks. People who land here are literally following Raven’s tracks in the country, the tracks that made the country. It is not only a site of refuge but of prospect



Language and landscape among the Tlingit 287

(Appleton 1975), as this is where goods were traded between coast and interior peoples via yet another yakwdeiyí, the canoe road up the Alsek River (Aalseix’) to the interior Athabaskan peoples. These interior groups were known as Gunanaa in Tlingit and thus Dry Bay became known as Gunaaxoo, “Among the Interior Peoples”. The trail of Raven continues up the river where his misadventures are inscribed in the physiographic landscape as the natural laws of the country, which are held to this day as shagóon, or “heritage and destiny”. Thus, yakwdeiyí have perceptual, affordance, cultural interest, and biospiritual salience. Yakwdeiyí may be spiritual as well as practical refuges. Culturally, if not linguistically, they are basic features, which connect land and sea and constitute a critical pathway of marinescape apprehension and engagement.

7.

Conclusion

Landscape classification is motivated by environmental perception and practice in relation to salient physiographical, metaphysical, and cultural processes. As recent work in the fields of ecological psychology, cognitive linguistics, phenomenology, and anthropology suggests, perception and practice are not so easy to separate (Levinson 2008). Practice becomes a way of perceiving, acting a way of knowing, and classifications emerge within the exigencies and affordances of everyday life on the land. Cognitive partitioning of the landscape domain, then, involves reckoning not just perceptual salience but cultural interests and dynamic processes of landscape change and interactions. These extend across spatiotemporal scales and generations of experience. Thus, just as critical to Tlingit environmental perception as acting on the world are ancestral lines, or paths (see Ingold 2007, 2000:â•›146; Thornton 2008:â•›118ff.) that link one to particular networks of places, people, spirits, and past events. As coastal dwellers in a maritime belt rich in islands, bays, rivers, and coastal mountains, Tlingits oriented themselves fundamentally between the seaward Pacific ocean and the inland watersheds and mountains that defined the boundaries of their living space. Geographical names cluster on the land-sea interface, drawing attention to critical affordances and cultural interests in these places and their controlling forces. The relative “thickness” and contours of Local and Traditional Knowledge (LTK) are indexed by the density of named sites and graded landscape classifications. Biodiversity and cultural diversity may be similarly indexed by differences in naming patterns and classification of landforms between island and mainland Tlingits and the canoe roads that link them.

288 Thomas F. Thornton

References Abram, David. 1996. The Spell of the Sensuous. New York NY: Vintage Books. Appleton, Jay. 1975. The Experience of Landscape. London: John Wiley and Sons. Cruikshank, Julie. 2005. Do Glaciers Listen? Local Knowledge, Colonial Encounters, and Social Imagination. Vancouver BC: University of British Columbia Press. De Laguna, Frederica. 1960. The Story of a Tlingit Community: A Problem in the Relationship between Archeological, Ethnological, and Historical Methods [Bureau of American Ethnology, Bulletin 172]. Washington DC: Government Printing Office. De Laguna, Frederica. 1972. Under Mount Saint Elias: The History and Culture of the Yakutat Tlingit, 3 Vols. Washington DC: Smithsonian Institution Press. Emmons, George T. 1991. The Tlingit Indians. Edited with additions by Frederica de Laguna [American Museum of Natural History Anthropological Papers 70]. Seattle WA: University of Washington Press and the American Museum of Natural History. Garibaldi, Ann & Turner, Nancy. 2004. Cultural keystone species: implications for ecological conservation and restoration. Ecology and Society 9(3): 1–18. Gibson, James J. 1979. The Ecological Approach to Visual Perception. Boston MA: Houghton Mifflin. Gregory, Derek. 1994. Geographic Imaginations. Cambridge MA: Blackwell. Hunn, Eugene S. & Meilleur, Brian. 2010. Toward a theory of landscape ethnoecological classification. In Landscape Ethnoecology, Leslie M. Johnson & Eugene S. Hunn (eds), 15–26. New York NY: Berghahn. Ingold, Tim. 2000. The Perception of the Environment: Essays in Livelihood, Dwelling, and Skill. London: Routledge. Ingold, Tim. 2007. Lines. London: Routledge. Jackendoff, Ray. 1987. Consciousness and the Computational Mind. Cambridge, MA & London: MIT Press. Lakoff, George & Johnson, Mark. 1999. Philosophy in the Flesh: The Embodied Mind and its Challenge to Western Thought. New York NY: Basic Books. Levinson, Stephen. 2008. Landscape, seascape and the ontology of places on Rossel Island, Papua New Guinea. Language Sciences 30: 256–290. Mark, David M., Turk, Andrew G. & Stea, David. 2010. Ethnophysiography of arid lands: categories for landscape features. In Landscape Ethnoecology, Leslie M. Johnson & Eugene S. Hunn (eds), 27–45. New York NY: Berghahn. Merleau-Ponty, Maurice. 1962. The Phenomenology of Perception. London: Routledge and Kegan Paul. Nyman, Elizabeth & Leer, Jeff. 1993. Gágiwdul.àt: Brought Forth to Reconfirm; The Legacy of a Taku River Tlingit Clan. Whitehorse YT: Yukon Native Language Centre, Yukon College; Fairbanks AK: Alaska Native Language Center, University of Alaska. Swanton, John R. 1908. Social condition, beliefs and linguistic relationships of Tlingit Indians. In Twenty-sixth Annual Report, Bureau of American Ethnology, 391–485. Washington DC: Government Printing Office. Swanton, John R. 1909. Tlingit Myths and Texts [Bureau of American Ethnology Bulletin 39]. Washington DC: Government Printing Office. Thornton, Thomas F. 1995a. Place and Being among the Tlingit. PhD. dissertation, University of Washington, Seattle.



Language and landscape among the Tlingit 289

Thornton, Thomas F. 1995b. Tlingit and Euro-American toponymies in Glacier Bay. In Proceedings of the Third Glacier Bay Science Symposium, 1993, Daniel Engstrom (ed.), 294–301. Anchorage AK: National Park Service. Thornton, Thomas F. 1997a. Anthropological studies of North American Indian place naming. American Indian Quarterly 21(2): 209–228. Thornton, Thomas F. 1997b. Know your place: The organization of Tlingit geographic knowledge. Ethnology 36(4): 295–307. Thornton, Thomas F. 2008. Being and Place among the Tlingit. Seattle WA: University of Washington Press. Thornton, Thomas F. 2011 [Forthcoming]. Haa Leelk’w Has Aaní Saax’u: Our Grandparents’ Names on the Land. Juneau & Seattle: Sealaska Heritage Institute & University of Washington Press. Thornton, Thomas F. (with Martin, Harold P.). 1999. What’s in a name? Indigenous place names in Southeast Alaska. Arctic Research of the United States 13 (Spring/Summer): 40–48. Varela, Francisco J., Thompson, Evan & Rosch, Eleanor. 1991. The Embodied Mind: Cognitive Science and Human Experience. Cambridge MA: The MIT Press. Veniaminov, Ivan. 1984. Notes on the Islands of Unalaska District. Translated by Lydia T. Black and R. H. Geoghegan, edited by Richard A. Pierce. Kingston ON: Limestone Press. Waterman, Thomas T. 1922. The geographic names used by the Indians of the Pacific coast. Geographical Review 12(2): 175–194.

chapter 14

Language, landscape and ethnoecology, reflections from northwestern Canada Leslie Main Johnson

I draw on Witsuwit’en, Kaska and Gitksan landscape research to investigate similarities and differences in landscape terminologies, and ethnoecological implications of landscape kinds in northwestern British Columbia and the southern Yukon. Kaska and Witsuwit’en are Athapaskan languages while Gitksan is a Tsimshianic language. Gitksan and Witsuwit’en share similar landscapes and some aspects of social structure, though they are linguistically distinct, and Witsuwit’en and Kaska share some aspects of traditional economy and language, though there are differences in landscape and social structure. This three-way comparison allows exploration of the interaction of language, landscape and ecological perspectives.

1.

Introduction

The term ‘landscape’ has a number of meanings in different disciplines and in common speech. The humanist conception of landscape is a prospect, a viewscape, while the ecological conception of landscape comprises a range of physical, hydrographic and biotic features and fluxes, an extent of lands and waters which serve as habitat to a variety of organisms, and which constitute an ecological system. These concepts are discussed in Johnson and Hunn (2010) and Johnson (2010a) and by numerous authors cited in those two works. In an ethnoecological approach, landscape comprises an areal extent on the order of drainage basins and surrounding interfluves, and may be thought of as the area of land with which human groups normally interact, travel in or have knowledge of … for the groups I work with, largely equivalent to homelands. I conceive of landscape as comprising both large, more distant features and what Granö (cited in Mark et al. 2010) has called “the proximity” – that is both the mountain and the smaller, more specific features of the mountain such as a

292 Leslie Main Johnson

meadow, fell-field or tarn; or the river and the smaller features of the river such as eddies, boils, back waters, and side channels. Associated with ‘river’ would also be features such as sand bar, gravel bank or quicksand, and features like ‘its bank,’ or ‘deep water’. Snow and ice features are also significant in places with a strong winter season, or at high altitude. Along with physical characteristics, I consider vegetation a significant part of landscape, or “the Land”. Finally, human created features such as trail, village or camp are also appropriately conceived as features of landscape. While ecologists are often concerned with habitat, and this concept is much employed in ethnoecological literature, my concern is more with kinds of place recognized by local peoples, whether or not these are evidently conceived as habitat for specific plants or animals.

2.

Unpacking “ethnoecology”

I espouse an “ethnoecological” approach to studying cultural knowledge of landscape. The term “ethnoecology” has been used in many ways to describe local (cultural) understandings of, use of and relationship to environments (e.g. Toledo 1991, 2002; Nazarea 1999; Nabhan 2001; and others). Ethnoecology can be conceived as the broad domain of local understanding of the environment, of the land and the entities that dwell there, and of the relationships among them, including the relationships of people to other living beings and the land (Johnson 2000, 2008). My particular interest is in how people understand and distinguish significant features of their local environment, and the meanings or significance of these sites. I hypothesize that the suite of features distinguished will reflect characteristics of the landscape itself, the way of making a living of local people, and aspects of the language(s) they speak. There are two main aspects to an ethnoecological approach to landscape: one deals with structure – the range of locally recognized place kinds or ecotopes, how they are recognized and named, and their entailments or “affordances” (Gibson 1979, cited in Thornton 2008); and the other emphasizes relationships, movement, flow, and meaning. Both in their different ways examine the relationship of local cultural groups to their homelands or local regions, draw out ecological implications from local landscape knowledge, and use knowledge of the land to gain insight into local ecological concepts and relationships. In this paper I will explore landscape and language in a comparative examination of landscape ethnoecology of three indigenous groups in northwestern British Columbia and the adjacent Yukon Territory in western Canada, highlighting features of regional landscape and ways of life to provide background for this



Language, landscape and ethnoecology, reflections from northwestern Canada 293

Yukon

NWT

Pelly R.

Mackenzie R.

Teslin L.

Liard R.

Stikine R.

British Columbia Babine R.

Alberta

Skeena Bulkley R

Fraser R.

CA NA DA USA Figure 1.╇ Homelands of the groups discussed in this chapter. From south to north, Witsuwit’en territory is in the darkest gray, Gitksan territory, just north of the Witsuwit’en, is in a medium gray, and the Kaska Dena lands (pale gray) extend from northeastern British Columbia into the central Yukon. Locations are indicative only and are not definitive land claim boundaries.

examination (Figure 1). I use a comparative framework to tease out influences of language, culture, way of life, and the nature of the landscape itself on local understanding of landscape.

294 Leslie Main Johnson

3.

The people and the landscapes

3.1

Gitksan

The Gitksan, a nation of about 13,000 (Gitksan Chief ’s Office, http://www.gitxsan. com/html/who.htm) have their homeland in the Skeena River (Ksan) drainage in British Columbia, a large river draining into the north Pacific, which hosts substantial runs of five species of Pacific salmon and steelhead trout. The bulk of their traditional territories are in the densely forested Coast Mountains of British Columbia, glaciated ranges that reach higher than 1,800 m (6,000 feet) above sea level. The Gitksan are northwest coast peoples who speak a Tsimshianic language allied to Nisga’a, Coast Tsimshian (Sm’algyax), Southern Tsimshian, and more distantly to some languages in coastal Oregon. Their traditional economy was based in river salmon fishing, hunting large and small game, and harvest of key plant resources, especially berry species, tree cambium, and a couple of root foods (Johnson 1997; Daly 2005). Trade with both coastal and interior peoples has been a part of the local economy for millenia, with evidence for long-distance connections in the region extending back 4,000 to 5,000 years (see discussion in Johnson 1997). The Gitksan seasonal round featured dispersal to summer fishing sites and other resource areas in the summer season, and centralized residence in permanent winter villages, with a winter feast season; however, in the fur trade period, some families trapped in more distant territories for most of the winter, returning to central villages only for brief periods in the winter season (see Daly 2005 and Johnson 1997 for more background on Gitksan economy and culture). Gitksan (Sim Algyax or Gitxsanimx) is an ergative language (Rigsby 1986), with a rich system for indication of location. Gitksan culture is characterized by organization into exogamous Clans [Phratries] with constituent House groups (Wilp), headed by hereditary chiefs (Simgigyet) who own the territories and other resource sites and govern on behalf of their house members. Gitksan society is hierarchical, and inheritance is through the mother’s line. House territories (lax yip) are typically bounded tributary watersheds or mountains, with firm boundaries. A series of other types of smaller resource sites, such as specific fishing stations, smokehouse sites, and berry patches, also exists. These might be nested within larger hunting territories, as the numerous fishing stations along the Skeena River north of Kispiox Village, which belonged to a number of different House groups, though the surrounding hunting territory might belong to a single chief.



3.2

Language, landscape and ethnoecology, reflections from northwestern Canada 295

Witsuwit’en

The Athapaskan-speaking Witsuwit’en (Bulkley River people) live immediately to the south of the Gitksan. Like the Gitksan, with whom they feasted and intermarried through the generations, Witsuwit’en have Chiefs (Dinï ze, Ts’akë ze, Skiy ze), Clans, and Houses (Yikh), and inherited names pass through the mother’s line. The landscape they inhabit is in the valley and flanking mountain ranges of the Bulkley River (Widzin Kwikh), a major tributary of the Skeena (Ksan), and the adjacent headwaters of the Fraser River system in the region called the Nechako Plateau or Lakes District, a rolling upland with forested ridges, grassy meadows, wetlands, and many lakes. Oral histories and phonological evidence suggest ancient residence in the Bulkley Valley area and prolonged cultural and linguistic contact with the neighboring Gitksan (Rigsby & Kari 1987). The Witsuwit’en share the concept of bounded territories as owned property with the Gitksan, and knowing boundary markers and history is likewise important. For an extended discussion of Witsuwit’en territory and customary law, see Mills (1994, 2005). For Witsuwit’en, Clan may be more important than the specific House in shaping territory use. Complex arrangements determining access were characteristic of both Witsuwit’en and Gitksan and involved both one’s own rights, those of the spouse, and those of the father. As with Gitksan, inheritance of titles, House and Clan affiliation is through the mother. However, in contrast with the Gitksan and in common with other Northern Athapaskans (Dene), the Witsuwit’en seasonal round typically involved winter dispersal to distant hunting and trapping areas and convergence on highly productive fishing sites in summer. For the Witsuwit’en, this was spatially restricted canyon salmon fishing sites at Moricetown (Kyah Wiget) and later Hagwilget (Tse Kya or Tsë Cakh), where there were permanent villages established and a summer potlatch season. As with Gitksan, there were specific more restricted properties such as fishing stations in the canyons, as well as hunting territories and rich resource areas like groundhog grounds. There were also named, owned berry patches. For both Gitksan and Witsuwit’en, management of owned berry patches by fire has been documented (Johnson 1999; Daly 2005; Trusler & Johnson 2008).

. The terms indicate Male Chief, Female Chief, and Heir. . Trespass by a member of another nation was punishable by death, and oral histories, such as the raid on the Kemano people by Kweese, detail wars started by such killings (Pat Namox, interview notes).

296 Leslie Main Johnson

3.3

Kaska

Kaska Dena live in the drainage of the Liard and Pelly river systems in northern British Columbia and the Yukon Territory, including the slopes of the Logan Mountains along the Northwest Territories boundary. The landscape is mantled in a boreal forest mosaic of forest and wetlands, with lakes of a wide range of sizes in lower elevations and valley bottoms, and is traversed by major rivers. The Cassiar Mountains, Northern Rockies, Simpson Mountains and Logan Mountains all cross parts of Kaska territory, providing access to large alpine areas and rugged slopes. Kaska is an Athapaskan language, which is, like Witsuwit’en, a synthetic language with templatic morphology (Kari 1989a) and which, unlike Witsuwit’en, retains tone. There is some dialect range within this region, and the reference of the term ‘Kaska’ has expanded over time from the local groups near Dease Lake in the Cassiar Mountains to encompass people of a number of local groups on the Lower Liard, the Upper Liard, the Dease River, the Frances Lake and Pelly Banks area, and the Pelly River in the area now known as Ross River. The present Kaska Dena Council also includes the Fort Ware people, who have been described as Sekani. The language of the Fort Ware people in particular shows some similarities in lexicon to Witsuwit’en. In contrast with both Gitksan and Witsuwit’en, Kaska have a highly egalitarian society based traditionally on hunting large ungulates, trapping, river and lake fishing for whitefish and grayling, and hunting for small game. People had extensive traplines and family hunting areas and tended to gather at fish lakes, or later trading posts, in the summer season. Seasonal movement among harvesting areas and shifting of areas of hunting focus in response to local movements and abundance of game is part of the adaptive interaction of Kaska with their homeland. Salmon are lacking in the Liard River system, but are present in the Pelly and in the Stikine. There is also a long history of use of alpine areas for both caribou hunting and snaring of gophers, Arctic ground squirrels documented for the region (Kuzyk et al. 1998; Hare et al. 2004; Andrews et al. 2009). Caribou hunting and moose hunting remain very important to the Kaska mixed economy and to their identity. River fishing is also popular, and people still engage in fur trapping and trading. Small game like grouse, ptarmigan and rabbits are also taken. Cranberries, blueberries, strawberries, raspberries and crowberries (locally called “blackberries”) are prized in season. Though I have not recorded evidence of use of fire for habitat management for the Kaska, wildfires are common and extensive in the boreal forest, and evidence of past burning is certainly recognized, as are ecological changes immediately following fires.



4.

Language, landscape and ethnoecology, reflections from northwestern Canada 297

Methods and approaches to landscape research

The research described in this paper was carried out over an extensive period from the mid-1980s to the present. Aspects of this research are reported in Johnson (2000, 2008, and 2010a), Johnson & Hargus (2007), Trusler & Johnson (2008), and Johnson & Hunn (2010). My research methods have been eclectic, reflecting my background in ethnobiology and cultural anthropology, my training in western-oriented field ecology, and my local residence in the Skeena Valley from 1977–1991. I initially approached Gitksan landscape knowledge from the perspective of elucidating habitat knowledge for culturally important plants, recognizing that habitat descriptors in local English did not easily translate into geographic features I could recognize. This caused me to realize that, like for biological kinds, terms for landscape features could not be presumed to have a one-to-one correspondence between languages, and that local knowledge and terminology for the features of landscape were worthy of investigation. I have used a mixture of linguistic and visual methods for both elicitation and recording of landscape information and have carried out informal conversations and interviews in peoples’ homes (primarily Gitksan and Witsuwit’en), and I have also spent time on the land and at formal language learning workshops where ‘land’ words and habitats were discussed (primarily Kaska). Gitksan and Witsuwit’en research employed photo elicitation (Collier 2001) while the Kaska work involved more photographic documentation in the course of travel on the land. I have used local dictionaries and bilingual recorded narratives as well to glean further landscape-related vocabulary in all three languages. There is a rich set of landscape terms included in the Kaska language book Guzāgi K’ū´gé’ (Kaska Tribal Council 1997), but the discussions which follow are based largely upon my own work and workshops in which I was a participant, as the referents and elicitation contexts of the dictionary committee’s work were not known to me. I have also analyzed narratives and informal conversation to gain a sense of how landscape is spoken about and how terms for features of lands, waters and vegetation are used. My work has focused on generic terms for kinds of place rather than elucidating insights on understanding of and relationship to land from the names of specific places, though I have examined lists of place names for all three groups to gain a sense of kinds of features named, and the linguistic derivations of the names (Johnson 2010a).

298 Leslie Main Johnson

5.

Results

The insights into cultural understanding of local landscapes, ecological concepts and meaning of the land can be conveniently discussed in terms of the array of place kinds recognized, the local terms elicited and English glosses for these, and a sense of how landscape is perceived in ecological terms. In presenting and analyzing landscape vocabulary, I have found it convenient to break down the corpus into sets of terms that pertain to: topographic features, substrates, hydrographic features, snow and ice features, vegetation, hunting-related terms, terms related to human geography, and terms for sacred or powerful places (see tables throughout this chapter). Other ways of grouping terms might also be effective, and the groupings I have made are not necessarily faithful to local conceptions. One might instead group vocabulary according to geographic areas within the landscape, such as mountains and uplands, rivers, wetlands, and so on. Wetlands, in particular, are difficult to place with reference to ‘water’, ‘substrate’ or ‘vegetation,’ necessarily combining all three elements. I found some terms difficult to place, such as ‘willow along the creek down there’ – is that ‘vegetation’ or ‘water’? Similarly, should timberline be tallied as ‘vegetation’ or ‘mountain’? These questions particularly highlight the difficulties separating vegetation from topographic features or hydrographic features, and of formal categories from ad hoc designations. The landscape is really all of a piece, and demarcations are often rather arbitrary (Mark & Turk 2003).

5.1

Topographic terms

Terms for topographic features such as ‘mountain,’ ‘ridge,’ ‘valley’ and so on are common to all of the sets of landscape terms I have recorded. It is always somewhat arbitrary to separate out topographic features from other types of feature sets, such as snow and ice features (including glaciers) and river features (including sand bars and canyons). Gitksan country, as I have indicated above, is quite mountainous, with extensive alpine, steep sloped valley walls, and the wide glaciated valleys of major rivers such as the Skeena River itself; and features such as kame terraces or river terraces and various forms of scree and talus are also found. The bulk of Gitksan territory falls within the Coast Mountains and the major rivers and tributary valleys and extends into the more subdued Bowser Basin and Kispiox valleys, reaching northward to the divide with the Nass, Skeena and Stikine drainages. The Gitksan landscape is highly mountainous, for the most part consisting of glacially scoured ridges and higher peaks which support permanent snowfields and glaciers as well as avalanche and debris avalanche tracks.



Language, landscape and ethnoecology, reflections from northwestern Canada 299

Slopes are densely forested with mixed coniferous forests dominated by western hemlock and red cedar with subalpine fir, mountain hemlock, hybrid spruce, lodgepole pine, and some deciduous trees in more rugged areas. The division of the landscape into bounded territories comprising side drainages and portions of mountains flanking the main rivers reflects seasonal utilization of the full range of altitudinal zones and indicates travel through the mountains by trail before the construction of modern roads in the twentieth century. Witsuwit’en country also includes a fringe of steep, forested and glacier-carved mountains with extensive alpine which are comprised of the Eastern slopes of the Coast Mountains, but the Bulkley valley is quite broad and mantled with various types of glacial, outwash, and colluvial sediments. There are hills formed by outcroppings of recent volcanics and large lakes as one moves eastward into the interior, where topography is more subdued. Wetlands and lakes of various sizes are common in less precipitous areas. Kaska country in the north is mantled by boreal forest, and is a mosaic of wetlands and hills, rivers, lakes, ponds, and glacial deposits together with areas of rugged mountains and both grassy and rocky alpine in the Cassiar Mountains, the Northern Rocky Mountains, and the Logan Mountains. Table 1.╇ Topographic terms and substrates Indigenous terms Gitksan terms laxk’elt k’elt E, k’ilt W sga’nist ts’i’winhl sga’nist gililix ts’imts’uu’lixs ts’imt’in usim ges ts’ilasxw biiyaakhl hahumxsim lo’op ts’imts’ilaasxw kslo’op tsaldem lo’op lo’op sdaats’isda ksiip xsiip

Approximate English equivalents hilly land top of hill, hill crest, ridge line, summit mountain mountain peak, summit upland ‘gully,’ ravine ‘valley’; basin a narrow place on the mountain rock canyon (as in Kitselas, ‘people of the rock canyon’) sheer cliffs rock wall, sheer cliff (as the headwall of a glacier) a newer way to say cliff rock face where there’s lots of thin shale piled up rock, stone, small rock hill big square boulders on the side of mountains, blocks which have fallen from a cliff ‘black shale that slides’ talus accumulation under cliffs; also unstable scree or inside of moraine sand

Translation

‘in the gully’ ‘in the valley’

‘wall of stone’ ‘in the canyon’ ‘thin rock’

300 Leslie Main Johnson

Indigenous terms

Approximate English equivalents

laxxsiip psa E, pse W yip

sand area, beach soil and clay earth, land

Witsuwit’en terms cis cis k’it dzilh dzilh tsëykh

hill, ridge hill slope mountain pointed mountain

ts’ik weggiz, wiggiz dzilh k’it dzilh ïggiz, dzilh ggiz wenin ’is wibegh wididlin tsë bï hon’a c’i’an yin bït tsë tsë hadït’ay say say k’it say titgut

ridge pass summit (lit. ‘on mountain’) mountain pass sidehill cliff cliff cliff cave den, hole, cave swale, valley, depression [any size] rock, boulder ‘moraine’ (lit. ‘rock sticks out’) sand, fine gravel sand bar quicksand

Kaska terms tsé’ hés chō, hés tlétāgi, gunentsįndle kéhnesí’, kę´ nazī´ tsātl’ah tātege kādejhalé witl’ús ts’ā zā chāzā tsį´h nán

Translation

mountain (esp. rock mountain)* mountain top of the hill hillside [south side, sun side], sidehill (open south-facing slope) valley in the mountains ‘upwards’ saddle, “where mountain joins” valley, gully, creekbed and along the creek (mountains) when it’s all sand, sandy area quicksand red ochre earth (also “the Land,” which is sacred or powerful)

* Rock mountains and mountains with grassy meadows and more subdued topography are differentiated.

Gitksan landscape terms recorded to date include a number of terms for topographic features, which grade into terms for features like talus accumulations and block accumulations at the base of rock faces. See Table 1 above and Johnson (2000 and 2010b) for further discussion. I recorded little that relates to soils;



Language, landscape and ethnoecology, reflections from northwestern Canada 301

terms for basic substrates were present, but for Gitksan, who did not traditionally cultivate crops, an elaborate terminology indicating different classes of soils which may require different handling when clearing and cultivating gardens is not needed. Witsuwit’en landscape terms also reveal a number of terms for topographic features, including hill slopes, ridges, mountains, hills, mountain passes, cliffs, valleys, and moraines or rubble fields. In comparison to Gitksan, there is more elaboration of terms which indicate direction or relationship in their semantic structure (characteristic of Athapaskan languages in general; see Kari this volume, and Moore 2002). Substrates such as sand, quicksand, and rock are distinguished. Like Witsuwit’en and Gitksan, Kaska landscape terminology has terms for mountains, hills, peaks, valleys, and the like. Kaska have terms for south slopes and terms, like Witsuwit’en, for bare areas. Substrates such as sand, quicksand and rock are distinguished. Discussion of topography prominently involved locative words which place speakers on the landscape and moving, up slope or down.

5.2

Water terms

Water terms are prominent in all three areas (see Table 2). Large rivers, important for fisheries and as routes or obstacles to travel, traverse all three regions, and a wide array of lakes, ponds, smaller creeks, sloughs and backwaters are also recognized. Dividing off water terms is somewhat challenging, because features of the banks, including their composition and topographic prominence, and wetlands both blur the land/water divide. In this region of permanent water, discussion of waterways without the banks or margins does not make sense, so such terms are included here. Wetlands are more prominent in Witsuwit’en country and especially in Kaska country, where extensive areas of permafrost and rolling glaciated topography help to produce relatively large areas of peatlands. In contrast, aspects of the rivers themselves, ponds, lakes, tarns, waterfalls, and the like are more conspicuous in Gitksan country, though vegetated swamps certainly occur and are ecologically important. For all three groups, obvious features like river, confluence, waterfall, eddy, canyon, lake, and island are named. All three groups also named ‘high bank’, a feature of river shores highly significant to canoe upstream navigation, where tracking at the base of such high banks is challenging. They are also significant landmarks in river and overland navigation. All three languages also distinguished creeks flowing into, and out of, lakes. This would be anticipated for the highly relational linguistic style of Athapaskan languages, but such terms were also provided by the late Peter Muldoe (Gitludaahl) for Gitksanimx. In the Athapaskan languages,

302 Leslie Main Johnson

Table 2.╇ Water terms, including river terms but excluding wetlands and swamps Indigenous terms

Approximate English equivalents

Translation

Gitksan terms aks; xsi- [xsan, xsu-] baam’aks golim’aks k’ali’aks

river, stream running water, stream running waters or streams large river

a form of the term for water

wilnaawadihl’aks

confluence, where rivers come together

t’aamiks ts’oohlixs

tl’ook’

pond; slow side channel back channel, deep embayment, doesn’t have current a real back eddy, with current, where you set net a real back eddy, with current, where you set net standing waves, rapids ‘hilly water’ a boil ‘swelling’ flat water, a quiet place waterfall bay bay whirlpool impassable whirlpool, as at Kitselas Canyon; ‘maelstrom’ island sandbar where the water barely covers a rock, but there is no wave when sticks and leaves snag on a rock that’s just at surface high, dry river bank steep bank of a river steep river bank, steep eroded river bank cutoff, steeper river bank, cf. bluff rising water, keeps coming up, or ‘swelling’ high water, flood stage lake where a creek flows in from the back of the lake a creek that flows out of a lake a spring (not a swamp) where moose go, a muddy place;  ‘place of mud?’ salt lick, black mud mud

Witsuwit’en terms c’ikwah c’ikwah yez

river, stream creek

luuguuksbax luuguuksbax’ aks laxk’elt aks gitxw, gitwhlaks lemksimks ts’itxs k’aldixgaks wil luulamjax ts’a’lixs antk’ulilbisxw laxlikst’aa wisax / wisex ts’iliks ’niilok gwildim aks namk E nemkap W ’wiinamk’ E pteliks disleks t’ax; t’amsagalaan t’ax xsi t’ax gwanks E; gwenks W antl’ook’ E en tlook W

a form of the term for the upstream direction ‘where the waters get to know each other’

‘small stream’



Language, landscape and ethnoecology, reflections from northwestern Canada 303

Indigenous terms

Approximate English equivalents

tacëk bin k’ënlï bin ts’anlï, tëzdlï bis k’it diyik tlatine

mouth of creek or river creek flowing into a lake creek flowing out of a lake bank (of river, etc.) canyon eddy

talhtis tëwhilh dzen hanlï tak’iz k’ët nendli bin tadïz’ay tabegh binbegh nu

deep water, strong current deep water muddy water spring spring waterfall lake little pond, backwater shore lakeshore island

Kaska terms tū, chū man chūkéli * chū kinēli tūłídli kudache tsezele tū jedé’ tū dinídislīn tū wit’ī´lī, dzínadlī´n łénedlī´˛ł etū´de, deskwidle tahmā, tūmā manmā tsēkāge tsēza tse dzéh tsedzeh tamā tādzai tl’ átagī, bes

Translation

“Means the water is kind of€– backed into kind of pool”

water, river lake water flowing [=creek?] along the creek confluence where creek comes in spring* hot spring* waterfall large meander loop ‘long way around’ (on site and off site) whirlpool, eddy (from dictionary) island shore line, bank lake shore gravel shore, rocky shore, beach gravel shore, gravel rock canyon rock shoreline high rock bank [like along Yukon River] high bank

* The Kaska words for ‘spring’ and ‘hot spring’ are Liard dialect, from Kaska Tribal Council (1997); Elders used the English terms when conversing with me about these kinds of place.

304 Leslie Main Johnson

names like ‘its bank’ or ‘shoreline’ were evident, as were terms indicating the substrate of shoreline features (see especially Kaska terms). A series of Gitksan terms denote features of rivers which are significant for travel safety and for fishability as well, provided by the articulate and bilingual Art Mathews Jr. Dinim Gyet, an accomplished river salmon net fisher. In particular, the careful delineation of different kinds of backwaters with different amounts of current, significant for fishing, stood out. Eddies are also recognized and named in Witsuwit’en and in Kaska, and are important in river fisheries for salmon for Witsuwit’en and for grayling in Kaska country. ‘Spring’ was named in Gitksan and in Witsuwit’en, but the Kaska term was not recorded in my conversations with Kaska elders. The Kaska language book (Kaska Tribal Council 1997) records tsezele for the Liard dialect. Springs represent clean water sources and are particularly valued for healing use. Witsuwit’en and Kaska have a particular term for the mouth of a creek as well as for a confluence; the terms appear to be linguistically cognate or related.

5.3

Snow and ice terms

I have argued that in places with long winter seasons, snow and ice features are ecotopes or “kinds of place” (Johnson 2000, 2010a; Johnson & Hunn 2010). Glaciers and multi-year snowfields, where they occur, are persistent features recognizable and mappable in any system of geographic ontology. Features of ice or current year snow are more ephemeral, but nonetheless essential to understand for safe and effective travel on rivers or lakes, overland, or through mountains. They also effect the movements and ability to feed of other animal species, as well as the ability to procure fish in winter. A linguistic indication of the pervasive importance of snow is that the root for ‘snowshoe’ is one of the ancient roots in the Athapaskan language family. Table 3 details snow and ice terms; terms are recorded from my own field research except for Kaska, for which they are taken from the Kaska language book (Kaska Tribal Council 1997), as I have not conducted fieldwork there in the winter season. Recognition of the qualities of ice cover seems particularly richly developed for Kaska and Witsuwit’en, while qualities of snow (wet versus powdery) were well developed in Gitksan. The distinction of multi-year snowfields was also found for all three groups, and may be implicated in knowledge of the behavior of alpine species that are hunted, such as mountain caribou and mountain goats and sheep. . Hot springs and cold springs are differentiated in Kaska conversation in English. Cold springs are good sources of water, and hot springs are valued for bathing.



Language, landscape and ethnoecology, reflections from northwestern Canada 305

Table 3.╇ Snow and ice terms Indigenous terms Gitksan terms sbeek bumksim maaxws bumks yeesims

Approximate English equivalents

s’yunim maaxws s’yun hlo’omks ’wiluks

cornice powdery snow like sugar powder, powder [snow] powder snow that blows all around; even snowshoes don’t hold you up sticky wet snow snowdrifts, powdery blowing snow, any kind of blowing snow snowfields, snow on glacier glacier wet snow in early spring; cohesionless ‘slide snow’ wet snow that doesn’t stick together (from calendar)

g’ipx pdaalast

‘frozen over’; river ice that a person can walk on water on ice

lulitx

candling ice

’moos ’muuxws

Witsuwit’en terms lho lho tl’ët lhim tin yis lhk’ëc’ots’iyh witlat, tiltlat, witiltlat witlatn, witlat k’it tinïnel winïnil tin telhtla Kaska terms tan ųts’íh, kénets’íh łēhtené’ tan akadaston tan cho’ ten ts’ih tan łagāton’ úhtsíh dēhtu’

glacier glacier foot ice (chunk) ice (flat) snow snowdrift, snowfield slide, avalanche avalanche track, slide area “ice is getting soft” “they say when you can’t walk on it [the ice]” “When the ice start to go, eh, river takes it away … pretty near like an avalanche” ice “glacier” (overflow ice) glacier (permanent) broken ice chunk ice clear ice cracked ice glare ice overflow, slush

Translation and notes

‘it sticks’

from the verb ‘to slide’ “the sun hits your trail in snow and it gets wider” (either in cold weather or in March) “the sun hits the ice in March and it becomes like icicles”

lit. ‘glacier front’

‘big ice’

306 Leslie Main Johnson

Indigenous terms

Approximate English equivalents

íhts’íh tan mā tezū´s, ten zús tagolé’ tan astsįdze tizę¯ tan dakít tan da’adze ten deténéston lî dahtané’ chuhtłąji kiditlah, zas dzedetl’ich

overflow ice ice edge loose ice, honeycomb ice rough ice shell ice slush ice thick ice thin ice ice jam crusted snow snowdrift snowslide, avalanche

5.4

Translation and notes

Slides

Related to snow and ice terms is a class of terms most clearly developed in Gitxsanimx: ‘slides.’ In the precipitous mountain country of the Gitksan, steep slopes limit snow stability, and avalanches and avalanche tracks are common features. However, these same steep slopes and unforested trackways may also be prone to debris avalanches and other types of slide activity; thus the Gitksan verb for sliding forms the root of terms for several types of slides. Table 4 is a short table of terms denoting various types of slides and slide areas in Gitksan, Witsuwit’en and Kaska. Some of the terms also appear above in Table 3, as they are both snow terms and slide terms. Again, the adaptive value of recognition of slide areas in mountainous terrain, and of understanding the risk of snow, rockslides or debris avalanches and what conditions they occur in, is evident. The habitat association of the rich and tangled regrowth on slide areas with certain food plants (ax or spiny woodfern rhizome for Gitksan) and with animal habitat (black and grizzly bears find abundant food on laxensuuks) provide other reasons for recognition of these kinds of place. Witsuwit’en elders emphasized that one must not camp near a creek (that is flowing down an avalanche area) in late winter or early spring because of the risk of a slide, and other narratives described ways to recognize when a trail crossing a laxensuuks was safe to traverse (Dinim Gyet narrative in Johnson 2010a:â•›Chapter€3). The significance of major debris slides on the landscape is underscored by Elders’ comments that ‘rockslide’ (Kidizah) is the name of the creek flowing into the place called Good Hope Lake in English, and should be the name of creek and lake (Angel Carlick, notes, August 29, 1998).



Language, landscape and ethnoecology, reflections from northwestern Canada 307

Table 4.╇ Slides Indigenous terms

Approximate English equivalents

Translation or comments ‘it slides’ ‘place-slides?’

hlo’om sga’nist ’yagahlo’o hlo’om gan laxensuuks

slides avalanche track, place where it slides every year rockslide or landslide snowslide, avalanche blowdown? or a landslide involving trees? landslide or snowslide scar; has slide alder

Witsuwit’en terms witlat, tiltlat, witiltlat witlatn, witlat k’it

slide, avalanche avalanche track, slide area

Gitksan terms hlo’o enhlo’ W, anhlo’o E

Kaska terms wítl’at kiditlah, zas dzedetl’ich kidizah

5.5

slide area (can refer to rockslides or snowslides) snowslide, avalanche; also rockslide rockslide

‘slide-mountain’ ‘timber avalanche’ ‘logs on it’; the suuks are the down logs on the slide area

cognate with the Witsuwit’en term and the Fort Ware term from dictionary real name of Good Hope Lake

Vegetation

Vegetation is described or named by all three groups. Northwest British Columbia and the southeast Yukon are both predominantly forested landscapes, and therefore places which do not have tree cover appear to be more salient and worthy of designation. Treeless places in fact are the marked condition. The Skeena Valley occupies a transitional zone in the Coast Mountains and the northern territories are transitional to boreal forest. Formally the vegetation is described as variants of the Coastal Western Hemlock Biogeoclimatic Zone, the Interior Cedar-Â�Hemlock Zone, the Sub-boreal Spruce Zone, and higher elevation Mountain Hemlock Zone and Engelman Spruce-Subalpine Fir Zone (British Columbia Forest Service 1988). Above this lies the rich alpine recognized as ‘meadow’ and used for a number of key resources including mountain goats, formerly caribou, ‘groundhogs’ (hoary marmot), berries, and medicines. Witsuwit’en territory ranges from the more coastal forest types in the inner edge of the Coast Mountains to broad areas of Sub-boreal Spruce, where white spruce, lodgepole pine (now decimated by mountain pine beetle) and seral communities of aspen and other deciduous trees predominate. Witsuwit’en recognition of alpine vegetation and resource use is similar to Gitksan. Kaska country is mantled with boreal forest and has extensive alpine zones above about 1,220 m (4,000 feet). For Kaska too, alpine zones are

308 Leslie Main Johnson

seasonally important for caribou and medicines, and higher elevations (the dwarf birch zone recognized by ecologists) also provide berries. Designation of vegetation seems somewhat ad hoc and fluid, especially in terms of plant communities. Physiognomy of vegetation is evident, with terms indicating ‘meadow,’ ‘brush’ (often called ‘willows’ by Athapaskan speakers in English), and features like timberline. For Witsuwit’en, the concept of edge extends to several types of ecotones where physiognomy differs across the boundary, extending from timberline (an edge marked in colloquial English) to ‘edge of mountain alder’ – the abrupt edge of an alder thicket growing on deltaic gravels (Dan Michell, photo notes, August 19, 2005). Similar ‘edge’ terms can be discerned in Kaska. Included in Table 5 below are wetlands, locally called ‘swamp’ or sometimes ‘meadow,’ which comprise an important class of (relatively) treeless habitats that are important for a number of resources and as animal habitat. Willow thicket areas are one of the forms that wetlands may take, and these are often critical winter habitat for moose. Sedge meadows are another. Labrador tea and sphagnum moss (for diapering) are characteristic of sphagnum peatlands. Sometimes characteristics such as winter snow character are among the defining traits of open areas, which are easy to traverse in winter on snowshoes (and more recently by snow machine). Table 5.╇ Vegetation types Indigenous terms Gitksan terms lax’aamit lax’amaaxws laalax’u sbaaytgan sbagaytgan sbagaytgangan sbagayt-am’mel sbaayt sginist sbaa ts’ex sbagadegantx laxsga’nist am ‘melmgaliaks luulaxsuuks gakslax sga’nist

Approximate English equivalents

Translation or comments

meadow (snowbed areas and other treeless places) meadow (alpine and other treeless flats) swamp, wet meadow, muskeg forest forest mixed forest

“place that’s good, that has no trees”; ‘prairie’ ‘prairie’

‘among the trees’ ‘among the trees trees’ (reduplicated) cottonwood forest ‘among the trees, cottonwoods’ pine grove, pine stand ‘place where there’s pines’ scrubby coniferous growth (juniper), krum- ‘juniper place’ holz (timberline) forest ‘out in the bush, in the forest’ forest area if it is up a mountain floodplain cottonwood, cottonwood-alongthe-river dense scrub regrowth in old slide area timberline (the actual line dividing forest growth from alpine)



Language, landscape and ethnoecology, reflections from northwestern Canada 309

Indigenous terms

Approximate English equivalents

ts’i’naast

burnt over patch (for berries or deer browse); clearing burnt over area ‘place that is burnt or charred?’ “all the timber coming up again” after the burn berry grounds too much brush or undergrowth on the berry patch berry patch trail berry patch trail

lax’anmihl lumks tsee gantx ansimaa’y maaxsgan genimsimaa’y E ginimsamaa’y W Witsuwit’en terms ts’ikh scinlegh wize begh wik’in k’it widïnk’in’ k’it niwdïzk’an nit’ay k’it wizulh k’it tl’o k’it tl’otl’is (k’it) c’iye (k’it) ts’al k’ët witsil k’it c’ato’, lht’ato’ k’ëndlih c’ato’ dic’ah, dicin tah widits’itl ts’o co tah tighiz co tah k’ëndlih co tah Kaska terms tūtsel, tūtsel mā tūtsel tu łetese elīn, łetesgwech’edi łíni tsēlē’

“mountain juniper” dwarf trees at timberline, krumholz timberline timberline burned area burned area burned area berry picking ground open area, also above timberline; bare ground with nothing on it meadow, open grassy area (e.g. a lawn, a grazed slope, alpine meadow) meadow, marsh (where large grass grows, e.g. sedge meadow) swamp (where moss grows?) swamp damp place swamp willow swamp (in the) bush (in the) bush, among the trees/sticks (for mixed forest) thick growth, “jungle” “big spruce country,” spruce forest area “big poplar country,” poplar woods “big willow country,” willow swamp, willow thicket swamp slough (wet sedge meadow, no current) slough slough slough (narrow; wooded shores)

Translation or comments

‘edge of alders’

‘grass on it’

(lit. ‘diaper place’)

‘it’s really brushy’

(lit. ‘among big willow’)

term seems to be related to the word for creek in Fort Ware

310 Leslie Main Johnson

Indigenous terms

Approximate English equivalents

tū tī´li, tū tilī˛

slough (wide, recently abandoned main channel) moss (sphagnum and sites with sphagnum) clearing, dry hillside, dry clearing open place burned place [long ago burned] small trees, poles on old burn (mountainside?) alpine, mountain alpine, mountain meadow [lit. grass] (from dictionary) white spruce forest forest, timber, tall trees

ts’ātl gunedęje gunejęje, wanitsįndle chōladé, sa’a gukādek’ān desk’ese tses héskage nedudze hése tl’oge gat chō tah, gat tah dechen chō, łedu tah, łedu, tahche tsezel, detah néhsban gō˛dze tah gó˛dzets’edlī ts’adli naw’a gū´le’ chō tah, gū´le’ chō chūkéli gū´le neyé chū kinēli gū´le dā’a

open small lodgepole pine area lodgepole pine stand “pine brush” “brush” (thick coniferous growth) “brush” (probably thick coniferous growth) willow along creek or river willows growing along creek brushy willow area along creek

Translation or comments

‘among big willows’ this may be an ad hoc description rather than a fixed term this may be an ad hoc description rather than a fixed term

Some of the terms listed above might as well be placed in water terms, as wetlands comprise site types with a suite of hydrographic, physiographic and vegetation traits. In particular, the terms for various kinds of features translated as ‘slough’ are ambiguous in their proper placement. Terms are also found which recognize fire as a factor on the landscape, and Gitksan in particular has special terms for brush on berry patches, formerly managed by Gitksan and Witsuwit’en by landscape burning (Trusler & Johnson 2008). Kaska recognize fire and its effects, but apparently did not manage berry crops through burning (Linda McDonald, personal communication, 2010).

5.6

Features of the cultural landscape: Hunting, trapping, trails, camps  and powerful places

A heterogenous array of significant cultural ecotopes deals with human occupance of the land and reveals important aspects of ways of living on the land. This group of place descriptions includes terms related to hunting and trapping, trails, places of habitation. In contrast to places of human use and habitation



Language, landscape and ethnoecology, reflections from northwestern Canada 311

Table 6.╇ Features of the cultural landscape Indigenous terms

Approximate English equivalents

Translation or comments

Hunting and trapping areas, trails, campsites and villages Gitksan terms ensimetx traditional hunting areas [for mountain goats] metx is mountain goat gena metx goat trail ginimxsga’nist goat hunting trail ‘mountain trail’ genimsiilinasxw hunting trail genim jap trap trail genx trail ksdaamoos hand or foothold on cliff endilgan beaver dam goot beaver lodge, from the shape when the pond is dry ‘heart’ antl’ook’ E, en tl’ook W a muddy place, used by animals as a mineral lick ansinhun E, ensinhun W place away from the village where you do fishing anjok E, enjok W campsite, dwelling place (e.g. berry camp, fish camp) lax galtsap village Witsuwit’en terms lhiyil c’itiy wikë tiy ts’ikh witlusbït tsa ’ilh c’ikën co’ën k’it c’itok’ët lhëtl’is c’ididlet Kaska terms kedā tane gah tane híh tane atane alē´s tlétāgī

goat trail foot trail dwarf trees at timberline, krumholz (goat bed at edge of cliff) “just a green, green area, where they go out and [get that] groundhog.” (ditnï) beaver dam (beaver) lodge lookout moose watering hole; salt lick mud lick

ko˛ tsa ko˛

moose trail rabbit trail game (animal) trail (from dictionary) trail lick [can specify by species that uses it] lookout [e.g. Billy Lake and McKinnon Lake lookouts] camp beaver lodge or house

Spiritual places Gitksan term sbilaxnok

a place of spiritual power and danger

contains the word for ‘eye’

possibly a particular use of the term for ‘high bank’ ‘beaver camp’

312 Leslie Main Johnson

stand powerful places, which are often remote from habitation (but not always), and which require particular care to demonstrate respect and proper behavior. Hunting and travel on the land are important domains of cultural knowledge for all three groups. The significance of mountain goats for the Gitksan, marmots (ditnï ‘groundhog’) for Witsuwit’en and moose for Kaska were evident. Small game also figures, as both ‘moose trail’ and ‘rabbit trail’ were taught to me while berry picking with my Kaska Elder teacher. Beaver dams and beaver lodges were also named, underscoring the historic importance of beaver for both subsistence and fur trade economies. All three cultures also recognize salt, mineral or mud licks for animals, which appears to be universal among hunting cultures (see discussions in Johnson & Hunn 2010 and Johnson 2010a). Both Kaska and Witsuwit’en have words for lookouts, places where a hunter can gain an overlook of an area of potential swamp or meadow areas where animals may be feeding. The linkage of lookout, trail, and camp, and with meadow or tūtsel area were made clear travelling with my Kaska Elder teacher, and helped me to understand better things I had earlier been told by Witsuwit’en elders. Places of human habitation are recognized, and to some degree form a contrast with ‘out on the land.’ Gitksan, who had high pre-contact population densities and permanent villages, have a word for ‘village’, lax galtsap, which differs from the word for ‘camp,’ anjok [place where you camp]. Kaska usually dwelt in smaller groupings, and travelled widely on their lands; for Kaska camp is ko˛. I found the extension of the term ko˛ to the dwelling of the beaver perhaps telling. I did not record Witsuwit’en terms for village or camp, but the word for ‘house’ yikh appears to be incorporated in the term for ‘our village,’ niwhkayikh, included in the local publication Tsë Cakh Wit’en (The Hagwilget (Tse-Kya) Band 1995). Sacredness of land itself is fundamental in all three cultures, and especially powerful places and places of history are recognized (Johnson 2010a, b). A special term for a dangerous kind of power place was only recorded in Gitksan. Such places should be avoided by those with insufficient power to protect themselves, and their locations are not generally divulged to avoid intrusion by the unprepared.

5.7

Orientation and movement on the land

Thus far, I have described terms from a series of thematic ‘kinds of place’, which ultimately presents a static view of landscape. People orient and move on landscape as well as designating its features. Athapaskan languages have enormously elaborated terminologies for designating relative location (e.g. The Hagwilget (Tse-Kya) Band 1995; Hargus 2007; Moore 2002) which, I have argued, places people in motion on the landscape (see Johnson 2010a). In mountainous landscapes,



Language, landscape and ethnoecology, reflections from northwestern Canada 313

Table 7.╇ River and slope directional terms Indigenous terms

Approximate English equivalents

Translation

Gitksan terms River directional terms, upland/lowland – axes of orientation gyeets downstream area, down river gigeenix upstream area, up river gew open area near the river gililix uphill area away from the river Witsuwit’en terms Drainage and slope directional terms c’ikwatsa’ downstream dikwanikh ‘facing upstream’ yikhina down the slope dikhina up the slope Kaska terms River and slope directional terms ahdā ahdege, tahgáh dege (tahgáh) nō´nē´, (tahgáh) anúsd첒 kúda digé kúda ats’ā´

downriver [from dictionary] upriver [from dictionary] across (the river) [from dictionary] uphill [travelling uphill] downhill [travelling downhill]

orientation by cardinal directions is less meaningful than orientation by slope and drainage direction. Gitksan, a Tsimshianic language, shares this basic set of perceptions with Witsuwit’en, and the landscapes occupied by the two groups make this natural. Kaska, along with other Athapaskan languages (Kari 1989b and Kari, this volume), also focuses on direction of drainage, and of upslope/downslope (see Table 7). There are also sets of terms in all three languages which designate relative location (under, over, among or across, beside, behind, ahead). Gitksan tends to focus on relative position of features or people – ‘on mountain,’ ‘beneath the pitch pines’ – while Athapaskan languages have highly elaborated and inflected locative words and deictics which carefully designate direction and distance from the speaker, language well adapted to people of the trail. Athapaskan languages differ in fundamental ways from other language families in the prevalence and elaboration of relative directional terms. Gitksan shares with the two Athapaskan languages the importance of river and slope direction, as opposed to a cardinal direction system, for orientation. However, the elaboration of what one might see as river- or trail-based relative distance and direction is not shared, and is not apparently fundamental to Gitksan perception of place in the world.

314 Leslie Main Johnson

Table 8.╇ Spatial and directional terms Relational terms (after Johnson 2000) Gitksan terms

Meaning

Translation

Example

miin win lax an

at the foot of; in front of place of, place where ‘on’ where, place when

beneath#mountain

miinhl sg’anist

on hills#water where#mud (mineral lick)

laxk’elt aks antl’ook’

These relational terms occur both in place kind generics (terms for landscape elements or ecotopes) and in toponyms. Witsuwit’en system Directional adverbs set 1* Witsuwit’en term

Term for an Indefinite, Unbounded Area

Meaning

dik yik nik tsi’n nu’ de’ nis nï’ ‘in, ‘ats nyen

diwih yiwh niwh tsewh nuh dewh nisewh nïwh ‘awh neywh

up down uphill downhill upstream downstream ahead behind side across

* These can be adverbs or postpositions; postpositions have modified forms. Source: Tsë cakh Wit’en, Hagwilget (Tse-Kya) People, traditional words and texts from Band members (Hagwilget Band 1995:â•›53). There is also a full set of prefixes that indicates distance from the speaker with five gradations, ranging from closest to speaker to furthest from speaker: de-; n- ; wi- ; u- ; ï-. (Hagwilget Band 1995:â•›53) Directional adverbs, set 2 At place

From place

General area

Meaning

niggit tset nut det diggit yiggit nisit nït nat nyedzit

niksdïts tsets nuuts, nusdïts dets, desdïts diksdïts yiksdïts nisdïts nïts, nïsdïts ‘ats, ‘asdïts nyesdïts

niwhiy tsewhiy nuwhiy dewhiy diwhiy yiwhiy nisiy, nisewhiy nïwhiy ‘iwhiy, nawhiy nyests’iy

uphill downhill upstream downstream up down ahead behind side across

(Hagwilget Band 1995:â•›54–55) These terms also occur in nominal form (see Hagwilget Band 1995:â•›55).



Language, landscape and ethnoecology, reflections from northwestern Canada 315

Kaska directional stems áné nā´né tsā´ degé yegé dā´ nā´ nā´sdį’ dā´sdį’

off to side across down ahead, above, upstream down below, underneath, ?back along trail back, below, downstream beside across behind

Directional stems from Kaska 100 class taught by Josephine Aklak and Patrick Moore, 2002. Spellings as provided by Patrick Moore. Kaska directional with distance markers: ‘across’ nā´né across jahnā´né close by across jūhnā´né close by across kū´hnā´né across farther away (both people know place) ahnā´né across farther away (neither knows exactly) denā´né across farther away (speaker knows place, but addressee doesn’t know place) Directionals from Kaska 100 class taught by Josephine Aklak and Patrick Moore, 2002. Spellings as provided by Patrick Moore.

6.

Discussion

Although none of the corpora of geographic terms reported here are definitive and complete, some sense of the significance of different landscapes and relations to ways of making a living seems evident in the material here discussed. Terms for different kinds of snow, snowfields and glaciers, and terms for various kinds of rock faces and cliffs, avalanche tracks of different types, ridges, peaks, narrow places, and handholds on a rock face reflect the Gitksan need for an accurate vocabulary to communicate about how to travel through and on the mountains for activities like mountain goat hunting (requiring very detailed knowledge of topography and animal behavior, as well as superb climbing skills, before fire arms were introduced), berry picking, groundhog (marmot) hunting, and medicine picking. Slide areas are obvious features on the steep sloped mountains, with distinctive vegetation or none, and constitute both hazards to travel in the mountains and sometimes access routes, depending on season. They may also be utilized by or associated with animals, as grizzly bears on lush slide areas in the summer, or important food plants such as fern rhizome in the run-out zones. For Gitksan, en€hl’oo ‘slide’ appears to be a class of place kinds, including snow or rock slides

316 Leslie Main Johnson

of varying frequency, and slides which are snow or debris. The reality of assessing travel hazard in mountainous areas with high winter snowfall means that slope stability and the recognition of slide areas (both snow and rock) characterizes all three languages. For Gitksan, the finely detailed vocabulary for rivers and creeks and water stages of the Skeena, a large and swift river with great seasonal variation in flow and somewhat unpredictable winter freeze-up, reflects the need to travel on or over the river in the open water season, the need to be able to set fish traps or nets and to know where to fish for the five species of Pacific salmon and steelhead trout, and the need to travel on or along the river in the winter frozen season. The term for ‘large river’ (k’aliaks, focally the Skeena) is different from the term for other rivers or creeks (xsi- or xsu-) and contains the generic term for ‘water’ aks. In Athapaskan languages, there is some tendency to equate the word for ‘water’ with the word for ‘stream’ (see Kari 1996.) Eddies are also recognized in all three languages and are significant for both fisheries and navigation. Athapaskan languages tend to have very well developed terminology for many kinds and textures of ice, which are needful to recognize to be able to travel safely on rivers and lakes in the winter season. Gitxsanimx also names a number of different kinds of ice and snow which have implications for travel ease, safety, and snow stability. Vegetation, in contrast, is rather undifferentiated for all three groups, though most of the conspicuous plants of their environments are named at one level or another, and people know where and when to find plants of interest. Forest, as the relatively unmarked condition, is called in Gitksan and Witsuwit’en by terms which literally mean ‘among the trees.’ In Kaska one can refer to areas of tall trees (dechen chō, łedu), or tall trees at the mouth of a creek (tache). For Gitksan one can indicate specific types of forest by appending the name of the species, as in spaayt sginist [among the trees-pine Pinus contorta] or spagayt am’mel [among the trees-cottonwood Populus balsamifera, ssp. trichocarpa]. For Witsuwit’en and Kaska, one can indicate species by indicating the tree kind, usually along with ‘big,’ and the term that means ‘among,’ as in tso co tah [among big spruce, Witsutwit’en, idiomatically ‘big spruce country’] or gat chō tah [among big pine, Kaska]. Kaska also contains terms for ‘willows’ (a broad and significant cover type in Dene ethnoecology – see Johnson 2008), formed similarly to the terms for forest as gū´le’ chō tah ‘among big willows.’ Some of the terms recorded in Table 5 are more complex than this and give a sense of the relationship, movement and flow that are characteristic of Northern Dene ethnoecology. I found it quite difficult to decide what were terms with stability, and what were spontaneous descriptions in some cases.



Language, landscape and ethnoecology, reflections from northwestern Canada 317

For Gitksan, if the forest is on a mountain slope (the typical condition for the slopes of mountains in Gitksan country) one can simply indicate lax sg’anist ‘on mountain.’ For Kaska and Witsuwit’en, the comparable terms (hés kage and dzilh k’it) are more likely to indicate being in the alpine, or a mountain summit. There are specific terms for timberline, well marked on mountains in all three areas. The Gitksan term gaskslax sga’nist contains the root for ‘mountain.’ One of the Witsuwit’en terms includes the root for ‘edge’ begh and may apply to edges of alder other than at altitudinal timberline. I also recorded the term for a mountain pass in Witsuwit’en weggiz [pass], dzilh iggiz [mountain pass], and terms which differentiated rock and grassy topped mountains in Kaska (tsé’ and hés; see Table 1). Several different terms are present for open areas (lax ‘aamit, lax amaaxws, Gitksan; wizilh k’it, Witsuwit’en; gunejęje, wanitsįndle, Kaska); treelessness is a major defining characteristic, and characteristics of winter snow cover in these open areas are also implied by the terms in Gitksan. Open areas in valley bottom and alpine may be called by the same terms in both Gitksan and Witsuwit’en; the absence of dense woody vegetation seems to be the defining attribute. For Athapaskan languages in general, ‘grass on [area]’ seems to be the most frequent term for grass-dominated open areas (tl’o k’it in Witsuwit’en). The Kaska dictionary (Kaska Tribal Council 1997) gives tl’oge [grass] as the term for meadow. Wetland types, which to some degree combine physiography and hydrography with characteristic vegetation, are variably developed in the homelands of the Gitksan, Witsuwit’en and Kaska. In the mountainous and steep homeland of the Gitksan, wetlands are not especially prominent over much of the area. I only recorded one word for wetland or ‘swamp’ in Gitxsanimx: laalax’u. This applied equally to moss-dominated peatlands and to fens and areas dominated by sedges. There was in addition a term translated as ‘muddy place’ or ‘mineral lick,’ a term which pertains to hunting (see Table 6). Several Witsuwit’en terms were recorded for ‘swamp’; their exact range of reference was not entirely clear. Linguist Sharon Hargus translated one term as ‘moose’s watering hole,’ which Elder Dan Michell explained is ‘salt lick,’ that is, a mineral lick used by moose, comparable to the Gitksan term. The meanings of others were described as the habitat for diaper moss or for bog cranberries. In Kaska country wetlands are well developed, prominent on the landscape, and of substantial ethnoecogical significance because of their association with moose habitat, waterfowl habitat, and beavers, as well as having implications for traveling on the land. Terms like tūtsel seem to belong with water terms as much . A set of Witsuwit’en landscape terms contains the root for ‘edge,’ thus tabegh is translated as ‘shoreline,’ binbegh as ‘lake shore’ (literally ‘lake its edge’), and the abrupt edge of an alder thicket as wizëbegh ‘edge of mountain alder,’ one of the terms also used for ‘timberline.’

318 Leslie Main Johnson

as with vegetation, and were described as “swamp, where you get your food.” The fit between the English terms ‘swamp’ and ‘slough’ and the relevant Kaska terms is not tidy (see Johnson 2010b), and tūtsel can be used either depending on vegetation, habitat characteristics, and flow of the slough (old cut-off river channel or side channel). Tūtsel can comprise both sedge-dominated and moss-dominated wetlands, with or without trees, though one can apparently also refer to mossdominated sites as ‘moss’ tsatl, which has distinct ecological associations and travel implications, as well as offering a contrasting suite of botanical resources. Berries are a significant resource for all three groups, but the natural and human ecology of berries differs between the Skeena-Bulkley drainage and the Liard River basin in the Yukon. For Gitksan and Witsuwit’en, who managed black huckleberry Vaccinium membranaceum and lowbush blueberry V. caespitosum patches by burning (Johnson 1999; Trusler & Johnson 2008), and for whom berry patches were important owned properties, there are terms for ‘berry patch,’ and, in Gitxsanimx, for ‘too much brush on the berry patch.’ In contrast, there do not seem to be specific Kaska terms for ‘berry patch,’ and indeed the occurrence of fruiting is much more stochastic and shifts on the landscape from year to year. Neither does there seem to be active management of berries by burning. Instead the effects of wildfire are recognized, but firing does not seem to be practiced to enhance fruiting, perhaps because of the ecology of the berry species involved. In Gitksan, logs (as on an old burn) are named suuks; this term occurs in avalanche tracks with down logs, lax ensuuks or luulaxsuuks, and also to refer to floating logs on floodwaters. There is also a term for standing burnt snags. A series of terms also indicate productive sites on the landscape for hunting, trapping and fishing, the access trails, and the associated camps, and hunting-related entailments were implicit in terms for a range of landscape features. Various Witsuwit’en terms for places on mountains which are nominally vegetation types revealed interesting ecological associations: as bedding areas for mountain goats (Oreamnos americanus) or as green areas suitable for hunting hoary marmots . Aboriginal burning was suppressed in the 1930s and 1940s, and has not been reinstated (Johnson 1999). Most berry patches now occur in clear-cut areas, but these are less productive for the most part and more ephemeral than berry patches which are deliberately maintained by fire. Some berry patches also occur in natural burns. . “Berry patch” an simaa’y is an important ecotope, and must be maintained by firing in Gitksan territory. The focal species of interest is black huckleberry, though lowbush blueberry was also a focus of management in the past. Large quantities of berries were collected for winter and for the feasthall in the past, and berries retain high cultural importance today. Accordingly, there are several terms indicating burned-over area or clearing, burn, berry grounds, too much brush on the berry patch, and regrowing trees. There is also a term for a berry patch trail, genimsimaa’y.



Language, landscape and ethnoecology, reflections from northwestern Canada 319

(Marmota caligata). Witsuwit’en male Elders provided extensive commentary on the behavior of animals in relation to specific sites or vegetation types and speculated about the implications of descriptive terms, especially in the context of the grassy slopes and meadows in the Nadina (Nedïn’a) country and the lake called Poplar Lake in English. The Witsuwit’en toponym alludes to the behavior of caribou (now extirpated from the area) crossing at a narrow place (Dan Michell, interview transcript, 2005). When asking the late Pat Namox for a term for open, south-facing slopes prominent in the Nadina River and Upper Bulkley areas, he spoke at length about the ability to see far from such places and their use in hunting. He described them as coënk’it and explained that the term contained the word for ‘eye’ (interview notes, May 28, 1998). The Kaska hunting-related place kinds seem similar in concept to those recorded for Witsuwit’en, but the emphasis on ‘licks’ in hunting seems greater for Kaska, and the explanation of the association of a (blazed) foot trail with a hunting camp and a tūtsel area as part of the ‘lookout’ complex was much more explicit. Gitksan also contains terms which indicate hunting trail and hunting camp. Trails and camp areas are part of the network of properties on hunting territories. Trails in general in Gitksan can be designated by purpose (berry patch trail, goat trail, hunting trail) as well as by the species that uses it; this latter is also quite evident in Witsuwit’en and in Kaska (as well as English), where animal trails can be specified by species or a (human) foot trail may be indicated. Gitksan trails were often explicitly constructed or improved, necessary to be able to effectively traverse the rugged mountain terrain, access productive alpine areas above steep glacially scoured valley walls, and to cross turbulent and swift rivers via constructed bridges (e.g., interview notes with Gitluudahl, the late Peter Muldoe, 1994). Dwellings, villages, and camps are another important set of landscape terms. The Gitksan, who had permanent winter village sites, have a term for village laxgaltsap, which contrasts with ‘out on the land’ or ‘in the bush’ galdo’o. Places of more temporary habitation during the seasonal round are distinguished as ‘camp’ or jokxw, and these are differentiated by predominant resource activity, as in ‘berry camp,’ ‘fish camp,’ ‘mountain goat hunting camp.’ Such camps are part of the network of owned resource sites on House territories. In contrast, Kaska describe places of habitation as ko˛, and in the past, seasonally occupied camps as well as more temporary camps while traveling were the places of human habitation on the landscape. The sense of power in the land and of specific places of spiritual power or encounter is common to all three cultures. The sense of the sacred, or of powerful places, was present talking about and traveling on the land with my Kaska teachers. Stories that detailed the power or mythological history of certain named landscape features were elicited as we traveled near such areas, or by

320 Leslie Main Johnson

association between one powerful place where certain events were localized and another where a similar history had occurred. As for other Dene (e.g. Legat et al. 2001), the land itself is powerful or sacred, and things that are close to or dwell in the land require respect. In common with other Dene peoples, Kaska and Witsuwit’en recognize certain charged places as the sites where mythic events occurred. One Witsuwit’en site is a rocky ridge across a flat whose name means ‘stone beaver dam.’ Other unusual rock formations or colored stains on rocks may commemorate other events from oral narratives or from long ago. One unusual place kind apparent in Witsuwit’en narratives is what one might term ‘calendrical sites.’ Because the traditional Witsuwit’en seasonal round required very extensive travel from winter traplines to distant summer fishing sites, knowing when to begin traveling to arrive at the canyon in Moricetown in time to fish the chinook or spring salmon run was quite important. I was told about a specific site known in English as ‘the footprints’ on the Sam Goozley territory near the present town of Houston: one stood in this site and examined the location of sunrise on the distant ridge. When the sun rose behind a particular peak on the ridgeline, it was time to set out for Kyah Wiget (Moricetown) for summer fishing (Alfred Joseph, personal communication, 1991; Alfred Mitchel, April 19, 1988). The rising of the moon also indicated timing of deploying a fish trap. Gisday We (Alfred Joseph) indicated there were other such sites on other territories which allowed calibration of the annual seasonal travel cycle. The Witsuwit’en landscape and the Gitksan landscape share a common border at Hazelton and share many features. Accordingly, it is instructive to consider similarities and differences in the landscape corpora I have recorded. In comparison to Gitksan, Witsuwit’en shows more elaboration of terms which indicate direction or relationship in their semantic structure (characteristic of Athapaskan languages in general; see Kari, this volume, and Moore 2002). There also appears to be a greater elaboration of wetland terms, and possibly less elaboration of terms for various kinds of rock faces. People spontaneously commented on directions of flow of water – water flowing into a lake, water flowing out of a lake. In common with other Athapaskan languages like Kaska, Witsuwit’en terms which describe vegetation types seem to place the person in the vegetation type for taller woody vegetation, or to describe vegetation as ‘place of X’ (literally ‘X on it,’ as in tl’o k’it, ‘grass on it’). Various terms for slopes and mountains and other features contain the term k’it ‘on,’ thus dzilh k’it ‘on mountain,’ more idiomatically translated as ‘summit.’ A berry patch is nit’ay k’it ‘on berries’ or perhaps ‘berries on it.’ A barren area . This seems to be a very widespread pattern of designating vegetation types; see Martin (1995:â•›146) for Mixe examples from southern Mexico and discussions in Johnson (2008).



Language, landscape and ethnoecology, reflections from northwestern Canada 321

with little vegetation is wizulh k’it, that is, ‘nothing on it.’ Tlo k’it ‘grass on it’ is a meadow area. Again, these are similar to Kaska terms for similar features, though I have previously remarked the absence of a specific Kaska term for berry patch, which I believe reflects differences in ecology between the two regions. Considering similarities and differences between Witsuwit’en and Kaska, a number of aspects of their landscape terminology are shared: vegetation terms have similar structure; the locational word system is deeply embedded in Athapaskan languages and is thus shared; and a range of words are in fact cognate and seem to have similar meanings, in particular the terms for ice features. Similar in meaning is the ‘lookout’ complex, though the Witsuwit’en term explicitly indexes looking, while this is implicit in Kaska through the choice of ‘lookout’ as the translation given by my Elder teacher. As both groups are quintessential hunters especially of caribou and moose, this shared concept is evidently linked to shared ecological aspects of a major focus of traditional livelihood. I had the sense that wetlands and licks played a larger role in contemporary Kaska landscape understanding than for Witsuwit’en, which could reflect the geographic differences between the Bulkley Valley, where most of my Witsuwit’en work was centered, and the Upper Liard River/Francis River area, where my Kaska research was centered. It might also be related to the greater contemporary importance of hunting for Kaska.

7.

Conclusions

Landscape ethnoecology in northwestern Canada deals with people in landscape and people moving through landscape, a theme remarked upon by other scholars as well (see Collignon 2005; Thornton 2008; Davidson-Hunt & Berkes 2010; Johnson 2010a and references therein). The trail, or path as Thornton (2008) phrases it, links places along paths of human movement through the seasons. These places may be familiar named places, whose affordances are known. The characteristics and potentialities of more generic types of sites are also known, and ecotopes or place kinds can be conceived of as having entailments, that is, of having predictable characteristics with consequences for travel, presence of resources and ease of obtaining them, and so on. Landscapes have biophysical traits and features, and they also have human features, the ‘human geography overlay’ as I have termed it. There is also a layer of meaning which pertains to the sacred (and by implication, perhaps the polluted or profane) – that is, the “Land” also has power, and emplaces power, in ethnoecological understandings of landscape. This too is needful to know for those who interact with the land: where particular respect is necessary, places to avoid, places to garner power. For Dene and other people, certain sites congeal or instantiate stories which teach lessons

322 Leslie Main Johnson

one should bear in mind (e.g. Andrews 1990; Andrews & Zoe 1997; Andrews et al. 1998 for Northern Dene examples; and see Thornton 2008 for an in-depth discussion for Tlingit in Alaska; Basso 1996 for Western Apache; and Strang 1997 and Rose 2000, among others, for Australian examples). As Cruikshank and others (e.g. Cruikshank 1990:â•›56) have remarked, history is instantiated on the land through such sites as well as more personal recollections. The influence of human geography, of society, its structure, and its inscribed activity patterns shapes local landscape perception, drawing attention to some sites, features or relationships while downplaying others. The suite of features that the landscape itself offers will also shape local landscape perceptions, emphasizing features of importance in the local landscape, and failing to attend to others not locally prominent or significant. The elaboration of terminology relating to slides and avalanches and to cliffs and rock faces among the Gitksan is an example of this presented in this paper, as is the rich terminology describing fine-scale hydrographic phenomena, important for navigation and successful fishing on the large rivers of the Gitksan homeland. For the Kaska the primacy of hunting is clear in conversations about the land, as is a pervasive sacredness which demands appropriate respect. The Kaska are skilled travelers who move widely to take advantage of the resources offered by diverse places through the seasons. This was not a mere vagrant wandering; people had family hunting and trapping areas, and the seasonal pattern of movement had a logic and relative stability, though remaining dynamic in response to the change inherent in the boreal forest as well as in response to other peoples (indigenous and non-indigenous trading partners, for example). For Witsuwit’en, there is a rich sense of the dynamic landscape, where mountains memorialize events of long ago, are seen in anatomic analogy, and are significant as landmarks for people with a large traditional seasonal travel pattern. Places may be boundary markers, but names also describe their characteristics. A dynamic sense of river systems and relationships emerges. Description of travel in the mountains and avoiding its hazards and of the significance of snowfields and glaciers (now diminishing) are other themes that stand out. Landscape is an active partner in the human business of making a living, of dwelling in place. Landscape knowledge is thus of fundamental importance in people’s ecological relationships and ecological knowledge. The characteristics of language, culture, local economy and the features of the biophysical world all contribute to local landscape terminology, toponyms and discourse about landscape, which in turn encodes knowledge about places and directs attention to significant features of the landscape.



Language, landscape and ethnoecology, reflections from northwestern Canada 323

Sources of geographic terms in tables Gitksan: Art Mathews, Dinim Gyet; Peter Muldoe, Gitluudahl; Beverley Anderson; Tommy Tait; Sara Tait, Wihalite; David Green; Kathy Holland; Commission Evidence/Court Case information; Bruce Rigsby; Sadie Howard; Gitksan Interpreters’ Gitxsan Glossary; Gitxsan Dictionary; Edgar Good; Mary Johnson, Antgulilbisxw. Witsuwit’en: Pat Namox; Lucy Namox; Alfred Joseph; Sharon Hargus; More Stories of Moricetown Carrier 1979; Tsë cakh Wit’en, traditional words and texts from Band members, Hagwilget (Tse-Kya) People, 1995; Maryanne Austin; Alfred Mitchell; Charlie Austin; Margaret Austin; Josephine Michell; Sara Tait; Beatrice Morris; Lillian Lewis. Directionals: Tsë cakh Wit’en, traditional words and texts from Band members. Hagwilget (Tse-Kya) People, 1995:â•›53–55. Kaska: General geographic terms: Terms listed compiled from speakers from several dialect areas and also from Kaska dictionary (Kaska Tribal Council 1997); orthography provisional. Some terms might be considered more descriptions than names; this is a difficult distinction to draw in languages that feature relational terminology and which are synthetic and verb-oriented. The snow and ice terms are all from the dictionary, as my field research was not in winter season. I included Liard or Frances Lake words where possible, as these were the main dialects I worked with. Orientation and directionals: Directionals from Kaska 100 class taught by Josephine Aklak and Patrick Moore, 2002. Spellings as provided by Patrick Moore.

324 Leslie Main Johnson

References Andrews, Thomas D. 1990. Yamoria’s arrows: Stories, place-names and the land in Dene oral tradition. Unpublished report for National Historic Parks and Sites, Northern Initiatives, Canadian Parks Service, Yellowknife NWT. Andrews, Thomas D., MacKay, Glen & Andrew, Leon. 2009. Hunters of the Alpine Ice: The NWT Ice Patch Study. Blurb Publishing, Andrews, Thomas D. & Zoe, John B. 1997. The Idąą trail: archaeology and the Dogrib cultural landscape, Northwest Territories, Canada. In At a Crossroads: Archaeology and First Peoples in Canada, George P. Nicholas & Thomas D. Andrews (eds), 160–177. Burnaby BC: Archaeology Press, Simon Fraser University. Andrews, Thomas D., Zoe, John B. & Herter, Aaron. 1998. On Yamòzah’s trail: sacred sites and the anthropology of travel. In Sacred Lands: Claims, Conflicts and Resolutions [Occasional Publication No. 43], Jill Oaks, Rick Riewe, Kathi Kinew & Elaine Maloney (eds), 305–320. Edmonton AB: Canadian Circumpolar Institute. Basso, Keith. 1996. Wisdom Sits in Places: Landscape and Language Among the Western Apache. Albuquerque NM: University of New Mexico Press. British Columbia Forest Service. 1988. [biogeoclimatic zone map] Collier, Malcolm. 2001. Approaches to analysis in visual anthropology. In Handbook of Visual Analysis, Theo van Leeuwen & Carey Jewitt (eds), 35–60. London: Sage. Collignon, Beatrice. 2005. Knowing Places, the Inuinnait, Landscapes and the Environment [Circumpolar Research Series No. 10]. Edmonton AB: Canadian Circumpolar Institute (CCI) Press. Cruikshank, Julie. 1990. Getting the words right: perspectives on naming and places in Athapaskan oral history. Arctic Anthropology 27(1): 52–65. Daly, Richard. 2005. Our Box was Full: An Ethnography for the Delgamuukw Plaintiffs. Vancouver BC: UBC Press. Davidson-Hunt, Iain & Berkes, Fikret. 2003. Learning as you journey: Anishinaabe perception of social-ecological environments and adaptive learning. Conservation Ecology 8(1). Gibson, James J. 1979. The Ecological Approach to Visual Perception. Boston MA: Houghton Mifflin. Hagwilget (Tse-Kya) People. 1995. Tsë Cakh Wit’en, Traditional Words and Texts from Band Members. Hagwilget BC: Hagwilget Band. Hare, P. Gregory, Greer, Sheila, Gotthardt, Ruth, Farnell, Richard, Bowyer, Vandy, Schweger, Charles & Strand, Diane. 2004. Ethnographic and archaeological investigations of alpine ice patches in southwest Yukon, Canada. Arctic 57(3): 260–272. Hargus, Sharon. 2007. Witsuwit’en Grammar: Phonetics, Phonology, Morphology. Vancouver BC: UBC Press. Johnson, Leslie Main. 1997. Health, Wholeness, and the Land: Gitksan Traditional Plant Use and Healing. PhD Dissertation, University of Alberta. Johnson, Leslie Main. 1999. Aboriginal burning for vegetation management in northwest British Columbia. In Indians, Fire and the Land in the Pacific Northwest, Robert Boyd (ed.), 238–254. Corvallis OR: Oregon State University Press. Johnson, Leslie Main. 2000. “A place that’s good”, Gitksan landscape perception and ethnoecology. Human Ecology 28(2): 301–325.



Language, landscape and ethnoecology, reflections from northwestern Canada 325

Johnson, Leslie Main. 2008. Plants and habitats – a consideration of Dene ethnoecology in northwestern Canada. Botany 86: 146–157. Johnson, Leslie Main. 2010a. Trail of Story, Traveller’s Path: Reflections on Ethnoecology and Landscape. Edmonton AB: Athabasca University Press. Johnson, Leslie Main. 2010b. Visions of the land: Kaska ethnoecology, “kinds of place” and “cultural landscape”. In Landscape Ethnoecology, Concepts of Biotic and Physical Space, Leslie Main Johnson & Eugene S. Hunn (eds), 203–221. New York NY: Berghahn Books. Johnson, Leslie Main & Hargus, Sharon. 2007. Witsuwit’en words for the land – a preliminary examination of Witsuwit’en ethnogeography. In Alaska Native Language Center Working Papers in Athabaskan Linguistics Number 6, November 6, 2006, Siri Tuttle, Leslie Saxon, Suzanne Gessner & Andrea Berez (eds), 147–157. Fairbanks AK: Alaska Native Language Center. Johnson, Leslie Main & Hunn, Eugene S. (eds). 2010. Landscape Ethnoecology, Concepts of Biotic and Physical Space. New York NY: Berghahn Books. Kari, James. 1989a. Affix positions and zones in the Athapaskan verb complex: Ahtna and Navajo. International Journal of American Linguistics 55(4): 424–454. Kari, James. 1989b. Some principles of Alaskan Athabaskan toponymic knowledge. In General and Amerindian Ethnolinguistics, in Remembrance of Stanley Newman, Mary Ritchie Key & Henry M. Hoenigswald (eds), 129–149. Berlin: Mouton de Gruyter. Kari, James. 1996. A preliminary view of hydronymic districts in Northern Athabaskan prehistory. Names, a Journal of Onomastics 44(4): 253–271. Kuzyk, Gerald W., Russell, Donald E., Farnell, Richard S., Gotthardt, Ruth M., Hare, P. Gregory & Blake, Erik. 1998. In pursuit of prehistoric caribou on Thandlät, southern Yukon. Arctic 52(2): 214–219. Legat, Allice, Chocolate, Georgina, Chocolate, Madelaine, Williah, Pauline & Zoe, Sally Ann. 2001. Habitat of Dogrib Traditional Territory: Placenames as Indicators of Biogeographical Knowledge – Final Report. Submitted by Whàehdòo˛ Nàowo Ko`˛ Dogrib Treaty 11 Council to West Kitikmeot Slave Study Society, Yellowknife, NT, March 2001. Kaska Tribal Council. 1997. Guzāgi K’ū´gé’, Our Language Book: Nouns. Kaska, Mountain Slavey and Sekani. Watson Lake YT: Kaska Tribal Council. Mark, David M. & Turk, Andrew G. 2003. Landscape categories in Yindjibarndi: ontology, environment, and language. In Spatial Information Theory: Foundations of Geographic Information Science [Lecture Notes in Computer Science 2825], Werner Kuhn, Michael Worboys & Sabine Timpf (eds), 28–45. Berlin: Springer. Mark, David M., Turk, Andrew G. & Stea, David. 2010. Ethnophysiography of arid lands: Categories for landscape features. In Landscape Ethnoecology, Concepts of Biotic Physical and Space, Leslie Main Johnson & Eugene S. Hunn (eds), 27–45. New York NY: Berghahn Books. Martin, Gary J. 1995. Ethnobotany: A Methods Manual. London: Chapman and Hall. Mills, Antonia. 1994. Eagle Down is Our Law: Witsuwit’en Law, Feasts and Land Claims. Vancouver BC: UBC Press. Mills, Antonia (ed.). 2005. “Hang onto These Words”: Johnny David’s Delgamuukw Evidence. Toronto: University of Toronto Press. Moore, Patrick James. 2002. Point of View in Kaska Historical Narratives. PhD dissertation, Indiana University.

326 Leslie Main Johnson

Nabhan, Gary P. 2001. Cultural perceptions of ecological interactions: An “endangered people’s” contribution to the conservation of biological and linguistic diversity. In On Biocultural Diversity: Linking Language, Knowledge and the Environment, Luisa Maffi (ed.), 145–156. Washington DC: Smithsonian Institution Press. Nazarea, Virginia. 1999. Ethnoecology, Situated Knowledge, Located Lives. Tucson AZ: University of Arizona Press. Rigsby, Bruce. 1986. Gitksan Grammar. Ms, University of Queensland, Australia. Rigsby, Bruce & Kari, James. 1987. Gitksan and Wet’suwet’en Linguistic Relations. Report prepared for the Gitksan-Wet’suwet’en Tribal Council. On file, library, Office of the Hereditary Chiefs, Hazelton BC. Rose, Deborah Bird. 2000. Dingo Makes us Human: Life and Land in an Australian Aboriginal Culture. Cambridge: CUP. Strang, Veronica. 1997. Uncommon Ground: Cultural Landscapes and Environmental Values. Oxford: Berg. Thornton, Thomas F. 2008. Being and Place among the Tlingit. Seattle WA: University of Washington Press. Toledo, Victor. 1991. What is ethnoecology? Origins, scope and implications of a rising discipline. Étnoecologica 1(1): 5–21. Toledo, Victor. 2002. Ethnoecology: a conceptual framework for the study of indigenous knowledge of nature. In Ethnobiology and Biocultural Diversity, John R. Stepp, Felice S. Wyndham & Rebecca K. Zarger (eds), 511–522. Athens GA: International Society of Ethnobiology. Trusler, Scott & Johnson, Leslie Main. 2008. “Berry Patch” as a kind of place – the ethnoecology of black huckleberry in northwestern Canada. Human Ecology 36(4): 553–568.

chapter 15

Landscape embedded in language The Navajo of Canyon de Chelly, Arizona,  and their named places Stephen C. Jett

On the pictorial wings of placenames, imaginations soar. (Keith Basso 1990:â•›173) Humans interact with landscape by classifying and labeling a select multitude of the landscape’s limitless individual areas and features. Studying place names reveals much about language, perception, values, beliefs, environment, economy, and history. Like place-naming among other Athabaskan speakers, Navajo toponymic practice overwhelmingly produces descriptive names for landscape features, reserving commemorative and activity place-naming largely for human-modified places. Athabaskan languages employ an unusual number of topological and directional affixes and verb forms, which can condense and convey much information within brief descriptive place names. Lexeme frequencies hint at what Navajos see or saw as significant in their natural landscape. Place-naming facilitated possessing/controlling landscapes. The Navajo attached the itineraries and activities of mythological protagonists to the land via a web of place names; the associated stories and their descriptive names served as mnemonic guides for far-traveling Navajos.

1.

Introduction

Human beings live in and experience landscapes, and they interpret and alter those landscapes through cognition and through action, as well as through language expressing that cognition and action (for an overview, see Levinson 2003). People identify and specify places that they see as significant by assigning those places names that characterize and/or symbolize them. On first consideration, place names – especially merely descriptive ones (like Flat Rock, for instance) – may strike one as of minor and only particularist consequence. But in fact, place names are potentially highly informative regarding a

328 Stephen C. Jett

considerable number of significant things. This is because they are intersections of place, landscape, thought, language, perception, value, belief, history, economy, and society, and thus provide avenues of understanding toward all of these physical-environmental, cognitive, linguistic, and cultural phenomena (see Thornton 1997:â•›209–210). Place names reflect a people’s fundamental relationship to, and interaction with, its habitat, as well as that people’s often encyclopedic knowledge of its territory and beyond (see Afable & Beeler 1996:â•›185). The anthropologist of Northern Athabaskan Indians Richard K. Nelson (1994:â•›18), in “The Embrace of Names,” wrote of “[n]ames covering the terrain like an unbroken forest. Names that wove people profoundly into the landscape, and that infused the landscape profoundly into the people who were its inhabitants. Names that give special life to the terrain …”. As folklorist Susan W. Fair (1997:â•›467), writing of the Inupiat, put it, “Native teachings electrify each named place with an intimate conglomeration of activities, genealogy, history, memory, moral lessons, and future …”. And, averred the anthropologist of the Apache Keith Basso (1990:â•›144), Place names are arguably among the most highly charged and richly evocative of all linguistic symbols. Because of their inseparable connection to specific localities, place names may be used to summon forth an enormous range of mental and emotional associations – associations of time and space, of history and events, of persons and social activities, of oneself and stages in one’s life. And in their capacity to evoke, in their compact power to muster and consolidate so much of what a landscape may be taken to represent in both personal and cultural terms, place names acquire a functional value that easily matches their utility as instruments of reference.

The Athabaskan language group is, with Eyak, a subdivision of the NaDene linguistic family (which includes Tlingit), centered in northwestern North America (Krauss 1973), and is now fairly certainly linked with Yeniseian of southwestern Siberia (Ruhlen 1998; Vajda 2010, in Kari & Potter 2010). Ancestral Na-Deneans probably entered Alaska from Siberia before 6000 years ago and expanded into western Canada (Kari, this volume). The Athabaskan-speaking ancestors of the Navajos and other Apaches moved into the American Southwest from Canada before A.D. 1500 (Perry 1991; Haskell 1987; Ives 2003; on the Navajo [Diné] language, see Young & Morgan 1980, 1992; Yazzie & Speas 2007; on Navajo phonology, see McDonough 2003; on Western Apache, see Bray 1998). Northeastern Arizona’s Canyon de Chelly/Canyon del Muerto system is at the heart of today’s Navajo Country. In 2001, I published the book Navajo Placenames and Trails of the Canyon de Chelly System, Arizona, which inventoried and analyzed over 250 Navajo place names in and adjacent to the canyons. The bulk of the data used in this paper derives from that work (see also Jett 1970;



Landscape embedded in language 329

Wilson & Dennison 1995; Van Valkenburgh 1974, 1999; Brugge 1993:â•›19–20, 40–54; Linford 2000; Aton & McPherson 2000:â•›36–37).

2.

Place names as perception and description

Most Native American place names – including those of the Navajo and other Athabaskan speakers and the Na-Denean Tlingit – are descriptive. Except when referring to human-modified places, Navajo place names are seldom activity toponyms or commemorative in the strict sense; they thus closely and directly reflect landscape perception (e.g., Afable & Beeler 1996:â•›185; Bright 2006; Kari 1996:â•›445; 1989, this volume; Thornton 2008:â•›68; Basso 1996:â•›23; Baumhoff 1958). Although there is surprisingly modest repetition of true place names over the vast Navajo Country, some geographical terms are sufficiently common as to be generics. Navajo examples of generics include tsé ‘rock’ (as a feature) and tó ‘water, spring’ and even the two combined as tséya tó ‘water under rock’, i.e., ‘cave spring’. Another, more complex example is be’ek’id halchii’ ‘red-area lake’, a term for a chronically muddy pond or lake (Brugge 1993:â•›40). Geographic terms may be particularized by addition of the -í suffix. Individuated but common designations such as Tsé ’Ã“í’áhí ‘The Standing Rock’, of which there are at least five places so named in the canyon system and many more in the greater Navajo Country, may be thought of as semigenerics. (For additional generics and semigenerics, see Jett 2001:â•›196–197.) As with other Athabaskan languages (cf. Kari 1989:â•›129, 139, this volume; Basso 1984, 1990:â•›155; Goodwin 1932; Basehart 1974:â•›63–82; Baumhoff 1958), almost all Navajo place names are transparent; although a small number contain archaisms, very few are wholly unetymological. Most commonly, Navajo descriptive place names include one (or more) specifying elements added to a generic to form a binomial (or polynomial) appellation. Such names are normally very straightforward and describe one (or more) salient physical characteristics of the place. An example is Shąą’tóhí ‘The Sunnyside Spring’ in Canyon de Chelly, whose aspect is toward the midday sun. Still, in a few cases a rock or other feature will be named for its resemblance to something else (e.g., Tsídii Binii’í ‘The Bird’s Face’, a rock with an avian shape in Canyon del Muerto). Place name collection and analysis throw light on what locations or features the namers perceived as being notable, worth naming. In the Canyon de Chelly system, almost half of the Navajo place names are attached to natural features (most of the cultural place names refer to trails, and a few to prehistoric vestiges). The ten classes of natural referents of which there are two or more examples in the canyons are, in order of frequency: rocks (especially, standing rocks); canyons

330 Stephen C. Jett

and coves; caves/overhangs/ledges; gaps in rock; springs; stream entries, exits, and confluences; plants (as individuals or patches); cliffs; falls and stream narrows; and natural cliff stains that resemble something else. (The local frequencies of these and other referents vary, of course, as the physical environment varies areally.) Although most Navajo place names are thus quite specific, not infrequently the name of a particular feature may be used to refer to the entire general area (cf.€Brugge 1993:â•›40; cf. the Tlingit: Thornton 2008:â•›87–93). In strong contrast to Anglo-American place-naming, only about ten percent of de Chelly-system Navajo place names refer to individuals or to specific human-historical or mythological occurrences or to repeated human activities. Furthermore – again, in contrast to Anglo practice – allusion to or transfer of a name from one location to another, distant one (e.g., Athens, Greece, inspiring Athens, Georgia, and Athens, Ohio) is very rare among the Navajo, although transfers to nearby, comparable locations occasionally occur.

3.

Place names for place-possession

Because the land is a sacred creation (see Levy 1998) and of indefinite duration, whereas humans are merely land-users with limited powers and finite lifetimes, traditionally Navajos cannot individually possess parcels of Navajo Country and their natural landscape features, only successively utilize them for dwelling, for grazing and farming, and for wild-resource-collecting under a system of use rights (which expire if use is abandoned for a significant period of time; Haile 1968). Numbers of place names do include the third-person possessive prefix bi- ‘his/her/its.’ Nevertheless, with respect to natural features or places or even man-made trails, this simply implies customary use or, alternatively in the case of a trail, the individual’s having constructed it, and does not signal exclusive, transferrable possession. An example in the Canyon del Muerto branch is Tó€Dích’íi’nii Bitséyaa ‘Bitterwater Clan’s Cave’, named because Bitterwater Clan land-users customarily dried produce there. Navajos generally avoid attaching personal names or sobriquets to natural features, as doing so might attract too much attention to that person from the supernaturals (Jett 1970:â•›182; for similar Northern Athabaskan avoidance, see Kari 1989:â•›129, 142; Cruikshank 1990:â•›62; for the Tlingit, see Thornton 1993:â•›298).

. Jádí Habitiin ‘Antelope’s Trail up out’ in Canyon del Muerto refers not to pronghorns using the trail but to nearby Navajo pictographs of such animals or to hunters employing the trail en route to the hunting ground outside the Canyon.



Landscape embedded in language 331

On the other hand, traditional Navajo people perceive themselves as being corporately permanent and the Navajo Country as having been created exclusively for them, so one could say that Diné Bikéyah ‘Navajo’s Land’ (originally, Dinétah ‘Among the Navajo’, i.e., ‘Navajodom’) is possessed by the Dine’é ‘The People’ as a whole (see Jett 2002). It may also once have been the case that local bands were perceived as possessing their local territories; a number of clans€– whose memberships are now scattered – have place-derived names, suggesting that these clans originated as localized bands (Jett 1978:â•›357). Furthermore, some individuals were known by sobriquets derived from their home areas, reflecting the close association between person and customary-use area. One way by which a group signals its taking possession of the land is what the Norse who settled largely empty Iceland and then Greenland called Landnáma ‘land-naming’. In a sense, land-naming was land-taking and land-taming. Like Iceland and Greenland, much of what is now the Navajo Country may have been unoccupied when the Navajos’ ancestors arrived from the north something over half a millennium ago, the land having been abandoned by the Ancestral Puebloans (Anasazi) by A.D. 1300. The People did not adopt any pre-existing place names except, possibly, as calques, i.e., loans by translation, in a small number of cases (cf. the Alaskan Athabaskan Ahtna; Kari & Tuttle 2005:â•›5; Kari, this volume). A new web of place names and linked supernatural stories was applied to the land (the stories probably, in part, under Puebloan influence). Although

. E.g., Ts’ah Yisk’idnii ‘Sagebrush Hill People’, Nihoobáanii ‘Gray-streak Ends People’, Kin Yaa’áanii ‘Towering House People’, and Kin Łichíinii ‘Red House People’. Thanks to David M. Brugge for suggesting inclusion of this material. . On the other hand, it is instead quite possible that the ’Anaasází ‘Ancestral Aliens’ were driven out by the expanding proto-Navajo and proto-Apache (see Jett 1964). . Direct translation by Navajos was probably very rare. “The Navaho … evince little interest in any language but their own,” and in fact “had a deep-seated belief that contact with alien people or anything connected with them was potentially dangerous” (Hill 1948:â•›391). Athabaskan tongues are generally resistant to lexical borrowing (Callaghan & Gamble 1996:â•›111), and Northern Athabaskans seem not to have adopted any pre-existing placenames (Kari & Fall 2003:â•›40; Kari 1989:â•›129, 139, 144, this volume). However, bilingual representatives from neighboring peoples could have translated their placenames into Navajo for Navajo individuals with whom they interacted (as in trade, in marriage, or in servitude), occasionally leading to Navajo adoption of a foreign name through translation. To the extent that the Navajo incorporated Puebloan stories, certain placenames in those stories may have been adopted by translation. Although borrowing was/is rare, the Navajo word for ‘bison’, ’ayání (as in the modern placename ’Ayání Bito’ ‘Bison Spring’), likely reflects a rare pre-reservation-period borrowing since it appears to be a variant of widespread similar lexemes of the approximate form yanasa found, for example, in Muskogean languages, Tunica, Natchez, and Cherokee (see Haas 1978:â•›33).

332 Stephen C. Jett

local Navajo names of natural places continue occasionally to be coined, place names as a whole and the names of important individual places are seen, traditionally, as having been established in permanence by the Holy People long before the creation of the first Navajo clans. But these places and their names were established for the People and in the Navajo language, so their existence and their employment in the myths is, to traditional Navajos, prime evidence of the legitimacy of the Dine’é’s occupation of their territory.

4.

Place names as itineraries

James Kari (personal communication 2008) has noted, “Geographic knowledge and prowess are highly valued in all Athabaskan cultures”. “Athabaskan place names”, he stated, “are systematic multifunctional sign networks that are conducive to being memorized and that facilitate travel and occupancy over large areas” (Kari 1996:â•›443). He also spoke of “the reverence and the care speakers take when reporting sequences of names …” (p. 444). “The place names appear in networks,” he continued, “and constitute cognitive maps” (p. 464). The anthropologist of the Navajo Gary Witherspoon (1977:â•›34) noted that, for the Diné’é, “the essence of life is movement. [This] proposition … is firmly embedded in the structure and content of Navajo language …” (cf. Tlingit: Thornton 1993:â•›295). The scholar of Navajo mythology Karl Luckert (1979:â•›195) argued that, “[b]y way of these round trips [of protagonists in chantway myths,] the Athabaskan hunter mind laid claim to a southwestern homeland.” The Navajo see their country, bounded by the four directional sacred mountains, as foreordained for their occupation and use (see Jett 2006). One function of claiming the land by naming its various notable features (or, more accurately, by discovering the names assigned to places by the Holy People) was to create a ceremonial and pragmatic mental map of the territory so that the Dine’é might orient themselves within it and so that it could come to serve their human needs and desires. Navajos, traditionally a quite mobile people, are unusually preoccupied with the details of geography and of named places, and descriptive place names served as geographically and ceremonially mnemonic wayfinding devices (Francis & Kelley 2005; on Alaska Athabaskans, see, e.g., Kari 1989, this volume; Kari & Fall 2003; on the Tlingit, see Thornton 1997:â•›221, 2008:â•›70–80). Christopher Vecsey (1988:125) declared, “By knowing, naming, and acting willfully, Navajos can coerce even the gods to share their potency …”. Wrote the scholar of Navajo ceremonial literature Leland C. Wyman (1957:â•›36):



Landscape embedded in language 333

Place is of the utmost importance to the Navaho. The need is felt ritually to recapitulate mythical toponymy and topography in song and prayer. The geographical tales of long journeys of the protagonists of the myths almost literally bound the Navaho Country; at least they state its landmarks.

Agreed the anthropologist of the Navajo Richard F. Van Valkenburgh (1974:â•›17– 18): Their stories, whether religious or secular, are replete with accounts of natural features, locations, distances. To the Navajo, these details are important alike because of their sacred associations and because of their practical significance: to a people who did not write they served as the equivalent of our maps, and they guided the Navajo to sites for hunting and for gathering plant foods, ceremonial plants, minerals, and other objects used in ritual.

As Richard Nelson (1983:â•›39) wrote about the Navajo’s linguistic relatives in Alaska, “These names are like street signs on a mental map that hunters and travelers use to orient themselves. Many are linked to stories of the past or to beliefs about certain places; so they fill the landscape with cultural and personal meanings”. Among the Navajo, they facilitate(d) journeys for visiting, war, hunting, trade, religious pilgrimage, and so forth. The Navajo anthropologist Harris Francis and the Anglo anthropologist Klara Kelley examined this phenomenon among the Navajo and found that … Navajo ceremonial stories include verbal maps … the maps in the Navajo stories can be used for wayfinding. The Navajo verbal maps work by identifying routes of travel with sequences of named landmarks and cultural features. They show the direction of travel by anchoring those sequences to icons of the cardinal directions or other places in the cosmic framework. … The story sequence forms a guideline for travelers to know which places they need to reach, in what order, and in what direction. Not only can the stories function as maps, but in the past Navajos did use them as maps. …  (Francis & Kelley 2005:â•›98–99)

As mentioned above, Navajo, like other Apachean tongues, is an Athabaskan language, closely related to the speech of the Native peoples of western Canada’s boreal forest and forest edge (Hoijer 1956). Athabaskan languages are characterized by unusual spatial/topological specificity, and as a result place names contain sometimes surprising amounts of information in small packages (see Pinxten et al. 1983; Young & Morgan 1980, 1992; Henry & Henry 1969; Cruikshank 1990:â•›64; Kari, this volume). As I have written elsewhere, such “information is … communicated via verbal roots, many of which imply shape and position – as elongate and stiff, blob-like, and so forth, and as standing, sitting, lying horizontally,

334 Stephen C. Jett

spreading out, and so on (Hoijer 1945; Faltz 1998; Young 2000)” (Jett 2001:â•›186; see also Leer 1989; Smith 2000; cf. Tlingit: Thornton 2008:â•›80–85). There are also a number of affixes that can be added to simple noun stems and thereby indicate position or directionality. Postpositional suffixes of this sort include -yi’ ‘within’, -gi ‘at’, -k’i ‘on’, -yaa’ ‘beneath’, -ni ‘under’, â•‚gha ‘through’, -ta’ ‘between’, -tah ‘among’, -naa ‘around’, and -jí(go) ‘in the direction of’. Prefixes include bii- ‘inside’, ha- ‘upward’, na- ‘downward’, and ch’í- horizontally outward’ (see, inter alia, Kari 1976). And there are also bąąh ‘alongside’, yáá ‘up into the air’, naa ‘across’, dah ‘at an elevation’, and ’adah ‘downward from a height’. I give here two instances of these usages in Canyon place names. Tsé Dah Sitlé’é is Junction Rock, a lumpish sandstone butte sitting on a rock-cut terrace at the confluence of Canyons de Chelly and del Muerto. Tsé means ‘rock’ and dah means ‘at an elevation’. The si- that begins the verb sitłé’é implies that the subject is in a static position; the succeeding -tłé’é part of the verb specifies that the object is sitting like a mushy blob atop another object or on a substrate. A second example is Ch’ó Haazt’i’. Ch’ó is ‘douglasfir(s)’. The haaz- part of the verb indicates that the subjects extend up out (of the canyon), and the t’i implies that they do so in a line. A specific place name may be qualified in order to generate additional, associated place names. For example, the aforementioned Ch’ó Haazt’i’ ‘Douglasfirs Extend up out in a Line’ is an irregular row of trees that lends its name to nearby features (cf. Ahtna; Kari, this volume): Ch’ó Haazt’i’ Nástł’ah ‘DouglasfirsExtend-up-out-in-a-Line Cove’ (Fir Tree Canyon) and Ch’ó Haazt’i’ Ha’atiin ‘Douglasfirs-Extend-up-out-in-a-Line Trail’.

5.

Place names as history

As mentioned, except for many of those referring to settlements and other human-established features, Navajo place names are usually descriptive and are seldom specifically commemorative. Nevertheless, the stories (hané’) explaining the origins of the ceremonials typically involve long, arduous journeys of protagonists in search of curative knowledge and include long lists of the names of the places that those protagonists pass and at which the Holy People transfer ritual knowledge to them; in the great majority of cases, these are real and stillidentifiable places carrying Navajo names, which reinforces the Navajo perception that these myths are true and bind the People to this particular land (see Jett 1993, 2006; Watson 1964; Van Valkenburgh 1974; Kelley & Francis 1994; McPherson 1992; cf. the Tlingit: Thornton 2008:â•›111). With regard to de Chelly



Landscape embedded in language 335

(and as extendable to all of the Navajo Country), local Blessingway Singer Greyeyes Ben Stuart said, “And the [holy] people built the canyon and it became the Navajos’ … And there are different places that are named after the [stories of the ceremonial] sings …” (Harris 1965:â•›3–4). Thus, many place names are reflections of mythic history, in which the supernatural intersects the physical and the mundane in the form of particular named places – places that retain power owing to their roles in these mythic events (see Jett 1995); pilgrimages to these places and the invoking of their names in ceremonial song and prayer call upon that power. An example is the previously discussed Ch’ó Haazt’i’ ‘Douglasfirs Extend up out in a Line’, which refers to an actual line of trees in a tributary of Canyon del Muerto. According to the story of the Coyoteway ceremonial, at this location the leader of the Yellow Corn People leaped down from a cliff, and firs sprang up in the footprints that he made upon landing (Luckert 1979:â•›196–197). Some localities important in these stories have, in addition to their exoteric secular names, esoteric ceremonial ones. A case in point is Rock People Turned Into, near the mouth of de Chelly’s Monument Canyon tributary. This is a mass of columnar sandstone, which actually carries several names. A secular descriptive name is Tsé Dah Deeshzhaa’í ‘The Up-Spiked Rock’. Then, there is the ceremonial descriptive/commemorative name Tsé Ndadzisdlí’í ‘The Rock That People Turned Into’; a story tells that this landmark comprises humans who were petrified as punishment for disobeying local Holy People. Some sacred landmarks, such as the four cardinal-directional mountains, carry one or more ceremonial honorific titles as well as their exoteric names; and Francis and Kelley wrote of “prototype place names”, which are generalized esoteric names such as Dził Łáhdilt’éí ‘Mountain One’ that are applied to more than one geographic landmark (each of which has its individual exoteric label) and which, by their application, connect those features to general cosmography. The application of ‘Mountain One’ implies that the feature so named is an originating place, as on a supernatural being’s journey (Francis & Kelley 2005:â•›93). Regarding secular history, there is, of course, a history (or at least, a reason) for every place name. But whereas numbers of named natural features have historical or personal or activity-related stories associated with them, as is true in other Athabaskan-occupied regions, few of the de Chelly names themselves contain any reference to events. Still, there are a few instances. There is, for example, ’Adah ’Aho’doo’nilí ‘Where People Were Pushed Away Downward from a Height’, today’s Massacre Cave in Canyon del Muerto, where a Spanish punitive expedition killed perhaps a hundred Navajos in 1805. The name refers specifically to where a knife-wielding soldier struggled with a young Navajo woman attempting to climb up the rock from her hiding place in the cliff-shelter refuge; the woman

336 Stephen C. Jett

pulled the man over the edge, to both of their deaths. An example of a place name based on a repeated activity is a dune in lower Canyon de Chelly called Séí ’Adiléhé ‘The Adultery Sand [Dune]’; when ascending the canyon from the trading post at its mouth, this was the first significant place at which one encountered soft, clean ground and privacy (between the pile and the cliff) for fornication. As is the case with the other Apache peoples, Navajo places and their names are inextricably entwined with story. Interestingly, Klara Kelley and Harris Francis (2001:â•›42) found that when Navajos are taken into the field and questioned about place names, “the route itself dictates the sequence in which a person identifies particular places. At home, the person is more likely to mention places by telling a story or ceremonial sequence that interconnects a group of places”. Burenhult and Levinson (2008:â•›138) have written, “[P]lace names are one of the most conservative elements in a language”. Athabaskan languages appear to be particularly resistant to change (James Kari, personal communication, 2008; cf. Cruikshank 1990). Kari (1989:â•›129) concluded that Alaskan Athabaskan “oral place names seem to be quite stable and conservative over time” (also, Kari, this volume). Despite Navajo not being a written language before the twentieth century, nineteenth-century – and in one case, eighteenth-century – transcriptions/ translations of Navajo place names of Canyon de Chelly indicate complete continuity into the early twenty-first century. Moreover, the chantway myths and other sacred stories that provide cognitive maps contain nothing reflecting European presence or cultural introductions, suggesting minimum ages of over 400 years for the place names involved (Francis & Kelley 2005:â•›99). Therefore, we may suppose that the history of most Navajo place names – at least those in long-settled areas – is lengthy and that the names themselves have long persisted unchanged.

6.

Place names and making a living

Athabaskan speakers, including the Navajo, are noted for their pragmatism (Vanstone 1974:â•›125). Navajos depended – and still depend to some degree – directly on the natural resources within their habitat. Place names not only assist a person generally to find his or her way around in the landscape, as described above; in addition, some toponyms specifically identify a resource-producing site and thus aid in the processes of making a living and of engaging in certain cultural activities (Brugge 1993:â•›42; cf. Alaska’s Ahtna: Kari, this volume). One . In the wider Navajo Country, there is a modest percentage of placenames that memorialize battles, executions, and so forth (Brugge 1993:â•›42; for other Athabaskans, see Basso 1984:â•›27–32, 1990:â•›155, 1996:â•›23–24, 28; Kari 1989:â•›142, this volume; Cruikshank 1990:â•›56, 59, 60, 62–63).



Landscape embedded in language 337

de Chelly-system example of such a name is Sheep Point, known in Navajo as Dibé Dáád ‘[Bighorn] Sheep Obstruction’. Here, a sheer-sided point of upland projects from the rim out into Canyon del Muerto, via a narrow neck. This was an area of bighorn-sheep abundance; from the end of the point, men would drive bighorns toward the neck, at which a pair of hunters would lie in ambush where there was a natural wall that had but a single, narrow passage through it. Another instance is Chííh Hajitseełí ‘At Someone-Chips-out-Red-Ochre’. This is an outcropping in upper Canyon de Chelly where there once was a deposit of red ochre, mined as a sunscreen and as a pigment.

7.

Northern/Southern Athabaskan comparisons

The place-naming practices of all Athabaskan-speaking peoples so far studied share a rather lengthy list of commonalities. As I wrote in 2001: Correspondences include: stress on naming natural features instead of cultural ones (aside from trails); no borrowing of preexisting names; translatability of virtually all names; descriptiveness and literality as opposed to metaphoricality or allusiveness; non-specificity, i.e., a single name being applied to more than one feature of an area (but with specifying nouns used when ambiguity is a risk) or to an area larger than the named feature itself; avoidance of incorporating personal names in permanent place names; and frequent use of nominalized verbs as, or in, place names. Navajo differences from Alaskan Athabaskan practice are fewer but are also interesting. They include the same name being applied to more than one unrelated landmark, even fairly locally … [but see Kari, this volume, for repetition in Ahtna]. They also include Navajo use of alternative names for the same feature. These last include sacred versus secular names as well as more than one distinct secular name, and also, commonly, equal acceptability of alternative phrasings that convey meanings similar to more-or-less standard names.  (Jett 2001:â•›207)

These differences may reflect, at least in part, fusion with Puebloan culture following the pre-Navajo’s arrival in the Southwest, especially in the adoption of farming with its increased degree of sedentism and in the selective adoption and reinterpretation of elements of Puebloan religion.

338 Stephen C. Jett

8.

Conclusion

Onomastics – the study of names – is a subdivision of linguistics. Language is a means of dividing the vast, continuous world out there and the objects and interactions within it into bits small enough and discrete enough to be comprehended and communicated, and proper-naming is the application of this linguistic capacity to specific entities – in the case of place-naming, to particular localities (Thornton 1993:â•›295). Such names are enduring and shared identificatory labels that, in a sense, contribute to the creation of the cognitive reality of the places named, that help bring those places into the orbit of human consciousness, comprehension, and activity. While so doing, the names become nexuses where physical place and human observation, activity, and history conjoin. They represent particularly apt portals via which scholars may achieve improved understandings of landscape, perception, and language, as well as culture, history, and economy. As Susan Fair (1997:â•›478) wrote: Toponyms cluster on the land, drawing attention to complex associations between themselves and features of the landscape, … residence and land-use patterns, language and dialect, the transfer of information by gender, economic practices, local beliefs, history, morals and other traditional knowledge.

Rather than being insignificant, obscure, and peripheral objects of study, then, perhaps place names should be central to cultural, geographical, and even cognitive research.

References Afable, Patricia O. & Beeler, Madison S. 1996. Place-names. In Handbook of North American Indians 17, Languages, Ives Goddard (ed.), 185–199. Washington DC: Smithsonian Institution. Aton, James M. & McPherson, Robert S. 2000. River Flowing from the Sunrise: An Environmental History of the San Juan. Logan UT: Utah State University Press. Basehart, Henry W. 1974. Mescalero Apache Subsistence Patterns and Sociopolitical Organization: Commission Findings on the Apache [American Indian Ethnohistory: Indians of the Southwest 12]. New York NY: Garland. Basso, Keith. 1984. Western Apache place-name hierarchies. In Naming Systems: The 1980 Proceedings of the American Ethnological Society, Elizabeth Tooker (ed.), 78–94. Washington DC: The American Ethnological Society. Basso, Keith. 1990. Western Apache Language and Culture: Essays in Linguistic Anthropology. Tucson AZ: The University of Arizona Press. Basso, Keith. 1996. Wisdom Sits in Places: Landscape and Language among the Western Apache. Albuquerque NM: University of New Mexico Press.



Landscape embedded in language 339

Baumhoff, Martin A. 1958. California Athabascan Groups [Archaeological Records 16(5)]. Berkeley CA: University of California Press. Bray, Dorothy, with White Mountain Apache Tribe. 1998. Western Apache – English Dictionary: A Community-generated Bilingual Dictionary. Tempe AZ: Bilingual Press/Editorial Bilingüe. Bright, William. 2006. Native American Placenames of the United States. Norman OK: University of Oklahoma Press. Brugge, David M. 1993. An Investigation of AIRFA Concerns Relating to the Fruitland Coal Gas Development Area. Albuquerque NM: Office of Contract Archaeology, University of New Mexico. Burenhult, Niclas & Levinson, Stephen C. 2008. Language and landscape: a cross-linguistic perspective. Language Sciences 30(1–2): 135–150. Callaghan, Catherine A. & Gamble, Geoffrey. 1996. Borrowing. In Handbook of North American Indians 17, Languages, Ives Goddard (ed.), 111–116. Washington DC: Smithsonian Institution. Cruikshank, Julie. 1990. Getting the words right: perspectives on naming and places in Athapaskan oral history. Arctic Anthropology 27(1): 52–65. Fair, Susan W. 1997. Inupiat naming and community history: The Tapqaq and Saminiq coasts near Shishnaref, Alaska. The Professional Geographer 49(4): 466–480. Faltz, Leonard M. 1998. The Navajo Verb: A Grammar for Students and Scholars. Albuquerque NM: University of New Mexico Press. Francis, Harris & Kelley, Klara. 2005. Traditional Navajo maps and wayfinding. American Indian Culture and Research Journal 29(2): 85–111. Goodwin, Granville. 1932. Various unpublished manuscripts on Western Apache placenames. Tucson AZ: The Goodwin Archive, Arizona State Museum, The University of Arizona. Haas, Mary R. 1978. Language, Culture, and History: Essays by Mary R. Haas, Anwar S. Dil (comp.). Stanford CA: Stanford University Press. Haile, Berard. 1968. Property Concepts of the Navajo Indians [Anthropological Series 17]. Washington DC: Catholic University of America. Harris, Sally Pierce (interviewer). 1965. “Grey Eyes, Canyon de Chelly.” Doris Duke 695, unpublished transcript from the American Indian History Project. Salt Lake City UT: Doris Duke Western History Center, University of Utah; copy at Visitor Center, Canyon de Chelly National Monument, Chinle, AZ. Haskell, J. Loring. 1987. Southern Athapaskan Migration, A.D. 200–1750. Tsaile, AZ: Navajo Community College Press. Henry, Davis & Henry, Kay. 1969. Koyukon locationals. Anthropological Linguistics 11(4): 136– 142. Hill, Williard W. 1948. Navaho trading and trading ritual: A study of cultural dynamics. Southwestern Journal of Anthropology 4(4): 371–396. Hoijer, Harry. 1945. Classificatory verb stems in the Apachean languages. International Journal of American Linguistics 11(1): 13–23. Hoijer, Harry. 1956. The chronology of the Athapaskan languages. International Journal of American Linguistics 22(4): 219–232. Ives, John W. 2003. Alberta, Athapaskans, and Apachean origins. In Archaeology in Alberta: A View from the New Millennium, Jack W. Brink & John F. Dormaar (eds), 256–289. Medicine Hat AB: Archaeological Society of Alberta.

340 Stephen C. Jett

Jett, Stephen C. 1964. Pueblo Indian migrations: an evaluation of the possible physical and cultural determinants. American Antiquity 29(3): 281–300. Jett, Stephen C. 1970. An analysis of Navajo place-names. Names 18(3): 175–184. Jett, Stephen C. 1978. The origins of Navajo settlement patterns. Annals of the Association of American Geographers 68(3): 351–362. Jett, Stephen C. 1993. An introduction to Navajo sacred places. Journal of Cultural Geography 13(2): 29–39. Jett, Stephen C. 1995. Navajo sacred places: Management and interpretation of mythic history. The Public Historian 17(2): 39–47. Jett, Stephen C. 2001. Navajo Placenames and Trails of the Canyon de Chelly System, Arizona. Frankfurt: Peter Lang. Jett, Stephen C. 2002. The Navajo homeland. In Homelands: A Geography of Culture and Place across America [Creating the North American Landscape], Richard L. Nostrand & Lawrence E. Estaville (eds), 168–183. Baltimore MD: The Johns Hopkins University Press, in cooperation with the Center for American Places. Jett, Stephen C. 2006. Reconstructing the itineraries of Navajo chantway stories: A trial at Canyon de Chelly, Arizona. In Southwestern Interludes: Papers in Honor of Charlotte J. and Theodore R. Frisbie [Archaeological Society of New Mexico 32], Regge N. Wiseman, Thomas C. O’Laughlin & Cordelia T. Snow (eds), 75–86. Albuquerque NM: The Archaeological Society of New Mexico. Kari, James. 1976. Navajo Verb Prefix Phonology. New York NY: Garland. Kari, James. 1989. Some principles of Alaskan toponymic knowledge. In General and Amerindian Ethnolinguistics: In Remembrance of Stanley Newman, Mary Ritchie Key & Henry M. Hoenigswald (eds), 129–151. Berlin: Mouton de Gruyter. Kari, James. 1996. Names as signs: the distribution of ‘Stream’ and ‘Mountain’ in Alaskan Athabaskan languages. In Athabaskan Language Studies, Essays in Honor of Robert W. Young, Eloise Jelinek, Sally Midgette, Keren Rice & Leslie Saxon (eds), 443–475. Albuquerque NM: University of New Mexico Press. Kari, James & Fall, James A. 2003. Shem Pete’s Alaska: The Territory of the Upper Cook Inlet Dena’ina, 2nd edn. Fairbanks AK: University of Alaska Press. Kari, James & Potter, Ben A. (eds). 2010. The Dene-Yeniseian Connection [Anthropological Papers of the University of Alaska 5(1–2)]. Fairbanks AK: University of Alaska. Kari, James & Tuttle, Siri. 2005. Copper River Native Places: A Report on Culturally Important Places to Alaska Native Tribes in Southcentral Alaska [BLM Alaska Technical Report 56]. Anchorage AK: Bureau of Land Management, Alaska State Office. Kelley, Klara Bonsack & Francis, Harris. 1994. Navajo Sacred Places. Bloomington IN: Indiana University Press. Kelley, Klara Bonsack & Francis, Harris. 2001. Canyon de Chelly National Monument – ethnographic resources. CRM: The Journal of Heritage Management 24(5): 41–43. Krauss, Michael E. 1973. Na-Dene. In Linguistics in North America [Current Trends in Linguistics 10], T. A. Sebeok (ed.), 903–978. The Hague: Mouton. Leer, Jeff. 1989. Directional systems in Athapaskan and Na-Dene. In Athapaskan Linguistics: Current Perspectives on a Language Family [Trends in Linguistics, State of the Art Reports 15], Eun-Do Cook & Keren D. Rice (eds), 575–622. Berlin: Mouton de Gruyter. Levinson, Stephen C. 2003. Space in Language and Cognition. Cambridge: CUP. Levy, Jerold E. 1998. In the Beginning: The Navajo Genesis. Berkeley CA: University of California Press.



Landscape embedded in language 341

Linford, Laurence D. 2000. Navajo Places: History, Legend, Landscape. Salt Lake City UT: University of Utah Press. Luckert, Karl W. 1979. Coyoteway: A Navajo Holyway Healing Ceremonial. Tucson AZ: The University of Arizona Press/Flagstaff AZ: Museum of Northern Arizona. McDonough, Joyce. 2003. The Navajo Sound System. Dordrecht: Kluwer Academic. McPherson, Robert S. 1992. Sacred Land, Sacred View: Navajo Perceptions of the Four Corners Region [Charles Redd Monographs in Western History 19]. Provo UT: Brigham Young University. Nelson, Richard K. 1983. The Athabaskans: People of the Boreal Forest/Ts’ibaa Laalta Hüt’anna [Studies in History 27]. Fairbanks AK: Alaska Historical Commission, University of Alaska Museum. Nelson, Richard K. 1994. The embrace of names. In Northern Lights: A Selection of New Nature Writing from the American West, Deborah Clow & Donald Snow (eds), 14–21. New York NY: Vintage Books. [repr. from The Island Within (New York: North Point Press, 1989)] Perry, Richard J. 1991. Western Apache Heritage: People of the Mountain Corridor. Austin TX: University of Texas Press. Pinxten, Rik, Van Dooren, Ingrid & Harvey, Frank. 1983. The Anthropology of Space: Explorations into the Natural Philosophy and Semantics of the Navajo. Philadelphia PA: University of Pennsylvania Press. Ruhlen, Merritt. 1998. The origin of the Na-Dene. Proceedings of the National Academy of Sciences USA 95(11): 13,994–996. Smith, Carlota S. 2000. The semantics of the Navajo verb base. In The Athabaskan Languages: Perspectives on a Native American Language Family, Theodore B. Fernald & Paul R. Platero (eds), 200–227. Oxford: OUP. Thornton, Thomas F. 1993. Tlingit and Euroamerican toponyms in Glacier Bay. In Proceedings of the Third Glacier Bay Science Symposium, 1993, D. R. Engstrom (ed.), 294–301. Anchorage AK: U.S. National Park Service. Thornton, Thomas F. 1997. Anthropological studies of Native American placenaming. American Indian Quarterly 21(2): 209–228. Thornton, Thomas F. 2008. Being and Place among the Tlingit. Seattle WA: University of Washington Press. Vajda, E. 2010. A Siberian link with Na-Dene languages. In Kari & Potter, 33–99. Van Valkenburgh, Richard F. 1974. Navajo sacred places. In Navajo Indians III, 9–199. New York NY: Garland. Van Valkenburgh, Richard F. 1999. Diné Bikeyah. Mancos CO: Time Traveler Maps. [orig. pub. 1941] Vanstone, James V. 1974. Athapaskan Adaptations. Chicago IL: Aldine. Vecsey, Christopher. 1988. Imagine Ourselves Richly: Mythic Narratives of North American Indians. New York NY: Crossroad. Watson, Editha L. 1964. Navajo Sacred Places. Window Rock AZ: Navajo Tribal Museum. Wilson, Alan & Dennison, Gene. 1995. Navajo Placenames: An Observer’s Guide. Guilford CT: Jeffrey Norton. Witherspoon, Gary. 1977. Language and Art in the Navajo Universe. Ann Arbor MI: The University of Michigan Press. Wyman, Leland C. 1957. Beautyway: A Navajo Ceremonial [Bollingen Series 53]. New York NY: Pantheon Books.

342 Stephen C. Jett

Yazzie, Evangeline Parsons & Speas, Margaret. 2007. Diné Bizaad Bínáhoo’aah: Rediscovering the Navajo Language. Flagstaff AZ: Salina Bookshelf. Young, Robert W. 2000. The Navajo Verb System: An Overview. Albuquerque NM: University of New Mexico Press. Young, Robert W. & Morgan, William. 1980. The Navajo Language: A Grammar and Colloquial Dictionary. Albuquerque NM: University of New Mexico Press. Young, Robert W. & Morgan, William. 1992. An Analytical Lexicon of Navajo. Albuquerque NM: University of New Mexico Press.

chapter 16

Navajo landscape and its contexts* Carmelita Topaha

This essay, which began as an audio-recorded interview, reviews some of the ways in which particular features of the landscape are embedded in Navajo tradition and culture. Also discussed are some of the complications of integrating modern life in the USA with traditional beliefs and practices.

1.

Cultural aspects of landscape

My background is archaeology and when we used to excavate on the Reservation in the late 80s, I would be down in the kiva helping out or sifting for artifacts; and if there was a road nearby, people would stop by to see what was going on. Navajo people. And the grandmas would come out and when the grandmas saw me, they immediately used to scold me because, number one, I’m a Navajo woman, so I should not be in the Pueblo kiva. I mean, that’s another tribe, so I was being disrespectful. And so they would get after me and they would say “You better get out of there and go have yourself a sing”. And so, I would tell them “After this, after today is over, I have my corn pollen and I do my own little ritual”. And when you talk about landscape, a lot of this excavation took place over there in Lukachukai area, over there in Arizona. We were excavating the road over Buffalo Pass to Red Valley, and now, today, in 2010, it’s a paved highway, but prior to that we excavated the road from Lukachukai over the mountain. And they would stand there and, you know, I didn’t say anything back, and they would just ask questions and I would tell them what we’re doing. And then they would say, “Why are you touching human remains?” And I say, “No, you know, I don’t do that”. And they would say – they would point to the mountain and they would * This essay began as a transcript of an audio-recorded session with Carmelita Topaha and David Mark, conducted in Farmington, New Mexico, on Sunday, June 27, 2010. 1. A “sing” is a traditional Navajo ceremony performed to remove an illness, remove effects of a bad experience, etc.

344 Carmelita Topaha

point to the four directions and say “You know, this land was all given to us by the holy people and what you’re doing and what the Bilagáana people are doing is disturbing the people who came here to live”. You know, in Navajo tradition, the entire mountain range including the Chuska Mountains is considered to be a human form, a woman laying on her side. And you’re disturbing the inside of Mother Earth’s stomach. And you’re probably removing sacred things that she has, like maybe something that represented a necklace, because of what that particular mountain is; you know, it could have been a female or male, depending on where the drainage came from on the mountain, like the arroyo, that could be part of where the blood ran. And so, they just went into a lot of detail about landscape and doing archaeology. They said that the people before us knew where to live and they did not just come there to say, “This is our home right here”. They had to check it out before. They had to see certain things in the mountain. They had to see where the sun went down and what shadow or what sun rays touched various parts, depending on whether it was a female or male mountain and the various sacred parts of that mountain. Where was the heart? Where was the head? So all of these things related to why Navajo females should not go into the world of being an archaeologist. Different Navajo people, especially those who live there, they have their own knowledge of what that particular mountain has. When you talk to an outsider from different places, they usually, you know, they have their own stories of what was passed down to them. For instance, when we were excavating and the older people stopped by and the men would say, “Well, which way is the road going to go?” then I would say, “Well, I don’t know. I think it’s going to go, you know, this way and that way, but it’s going to go probably right here”. And the guys would say, “Well, we don’t want the road”. They go, “Is the road going to follow the old dirt path or dirt road up on the mountains?” And I go, “It probably will”. And they would say, “Oh good, because there are some areas that we don’t want to be disturbed because there’s some burials somewhere in some areas”. And so they said that was from the old people. And so they asked how many burial sites and artifacts we found and I said we won’t know until the end of the excavation period, but it was interesting to hear them talk – the local people there talk about Lukachukai itself and how it relates to their community, to their ceremonial – there’s just such an attachment to the mountain. So, the concern about our excavations was partly about the issue of the archaeological sites and places associated with death and the Anasazi, and partly about the fact that the mountain is the body of a Holy Being, with different parts

. “Bilagáana” is a Navajo term for ‘white’ people, especially English speakers.



Navajo landscape and its contexts 345

of the body in particular places. Many people still think about landforms that way, what the landforms and rivers represent in the origin stories and so on. Here is another example. When I came back from California, I used to take my aunties to the Window Rock Fair every year, three of my aunts, and my mom was long gone, passed away by then; and I used to take my aunts and they’d get all dressed up in their best skirts and they would wear their really nice squash blossom jewelry, and as we were going that way, they would tell me stories of the landscape. And one of my aunts, my Aunt Lucy would be the one, she was the storyteller, and anywhere we went she would just tell stories of what happened here and, you know, what do you call it, place names and things like that; and I didn’t realize, at the time when she was telling me these stories, that there was a connection. Now that I’m older, I realize that the reason she was telling these stories is because she wanted to pass those down to me, you know. She didn’t come directly to me and say “Carmelita, you know, I want to tell you these stories and I want you to remember them so you can tell your grandchildren”, and da da da da. They didn’t do things like that. You just had to be at the right time, not necessarily age-wise, but when the Elders would come to you, then it was your turn to take on this task; and I didn’t realize. And I always thought “Oh I’ll wait till I’m older and then the aunties, the grandmas will tell me these things”, you know, but you never know when these tasks are going to be told to you so you can take them on. Anyway, as we were going over the Narbona Pass going to Window Rock, to the Window Rock Fair, my aunt told me to stop. So I pulled over to the left side of the highway in an area where the ground is green, completely green on one side. And she would get out and she tossed her corn pollen, and my other aunts would get out and they would say things in Navajo: “Your spirit is still here”. I can’t even remember the old words they used to say in Navajo, but it was about paying homage to the woman, to the part of the female – I guess you might say finger of the mountain or branch of the mountain. And so we got back in the vehicle and she says, “The green part is the stomach”. She said “When the Warrior Twins got in a fight and the mom tried to break them up they accidentally hit her or punctured her and her stomach ruptured and there she lay and there the green out of her stomach came out and that’s why that area is there”. So that particular place is fenced off and is a sacred area to the Navajo people. A lot of the landscape is connected to the “origin stories” and the journeys of the holy beings. It’s all connected to where in the beginning, my aunts would tell me and my mom would tell me and my grandmother, where clans originated at that place and so on. My grandpa was from Crystal and everybody in the

. The author spent some of her childhood at a school in California.

346 Carmelita Topaha

Newcomb area knew that he was from Crystal, and so they would give some people names that are associated with that particular area. And the same way with my other grandpa, Sam Manuelito, he was from another area, from Newcomb, but where he lived wasn’t really in that area. They used to live close to the mountains, the Chuskas, so that’s where our clan is from, they’re called Tó’áhaní, which means something like “near the water clan”. Several years ago, a gentleman named Carl Todachini took a couple of classes from me at San Juan College, and at the time he was taking my classes he was about 80 years old, and he was so knowledgeable and so proud that I was talking about Hastiin Klah, our great-grandpa, and he would sit there and he says, “I remember all the stories”. And he would say things in Navajo and when he would say these particular Navajo words, I didn’t know what they were, if they were somebody’s name or if it was a place name, if it was a location, because I didn’t grow up in Newcomb. But he would say, “I’m going to take you back out there and I’m going to show you where all these places are. They have all these little things that – when I grew up there as a child, as a young boy, I would see your mom and your aunties or sisters and they would all gather and go out to the mountains and take the wagon”. You know, they didn’t even have a road at Narbona Pass at the time, but there was a place they used to go up to the mountain and they had different landforms, different land names for them. And I guess, today, I don’t know of anyone who knows those places because, I’m sad to say, Carl Todachini died€– I would guess about maybe five years ago. But he had so much knowledge of Shiprock, and to this day I could hit myself for not taking him out there to show me where those places are. I remember my sisters, we came up from California – I think it was in the 80s or late 70s – we came up from California for a funeral, and so I wanted to know where my grandpa Hastiin Klah lived. And my aunties, my mom, they all looked at each other and said, “We don’t go there. We haven’t been there since gosh”, but I said “Is it around here?” And they go “Oh yeah”. And I said “Well, why don’t you go there?” And they go “Well why?” And I said, “Well, I want to know. I mean, I want to see it”. So my mom says, “Okay”, she says. So we all hopped into the truck and I was driving and, you know, I don’t know if I can find it to this day, but anyway, we went to this area, it probably took us about 30 minutes to get there from Newcomb Trading Post, and we got there and it was – the old hogan was still there, some walls, old walls were still standing and a sheep corral, and my aunties just started crying and they just said “Oh” and my Aunt Lucy, she was born there, and she started crying, my, and they started saying Navajo words I wasn’t familiar with. And then, themselves, they started saying “You know what? We don’t even use those words any more”. Those were Hastiin Klah’s words, you know, when



Navajo landscape and its contexts 347

he was teaching the girls how to be responsible for his sacred ceremonial paraphernalia, each of his five nieces was taught to take care of certain things because Hastiin Klah was a famous Medicine Man in the 20s and the 30s. So he taught each of these girls how to take care of things and so they knew the songs, the astronomy, the landscape, the harvests, the plants, and they knew how to wash the Yeibichei masks, there’s a certain way to handle them, you don’t just lay them down. It has to be a certain cardinal direction. It has to do with, however, if it’s a male or a female, you know, whoever the patient is – a lot of special directional and seasonal knowledge is involved, and how it’s placed and when to pick plants, and all these things that the girls were taught. And I wish, when we went to Hastiin Klah’s old place, you know, I was always interested in history but I never took it more than that and I sat there and listened to their conversation, but they were talking about memories of when they were growing up with Hastiin Klah, and those words are no longer used today. So I was having a hard time trying to understand what they were saying. And they would laugh and say “Oh yeah, remember when he did this or when he did that?” And then they were talking about certain things about the Newcomb area, how come he chose that particular area to live at, you know. I mean, it’s a kind of flat area but there’s this pinnacle called Hwiishch’aa and it correlates with Cemetery Ridge. On the west side of the school at the end of the Chapter House, that’s an archaeological site; so there’s all these connections to landscape. Now that I’m older I think back about all these things and how they put in perspective why these things are that way. I know when Carl Todachini used to talk about Newcomb, he said it was just us; meaning us including the relatives – there weren’t any outsiders there; it was just the big family of Manuelito and Hastiin Klah. Hastiin Klah was very wealthy, meaning not as far as monetary value, but as far as having cattle, sheep and lots of wagons, and he was really a prominent Medicine Man. When he would have a sing, they predicted there was over maybe five hundred – at least four to five hundred people would arrive by wagons and horses and they would stay like a week. So, he was pretty popular in the 20s and 30s. But then, in the 1920s and 1930s, Hastiin Klah worked with Franc Newcomb and Mary Wheelwright and others to allow some of the stories and songs to be written down and published (Klah 1942; Wheelwright 1946).

. Hwiishch’aa is mentioned in Haile’s (1978) report on the Mothway (Frenzy Way); there is a part of the story where Scrap Picker Boy and his Grandmother coming through the Chuskas and come out near Newcomb, at Hwiishch’aa.

348 Carmelita Topaha

2.

Landscape terms

I’d like to comment on the project that David Mark, David Stea, and Andrew Turk are doing to make a dictionary of the terms for kinds of things in the landscape (Turk et al. this volume; Mark & Stea 2010). Whether it is worth doing, or whether it is just scratching the surface. My background is in archaeology and I always wanted to be an archaeologist, and I worked for the University of New Mexico archaeology company called OCA, Office of Contract Archeology, and through being an archaeologist, with a degree in anthropology, I was able to teach at UNM. And from there, when I started teaching, colleagues of mine would say, “Did you ever do this on the Res and that?” And I said, “No”. And so, I would start thinking “You know, that would be my next place to be – back on the Res.” So, I made my way back to the Reservation and I got hired by a Navajo archaeology company. And so I already knew the protocols, and when I taught my first class at UNM, it was called “Introduction to Native American Archaeology”; and the reason I taught that class was because all my archaeology friends, directors, and crew chiefs were all Anglos and they had no clue of native rights and beliefs about archaeology. So that began to be an interest of mine, how to develop protocols and to do things on the Res. I already had several strikes against me. It was already against my culture to be a female archaeologist, number one. And so, when you would go out there and do things, I always used to hesitate, you know, meeting grandmas, because they might scold me. The old men, the elderly men, they wouldn’t do that. And so, I just had to come up with something and say “Well, that’s part of the job. That’s what I wanted to do. You know we need to preserve our land. Who else knows the land better than the people here? Because they can come in and they can disturb a site, they can disturb a sacred area where you guys have been practicing or praying or singing and gathering, at this particular area that nobody else knows. So, how else to preserve things, so that you don’t have oil pads come in, you know, and put it right here where maybe there’s a burial or something that happened here, like, you know, 50 years, 75 years ago that you want to preserve?” So I would just tell them that and they would say, “Yes, okay, that’s fine”. So when I started going on the Res to do archaeology, I was fortunate; I went to a lot of remote areas and I would see the landscape and I was always interested, even before this project started, I was always interested in how people see the land. Because I would stand by their house and I would look out and see what the horizon was. And it would look like the different mountains and mesas and flat valleys are nothing at all, and I would say, “How did they see this?” And so by going out in the remote areas, I found a lot of Navajo people who knew so much about the landscape, especially the weavers, and the sheepherders. They would say things



Navajo landscape and its contexts 349

and they would tell me, when I go out there and do archaeology or go do survey, where their home site is going to be, and they will say “Oh just go walk down and go by that – where it slopes or where the end of that big boulder is, that’s where such and such happened”, and, of course, I didn’t know where these places were because that wasn’t my area. So I would have somebody go with me and I would walk out there and do an acre, maybe sometimes two acres, and I would ask them “Well, how did this become this name?” And they would say, “Oh, this happened. It was grandma.” It was always grandma who was naming these things, but there was so much history in how people viewed the land. Now that I have been with the Ethnophysiography Project (Turk et al. this volume), I believe it’s six years, I look back and I think “You know, there’s so much more still to this project”. We’re just scratching the surface, because now that I’ve seen the partial draft of the Navajo geography dictionary, I see a lot of appendices that could be in an edition, not just seeing a word along with picture but adding mapping and GIS. At least somebody is trying to preserve the landscape and the language. Who knows, in 50 years, 75 years, the places that we’ve gone to and talked to the elders and the people that took us to their sacred areas or, you know, favorite areas; who knows, those places may be gone, but at least we’re doing something to preserve history or preserve a community or a location with pictures and photographs. There’s an issue of outsiders getting involved in this because, on the one hand, things get preserved that might be lost otherwise, but then on the other hand the outsiders know the stories and may be cashing in on local culture by publishing books or articles. This happened with Hastiin Klah: he got in trouble with other Medicine Men for telling Mary Wheelwright and Franc Newcomb some of this material and allowing them to publish it, but it would probably be lost now if he hadn’t done that. Some might say maybe it would have been better, have been right for some of this knowledge to have been lost, rather than writing it down where anyone could see it. It’s a balance sometimes. When you may say it would be a given that it be lost, I think, on the female side of our Tsé taa’aanii clan, I think it’s, how would you say, it’s a plus that it’s not lost. My aunties, they still have the memories of what was told. I mean, nothing was written down, no photographs, no recording, no video. And at the time, you know, we weren’t into that technology. We didn’t even have it. I mean, on the Reservation, there were a lot of things we didn’t have. So I understand when you ask, you know, “Why don’t you go over there and see grandpa’s place?” and stuff; it’s simply forbidden because they have all those memories. They don’t need to go and check it out, as you might say. But going back to where you said it’s a doubleedged sword, and it is – you know – I mean, it seems to always take an outsider, meaning a non-Indian, to come in, to come to the Res and, how do you say it,

350 Carmelita Topaha

research. And throughout the history of the Navajo people there’s been many, numerous archaeologists, anthropologists, you name it, who have come to the Res and taken a lot of information, publish and, you know, they did their dissertation or whatever, and that’s part of the western world, you know, that’s part of academics in a university. I understand that part, but yet on the other side of the fence – I mean, I went to college, you know, I got a degree in anthropology and I would like to not study my culture. That’s a word I don’t really like to use, but I can’t seem to get into that because – well, the fact is, I know my culture, you know. Maybe if I went and did research on the landscape concepts of another Native American culture, maybe that would be a lot easier for me to do. But regarding the idea with doing this landscape language project with the Navajo people and language, I’m very grateful that [David Mark, David Stea, and Andrew Turk] approached me about participating in the project. It’s preservation you’re looking at, preservation of language and culture, and also they’re geographers and looking at the land and seeing how everything is situated, but once the Navajo people realize that this is being preserved and given back to the community and given to the children of the next generation, we hope to inspire young people to go into fields like geography and anthropology and to help preserve their culture and landscape, because where is it going to be at in 50 or 75 years, especially the landscape? The older people – you know, a lot of things happen with landscape and there are some stories written in communities of certain significance of landscape areas, but it’s still not enough. I guess it’s always like this when you do a project: you look back at all of your notes and data collected and maps and interviews and you always think “Well, it could have been done better this other way or whatever”, but you really don’t know that until you come to a certain time period and look at it and, you know, you go over it and you ponder it and you have feedback and you talk with your colleagues about it. I guess I won’t really know the full extent of it till I actually get that CD version of the Illustrated Dictionary of Navajo Landscape Terms (Mark & Stea 2010), and then put it in my laptop and then see all the gorgeous pictures and all the beautiful comments that were said about landscape and the Navajo language.

References Haile, Berard. 1978. Love-magic and Butterfly People: The Slim Curly Version of the Ajiłee and Mothway Myths. Flagstaff AZ: Museum of Northern Arizona Press. Klah, Hasteen. 1942. Navajo Creation Myth: The Story of the Emergence [Navajo Religion Series I]. Recorded and edited by Mary C. Wheelwright. Santa Fe NM: Museum of Navajo Ceremonial Art.



Navajo landscape and its contexts 351

Mark, David M. & Stea, David. 2010. Illustrated Dictionary of Navajo Landscape Terms, compiled with assistance from Larry King, Carmelita Topaha, Harding Yazzie, Jr., Andrew Turk, Jay Williams, and others. Informal publication, University at Buffalo NY. Wheelwright, Mary C. 1946. Hail Chant and Water Chant [Navajo Religion Series II]. Santa Fe NM: Museum of Navajo Ceremonial Art.

chapter 17

Navigating regional landscapes with Jicarilla personal narrative Elizabeth M. Lynch

Interpretation of how prehistoric hunter-gatherers may have charted the landscape they inhabited is often based on the physical materials left behind. The potential exists to interpret Jicarilla narratives as providing mental templates for movement through and perception of their environments. Personal narratives may have facilitated remembering significant paths and passing this knowledge on to others. These narratives are rich resources that inform our understanding of how the Jicarilla, and perhaps prehistoric plains people before them, may have perceived and travelled through the complex eco-corridors that buffered the Pueblos, extending onto the plains of northern New Mexico and southeastern Colorado. This chapter examines whether specific Jicarilla Apache narratives encoded navigation patterns within oral tradition. Personal hunting narratives are examined as pathfinding mechanisms. Creation stories and ceremonial landmarks mentioned in the text are treated as points of significance on the landscape. Cognitive maps were developed that reveal a series of patterned movements, which may prove insightful in interpreting prehistoric movement and spatial construction.

1.

Mapping, navigating and oral tradition

Navigating familiar and unfamiliar space offers unique challenges for most people, present and past (Golledge 1999:â•›15). Typical challenges include finding locales or places (those where food may be found and obtained, those important to maintaining and building social, economic, and symbolic relationships, etc.) and then finding one’s way home again. A parallel challenge is relocating significant places and then, when necessary, “communicating this spatial knowledge to others” (Golledge 1999:â•›15). The use of abstract maps, from those drawn on restaurant napkins to those created with global positioning systems, is one way that individuals living in complex societies have resolved the reproduction of this

354 Elizabeth M. Lynch

knowledge so that they may, with some efficiency and certainty, navigate through their physical environment. What emerges from these maps is an abstract vision of networks and symbolic places (Golledge 1999; Tuan 1975, 1977). Of course, movement and direction along these paths is at the discretion of the individual, who is shaped by and affects the cultural and physical world into which they are initiated. Personal narratives are effective means of encoding this knowledge by recording experience of events, movement and places (Basso 1996; Tuan 1976). This information, much like the giving of verbal directions, may, in turn, be mapped into abstract reproductions of symbolic and practical place knowledge. Lacking similar landscape narratives from prehistoric people, it is difficult to infer how they may have created symbolic and practical networks of their world. We may reason that they were once in a certain region, or at a particular place within a region, from the archaeological record. We may reason significance and symbolic meaning from the type of artifacts left behind (rock art panels, ceremonial figurines or pottery, luxury material artifacts, etc.) and from ethnographic accounts. However it is almost impossible to know how prehistoric hunter-gatherers may have moved between places (symbolic and domestic) on the landscape and created pathways through their world. Cognitive mapping, defined as “one’s internal representation of spatial information” (Golledge 1999:â•›15), can be used in conjunction with oral tradition to reconstruct movement between significant places. Oral tradition can be viewed as a symbolic framework to view the world and conceptualize one’s environment (Basso 1996; Tuan 1976). Points of departure and return, pathways, and places along pathways may be embedded within the symbolic representations of society (Tuan 1977) and retold through narratives. These stories then become a foundation for the creation of cognitive maps of the landscape and may provide invaluable insight into how prehistoric peoples may have viewed and navigated their environment. Successful navigation of the landscape requires an ability to read landmarks and make decisions about the appropriate course to follow (Kitchen 1996). Golledge (2003:â•›34) proposes that landscapes are made “legible” by the “sociocultural meanings of the surroundings, incorporating qualitative and emotional characteristics with the physical and spatial.” In groups with strong cultural heritage and social ties, this may have the effect of increasing the intensity of the legibility of the environment (Golledge 2003:â•›34). Golledge (2003:â•›36–39) discusses three ways of making the landscape “legible”: the use of frames of reference, landmarks, and paths and networks. Several ways exist to physically and mentally negotiate landscape through cognitive mapping. Of those discussed by Golledge (2003) and Allen (1999),



Navigating regional landscapes with Jicarilla personal narrative 355

“piloting” and “chunking” are most applicable to this study. Piloting occurs when “the current position is known and observed landmarks are integrated into a representational self-referencing system”. Chunking is a process through which “a reasonably long or circuitous route can […] be subdivided into a number of manageable chunks which allow an individual to develop a parsimonious memory structure to define a complex route” (Golledge 2003:â•›23, 29). Jicarilla texts recorded by Pliny E. Goddard (1911) can be used to construct mental maps of traditional Jicarilla landscape. In this context, traditional origin stories operate to create the frame(s) of reference discussed by Golledge (2003) which are needed to successfully find one’s way. The narratives act to center the individual on the landscape, providing a sense of belonging and security. Landmarks are the places to which the stories navigate, acting as lighthouses of a sort, as individuals ‘pilot’ across the plains to other landmarks or places. Networks and paths are the routes that connect places within the broader Jicarilla space. Paths may have been viewed by the storytellers as a means through a dangerous or unconscious space or as networks connecting places, but as discussed below the paths are patterned in space and are encoded in the narratives.

2.

Historical context

The Jicarilla are one of the Apachean groups that were part of a larger movement of Athabaskan language speakers who are believed to have begun migrating south, from Alaska and Canada, circa 1000 years ago (Hoijer 1956:â•›232, 1971; Tiller 1978). The exact route they navigated may have been through the Rocky Mountains or along the eastern slopes (Gunnerson 1969; Gunnerson & Gunnerson 1971; Huscher & Huscher 1942; Tiller 2000). Hoijer (1956:â•›232) places the Apachean languages in the southwest by about 600 years ago. The Jicarilla, as distinct from other Apachean speakers, appear in the writings of early Spanish explorers as they moved to stake the crown’s interest north of Mexico City (Schroeder 1974; Stanley 1962). Oral tradition can be used to investigate a people’s movement around their landscape (Echo-Hawk 2000; Tiller 2000). Traditional Jicarilla origin stories describe a gradual emergence from a dark underground world via several mechanisms (Mooney 1898; Opler 1936, 1994; Russell 1898). This may represent a movement from distant places, perhaps from northern latitudes that remained in darkness for much for the year, or it may represent a type of symbolic cultural birth (Echo-Hawk 2000:â•›273). From this emergence place, the ancestral Jicarilla moved throughout the world, facing many challenges, some choosing to settle along the way (Goddard 1911; Mooney 1898; Opler 1936; Russell 1898). While the

356 Elizabeth M. Lynch

Jicarilla origin stories are similar to those of other Apachean groups, by the time the Spanish Coronado expedition first encountered these Plains peoples, their ancestors had already settled in the region around modern day Taos, New Mexico extending east onto the plains to the Arkansas River (Opler 1961; Stanley 1962). Jicarilla origin narratives do not specify when the people arrived in the southwest, but place their traditional homeland in northern New Mexico and southern Colorado (Goddard 1911). According to Mooney (1898:╛199), while others who emerged with the Jicarilla wandered, the Jicarilla wished to live in the center of creation, which is placed at Taos, New Mexico. Their territory extended to four sacred rivers: the Arkansas (north), the Pecos (south), the Chama (west) and the Canadian (east) (Mooney 1898:╛200). The Jicarilla were traditionally divided into two main bands: those who lived in the mountain valleys west of Taos (Olleros) and those who lived on the plains east of the Sangre de Cristos (Llaneros) (Tiller 1978, 2000). However, this distinction may be a superficial one since the Jicarilla seem to have always considered themselves one people (Opler 1936; Tiller 2000). The Jicarilla narratives used in this study provide a record of a system of movement among places within the Jicarilla landscape as well as insight into their relationships with their neighbors. The traditional boundaries of Jicarilla Apache territory circumscribe a diverse human landscape mosaic (sensu Rodewald 2003:╛587). This mosaic, much like that encountered by the Spanish, included a complex system of sedentary Pueblo villages under the protection of the Spanish and often at odds with the Jicarilla (Stanley 1962) and eco-corridors (trails or paths through an often undomesticated but well-traveled landscape) through which semi-�sedentary hunter-gatherers migrated from homes near the Rio Grande, through the Sangre de Cristo Mountains, onto the plains to hunt buffalo.

3.

Reconstructing Jicarilla cognitive maps

This study explores oral tradition in the context of the construction and understanding of cognitive maps among Jicarilla. The texts used were recorded by Pliny E. Goddard and are divided into four parts, containing “myths,” “tales,” “traditions and personal experiences” and “information concerning industries and ceremonies” (Goddard 1911:â•›1–5). The stories were obtained from two Jicarilla male individuals: Casa Maria and Juan Pesita (Goddard 1911:â•›8). The texts were recorded in the Jicarilla language and help with the texts was provided by two other Jicarilla (Goddard 1911:â•›9). The latter may have included Ed Ladd, who provided Mooney with some Jicarilla narratives a decade earlier (Mooney 1898:â•›26; Goddard 1911:â•›9). Jicarilla Apache Texts (Goddard 1911) is in two formats: one is



Navigating regional landscapes with Jicarilla personal narrative 357

the text of the narrative in Jicarilla with the English translation below and the other is a free translation of the narrative. The texts used for developing the narrative maps below were related by Casa Maria. The Jicarilla place names are recorded according to Goddard’s translation and at times vary slightly. In reconstructing the cognitive maps, these texts are considered as artifacts or remains of the individual’s frame of reference (how he or she viewed and meaningfully ordered the world outside). The origin stories are viewed as a mental setting because they are social stories, uniting all Jicarilla in time and place. They provide the meaning and context for the landscapes and landmarks encountered as well as boundaries for the interpretation of events which occur to the narrator. The personal narratives, which may be further subdivided, are considered in light of movement between places. They create and define the paths and networks that connect the places in the space of the Jicarilla landscape. Both the origin stories and narratives, as recorded by Goddard, reveal significant places, those which have meaning beyond the individual because they are named and mapped in stories (Basso 1996). It is not a far stretch to infer that the remembering of these place names confirms the importance of the places. Basso’s Wisdom Sits in Places (1996) is an example of how the importance of certain landscape features is created and maintained by the social memory of the group. For the Jicarilla, this socializing process (Bourdieu 1991) begins at a young age so that by adulthood the significance of landmarks, the perception of landscape and the mental organization of the environment would have been deeply embedded in the mind (Tiller 2000). Goddard’s (1911:â•›242) consultant Casa Maria would have been between 12 and 14 when he was involved in the military encounters between the Jicarilla and Kit Carson (Goddard 1911:â•›125, 144). In Apache society grandparents are the traditional instructors of the young (Tiller 2000). Boys accompanied raiding and hunting parties at a young age, which facilitated the learning process by incorporating the physical landscape with the stories (Tiller 2000:â•›23). Girls, on the other hand, gained their knowledge from their grandmothers and would have developed a more detailed knowledge of resources close by the campsites (Tiller 2000:â•›23, 26). To reconstruct the distant Jicarilla movement across this landscape, abstract maps will be developed from the personal narratives recorded by Goddard (1911). Important places and ceremonies are mentioned in the origin stories and provide a sense of place within the order of the universe. Nodal maps (abstract constructs linking central places via a network of directional movements) are constructed from the personal hunting accounts. The origin stories (those recorded by Goddard 1911; Mooney 1898 and Russell 1898) seem to refer to an ancestral time, and while they are important providers of context and shared social meaning, they are not conducive to the drawing of maps. The maps themselves

358 Elizabeth M. Lynch

simply “map” place names, derived from the narratives and plotted directionally on a blank space. It would be misleading to attempt, at this point, to attach these places to real locations, unless noted as such in the text, or to real distances. Tiller (2000:â•›21) mentions the trails and hunting strategies that would have been used after the harvests had been collected in the fall. Further evidence for camp sites is noted in Tiller (2000:â•›15, 22, 25) and in Nordhaus (1995:â•›25). The groundtruthing of the narratives collected by Goddard to other traditional stories, oral histories, historical records and the archaeological evidence of this region is beyond the scope of this chapter but is important work that remains to be done.

3.1

Ancestral landscapes

The origin stories of the Jicarilla (Goddard 1911; Mooney 1898; Russell 1898) create the mental space in which all of the narratives occur. Through the telling of the origin story (how the Jicarilla became Jicarilla), we are brought into the present-tense of the individual narrator through the ancient-time emergence and subsequent odyssey of his people. The story provides the framework of the narrator’s past. It places the individual in the knowable and experiential world. Accounts of the origin stories vary in some of the details, but all accounts place the people in a dark underworld (Goddard 1911; Mooney 1898; Opler 1994; Russell 1898). Here, all creatures have the same ability to communicate with each other (Mooney 1898:â•›198). As the story continues, some of the people decide to leave but cannot reach the way out, and so they build various ladders and mounds to try to reach the light they see above (Mooney 1898:â•›198). Once successful, they begin to wander the earth around this place of emergence, leaving only an old couple behind (for a full account consult Goddard 1911; Mooney 1898; Opler 1994; Russell 1898). After emergence, the world is made safe for people to inhabit by a Culture Hero who rids the region of monstrous fish, elk and eagles who would threaten their survival (see Goddard 1911; Mooney 1898, Opler 1936; Russell 1898 for a full account of these adventures). Though other groups who emerged with them from the darkness have found places to live on the earth, the decision to settle is not an easy one for the Jicarilla. Finally, with the encouragement of the creator, they decide to settle near Taos, New Mexico, the center of the world brought into being by the Creator (Mooney 1898:â•›200). Sacred rivers surround this area, which is rich in hunting resources and plant foods (Mooney 1898:â•›200). This understanding of Taos as the “heart” of the Jicarilla territory and people is still strong today (Nordhaus 1995; Tiller 2000).



Navigating regional landscapes with Jicarilla personal narrative 359

The names of places where ancient events occurred provide a central point and a sense of boundaries, but more than that, they provide an overarching framework from which the Jicarilla could explore and know their landscape, which had been made safer for them. Details change depending upon the narrator but key aspects remain the same: the center of the world is always the area around the Taos Pueblo (Goddard 1911:â•›7) and the places where monsters (generally oversized animals such as elk, fish or eagles) were killed are to be understood as real places on the landscape. Goddard’s (1911) consultant related the creation story and the subsequent challenges of the Culture Hero but used few place names in doing so. Mooney (1898:â•›205) tells us that the eagle resides at “Standing Rock” (Tsé-aí), “a high, steep cliff westward from Taos, with two sharp peaks on the eastern and the western ends and a depression in the center between them.” Mooney (1898:â•›208) provides a place location, “Cieneguilla, east of the Rio Grande and southwest of Taos”, for another dangerous creature: “Rock that Runs” (Tsĕ nanlki’ň). According to Mooney’s (1898:â•›208) consultant, “the rock is still there, lying on a level flat€– a black rock as large as a house, with its ‘face’ to the west, and with a spot on the north and on the south side where the arrow went through, and red streaks running down from them where the blood ran down to the ground.”

3.2

Maps of personal narratives

The personal stories recorded by Goddard (1911) chronicle important events in the lives of Casa Maria and Juan Pesita. Some of the narratives provide details of specific incidents in time occurring at a stationary, and often unknown, place on the landscape. “A Duel Between Scouts” (Goddard 1911:â•›137) occurs while on a hunt. “A Captive Woman Attempts to Make Peace” (Goddard 1911:â•›139) takes place while the group is camped. “Pesita is Shot” (Goddard 1911:â•›152) is a narrative that occurs almost entirely in the area of Cimarron, New Mexico. These stories occur on the physical landscape and provide valuable mental constructs for the individual chronicler and his listeners. However, unlike the narratives which unfold across the landscape, they do not appear to relate to developing spatial information but appear to belong to another category: social memory. To facilitate mapping of movement across the landscape, I have assigned the Goddard personal narratives to three main categories of story: hunting journeys, raiding parties, and military excursions; though hunting is a factor in most of the stories. Keen knowledge of the landscape and the resources allowed the group to take advantage of opportunities for fresh food whenever it arose, as Casa Maria is quick to inform his audience. Distinctions between individual narratives are

360 Elizabeth M. Lynch

clarified by the storytellers themselves. For example, hunting narratives express the explicit intent of going on a hunt for a specified game animal. Any excitement that arises along the way is incidental to the main theme, though often these incidents provide valuable moral lessons. The narrative “The Arrows Fail on the Hunt” (Goddard 1911:â•›154) takes place in the fall on the plains, east of the village of Cimarron, New Mexico. The story opens with the group already encamped somewhere along the Canadian River, preparing to hunt buffalo. The group travels east to, and camps at, three places, “Mountains Stand”, “Saddle Washed Away”, and Balisoye, before they find buffalo. They remain in the vicinity of Balisoye for two days in order to prepare for the hunt because of the sheer number of buffalo encountered. On the next day the group moves to another place Gadjaeyi and camps “below in the arroyo” (Goddard 1911:â•›154). Over the next three days the Jicarilla use so many of their arrows hunting and have so much meat that they decide, prudently, to head back to Cimarron before they encounter any “enemy” (Goddard 1911:â•›155). The story begins at an unknown locality so I place the starting point east of Cimarron, heading in a northeasterly direction on the illustration of the abstract map (Figure 1). This is speculative, as the direction the group travels is not made clear. The potential routes to the north or towards the east, from the village of Cimarron, are numerous. It is possible that the place called Mountains Stand refers to the real visual landmarks emerging as one travels east from Cimarron: Laughlin Peak, Tinaja Mountain and Eagle Tail Mesa. The group could have also

Figure 1.╇ Map of Story 65 “The Arrows Fail on the Hunt”. The story begins at camp  on the Canadian River heads northeast (or east). The band returns to Cimarron.  The journey home takes four days with full load of meat (Goddard 1911:â•›154).



Navigating regional landscapes with Jicarilla personal narrative 361

taken a route through, or around, the mountains north of the Canadian River, then onto the plains south of the Arkansas River. The place name “Mountains Stand” and “Saddle Washed Away” (variation: “Saddle Floats Away”) are repeated in the narratives “A Fight with the Enemy on the Arkansas River” and “The Apache Meets a Texan” (Goddard 1911) (see Table 1 and Table 2). What is important about the map in Figure 1 is that it reveals how Casa Maria’s narrative encodes relational points throughout a large landscape. Further, since the group found a healthy herd of buffalo within four days’ journey of their starting camp, they were in familiar territory, with known landmarks, though their “enemy” was not too far away (Goddard 1911:â•›155). The return journey to Cimarron took four days. No place names were specified in the narrative and at least part of the journey was conducted at night (Goddard 1911:â•›155). They were laden with so much meat from the hunt that they were forced to abandon their supply of flour. The manner of the return journey suggests stealth (Goddard 1911:â•›155). In the narrative “The Horses of the Ollero are Stolen” (Goddard 1911:â•›141) we are told of an Ollero group who, after receiving some horses from Maxwell (identified as Lucien B. Maxwell of the Maxwell Grant by Goddard 1911:â•›250), decide to hunt buffalo on the plains. The impression is that Casa Maria remained Table 1.╇ A list of place names from “A Fight with the Enemy on the Arkansas River” (Goddard 1911:â•›133) sī ma lōn ye

Cimarron

k’e na da za

They moved out

kōl tcī de ye

Canadian River

na he za

They camped

dzīl tc’īdjai ye

Hills stand

na da he za

They camped

Lī yeL des eL ī ye

Saddle floated away

na da he za

They camped

ga lī sō ye

Carriso

da kwe na da he za

There they camped

sī ma lōn se gō ye

Cimarron dry

na da he za

They camped

ac dle ka dn la ye

Five peaks

na da he za

They camped

sīma lōn se gō k’e gō n ’a ye

Cimarron dry out of na da he za the canyon

They camped

dzīl ya dn ’a ye

Mountain stands up

na da de za

They camped

ge gōL gai ye kū he nl On the plain water ka na lies

a na da he za

There they camped

dī ge gō n keL ye

Level place

ī yąn ne da des tse e na Buffalo had been killed

na bec dī hī

Arkansas River

sai daL gai ye

White sand

ka na da za

They camped by water

na bec dī hī

Arkansas River

xa da ned n de

They rode

tse īn tcīnc yĩ hī

Tseintcincyihi

362 Elizabeth M. Lynch

Table 2.╇ A list of place names and associated actions from “The Apache Meet a Texan” (Goddard 1911:â•›148) dzīL tc’it djai ye

Mountains stand

da kwe na he za

There they camped

dzīL n tsai ye

Mountains large

na he za

They camped

dzīL n n kel le ye

Mountain flat

na he za

They camped

ga dja e ye

Gadjaeye

na he za

They camped

kaL de ’ĩ ’a ye

Cedar it stands

n na he za

They camped

sī gō lō xō ye

Cigorojo

n na he za

They camped

tsĩ gaL lĩ kō sīL ka

Wild horses lake

kū ’et dĩye

Water was not

na da he za

They camped

bōn da ye

Bondaye

da kwe na da he za

There they camped

ba la lō lō

Balalolo

gō ye na ye

It is named

a wa a sōL

Agua Azul

gō ye na ye

It is named

kōL tcī de ye

Red River

ka na da za

They camped

kōL tcī de ye

Red River

bis n da ci

Up

lī dō bī lañ gō cī

El Rito Blanco

mī yō a gwa ye

Millo Agua

na he za

They camped

na dōs tse ’aL ‘ī ye

Pipes they make

na he za

They camped

lī yeL des e Lī ye

Saddle washed away

na he za

They camped

tse da des lī ye

water flows over stones

na he za

They camped

kōL tcī de ye

Canadian River

na he za

They camped

sī ma lōn ye

Cimarron

na da zez n da

They camped

at Cimarron and is relating a story told by the Ollero when they travelled through Llanero country on their way back to their home near El Rito, New Mexico. We are told they had a good hunt for two days but decided to try for more buffalo. An unknown enemy stole some of their horses on the third night. The ensuing chase, the failed attempt to retrieve their horses, and barely having enough horses to bring home their meat seems to impart a lesson for appropriate behavior. The map of “The Horses of the Ollero are Stolen” (Goddard 1911:â•›141) is illustrated in Figure 2. As with “The Arrows Fail on the Hunt” (Goddard 1911:â•›154), the direction is a linear one which takes the group onto the plains to a location where it seems the group expects to encounter buffalo herds. The map is drawn to reflect an easterly direction, but even Goddard was unsure of the location of the end point. “There are two streams named kōLtcīdehī, ‘Red River’, besides the Canadian and some days travel from it” (Goddard 1911:â•›141). The cognitive map of the story illustrates clearly that camps, or stops, were made at known places on



Navigating regional landscapes with Jicarilla personal narrative 363

Figure 2.╇ Map of “The Horses of the Olleros are Stolen”. The story begins  at Cimarron camp, then moves east (or northeast) over the Canadian River.  The band returns to Cimarron (Goddard 1911:â•›141).

the landscape. Since the story was related by a different group of hunters, Casa Maria’s certainty about some place names suggests shared landscape knowledge between the eastern and western Jicarilla bands. The final story “A Deer Hunt” (Goddard 1911:â•›161) represents a more complex set of movements across the landscape (Figure 3). In this story, Casa Maria is west of Taos, New Mexico, camping near the town of Coyote. For ease of representation, I divided the story into three circuits based on the seasonal stages depicted within his account. The first circuit is the longest both in terms of the action and the distance covered. Casa Maria sets out to hunt deer with a small group. They are very successful. We are not told how long this hunt lasts; however, the distance between Coyote and Ojo Caliente is around 53 miles by modern routes. We are only told of two “camps” (Goddard 1911:â•›162) before they move on to the next camp. The group sells the deer meat at the Pueblo of San Felipe before the final stop at Cuchilla. Casa Maria stays here for a four-day festival before heading back to Coyote. The map shows the directions between the specific towns. Ojo Caliente is a village west of Taos, New Mexico and north-northeast of Coyote. However, the locations of “Stone Light” (SeasdzōLeye), “Ute Lies his Head” (Yōdabītsīlaye), “Spotted Mountain” (dzīL Lī l’ī djī), and “Mountains Stand” (Dzīltcīdjaie) remain unknown.

364 Elizabeth M. Lynch

Figure 3.╇ Map of the narrative “A Deer Hunt” (Goddard 1911:â•›161). The story takes place in three circuits lasting almost a full year. The first circuit begins and ends near Coyote, NM. The second circuit begins near Coyote and ends at a winter camp somewhere close to Gallinas, NM. The final circuit begins at the winter camp and ends at Tierra Amarilla, NM.

The second circuit begins from Coyote, NM and ends somewhere near Gallinas, NM. Casa Maria hunts alone this time and is very successful, taking deer and elk (Goddard 1911:â•›163). This hunt takes place between three towns: Coyote, Gallinas and Tierra Amarilla. Coyote is about 13 miles from Gallinas and about 37 miles from Tierra Amarilla. Casa Maria’s wife distributes some of the meat to relatives. They finally make camp near Gallinas around “Cebolla Canyon” (Cī gī ya he e gōn`ã e) where they find more deer. At another place near Gallinas, “Stone Flat” (Tse kel), they construct a large tipi and settle in for the winter (Goddard 1911:â•›163). The final circuit begins at the winter camp of “Stone Flat” (Tse kel). Casa Maria and his group are visited by Navajos, with whom they share meat (Goddard 1911:â•›163). The camp is moved to Tierra Amarilla in the spring. He is so successful hunting there that he is able to share with others. Again, his wife distributes the meat (Goddard 1911:â•›164). The story ends in Tierra Amarilla.



Navigating regional landscapes with Jicarilla personal narrative 365

These stories work well to illustrate Jicarilla narratives in spatial form. The stories and the narratives related by Casa Maria clearly indicate that hunting pathways in the Mountain Valleys followed circuitous annual routes in the ecocorridors between the Pueblos. Likewise, Casa Maria’s narratives describe a specific linear or elongated hunting strategy on the plains. Game varies from antelope, deer, and small game in the mountain valleys to buffalo on the plains. The Llanero route rarely intersects the villages of others, and camping stealth is practiced when moving through a landscape of “enemy”. West of the Rio Grande the Jicarilla camp nearby to, and interact easily with, the Pueblos, the Navajo and each other. The narratives also reveal a frame of reference for the Jicarilla that creates a meaningful existence, a significance of the Jicarilla being, incorporated within the landscape itself. The sacred places where Culture Hero killed the monsters hold real significance for the Jicarilla people, becoming transformed from a mere physical feature to a place with soul (Bachelard 1969). The remains of “Rock that Runs” (Tsĕ nanlki’ň), who was killed by Culture Hero, could be seen west of Taos. The presence becomes permanent and through the narrative serves to provide a mental template of Jicarilla place. Casa Maria’s narrative helps maintain the place within the memory of the Jicarilla tradition. The inclusion of young Jicarilla in the hunting experience and accompanying narratives ensured that the next generation had the cognitive toolkit to find game, secure it and return safely home. Trail systems were encoded in the stories passed from one generation to the next, as was the knowledge of good hunting areas and important landmarks. These narratives also embedded the knowledge of the whereabouts of potentially dangerous foes or advantageous allies. The stories also map or encode the changes that accompanied the shrinking home range of the Jicarilla (Tiller 2000).

4.

Conclusions

The mental mapping of Jicarilla regional landscape through narratives is a powerful tool to explain movement on the landscape of this region. The knowledge necessary to travel and hunt successfully within a large territory is provided by the origin stories and narratives which link actual paths to places of experiential significance in the environment. It seems clear that the strongly held beliefs of this closely bound group provided an overarching schema of landscape which was passed on to future generations. Though the stories discussed above were told by Casa Maria, many of his stories involve large groups of people. This raises the question of whether and

366 Elizabeth M. Lynch

how others may have experienced the landscape around them differently than the storyteller. They obviously saw the same landmarks, experienced the same rivers, helped with the hunt or provided support to the hunters, but would their individual cognitive maps have differed? It is fascinating to wonder what the other voices would add to the map. For instance, women would probably have had intimate knowledge of the areas immediately surrounding the camps (Tiller 2000:â•›28–29). The use of Jicarilla narratives to develop an understanding of cognitive mapping of landscapes for a particular group of historic peoples is a powerful tool, but is it applicable to the prehistoric past of this region? The present maps are not enough to answer that question but they are a point of embarkation to begin an exploration of indigenous landscape for the Southern plains. The Jicarilla occupied, for at least 200 years and probably longer, this vast expanse of land and adapted well to the main eco-zones: plains, canyons, and mountain valleys. They maintained internal cultural cohesion even while being influenced by close proximity to the Pueblos, sharing hunting corridors with the Navajo, developing close ties to the Ute and negotiating elbow room among the Spanish and Americans (Tiller 2002). To apply this method to earlier periods requires the development of a detailed and sensitive regional model, using other Jicarilla oral history and archaeological data and groundtruthing these places and trails with Jicarilla memory. Though the Jicarilla may possess the closest memory to this landscape, surrounding cultural groups such as the Pawnee, Comanche and Mescalero peoples may be able to provide overlapping spatial information. Our understanding of this region would benefit greatly from a study that incorporated modern Jicarilla stories and memories with a “groundtruthing” expedition of Casa Maria’s narrative. Building a comprehensive and overlapping indigenous knowledge of “place” in this region would greatly add to the archaeological understanding of the prehistoric landscape by providing an interpretative framework from which to view material remains.

Acknowledgements This chapter was inspired by my first reading of the stories Casa Maria related to P. E. Goddard, after I had worked extensively in the region of the Purgatoire River, Colorado and driven the modern route from Albuquerque though Coyote, Ojo Caliente, Abiquiu, Taos, Eagles Nest, and Cimarron, New Mexico. The landscape is powerful and Casa Maria’s stories more so. I could not have completed this chapter without the advice, comments, and encouragement of Mary Lou Larson (UW), Pamela Innes (UW), Michael Harkin (UW), and Scott



Navigating regional landscapes with Jicarilla personal narrative 367

Rushforth (NMSU), the support of my family and the other graduate students at the University of Wyoming. The Landscape and Language Workshop, Albuquerque, New Mexico was invaluable in providing support and focus for what could have been a dizzying array of ideas.

References Allen, Gary. 1999. Spatial abilities, cognitive maps, and wayfinding. In Spatial Information Theory: A Theoretical Basis for GIS [Lecture Notes in Computer Science 1661], Christian Freksa & David M. Mark (eds), 46–80. Berlin: Springer. Bachelard, Gaston. 1969. The Poetics of Space. Boston MA: Beacon Press. Basso, Keith H. 1996. Wisdom Sits in Places: Landscape and Language among the Western Apache. Albuquerque NM: University of New Mexico Press. Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge MA: Harvard University Press. Echo-Hawk, Roger C. 2000. Ancient history in the New World: integrating oral traditions and the archaeological record in deep time. American Antiquity 65(2): 267–290. Goddard, Pliny Earle. 1911. Jicarilla Apache Texts [Anthropological Paper of the American Museum of Natural History 8]. New York NY: The Trustees. Golledge, Reginald G. 1999. Human wayfinding and cognitive maps. In Wayfinding Behavior: Cognitive Mapping and Other Spatial Processes, Reginald G. Golledge (ed.), 5–45. Baltimore MD: The Johns Hopkins University Press. Golledge, Reginald G. 2003. Human wayfinding and cognitive maps. In Colonization of Unfamiliar Landscapes: The Archaeology of Adaptation, Marcy Rockman & James Steele (eds), 25–43. London: Routledge. Gunnerson, James H. 1969. Apache archaeology in Northeastern New Mexico. American Antiquity 34(1): 23–39. Gunnerson, James H. & Gunnerson, Delores A. 1971. Apachean culture: a study in unity and diversity. In Apachean Culture History and Ethnology, Keith H. Basso & Morris E. Opler (eds), 7–28. Tucson AZ: The University of Arizona Press. Hoijer, Harry. 1956. The chronology of the Athapaskan languages. International Journal of American Linguistics 22(4): 219–232. Hoijer, Harry. 1971. The position of the Apachean languages in the Athapaskan stock. In Apachean Culture History and Ethnology, Keith H. Basso & Morris E. Opler (eds), 3–6. Tuscon AZ: The University of Arizona Press. Huscher, Betty H. & Huscher, Harold A. 1942. Athapaskan migration via the intermontane region. American Antiquity 8(1): 80–88. Kitchen, Robert A. 1994. Cognitive maps: What are they and why do we study them? Journal of Environmental Psychology 14: 1–19. Laszlo, Errin, Artigiani, Robert, Coombs, Allan & Csányi, Vilmos. 1996. Changing Visions: Human Cognitive Maps: Past, Present and Future. Westport, Connecticut: Praeger Publications. Mooney, James. 1898. The Jicarilla genesis. American Anthropologist 11(7): 197–209.

368 Elizabeth M. Lynch

Nordhaus, Robert J. 1995. Tipi Rings: A Chronicle of the Jicarilla Apache Land Claim. Albuquerque NM: BowArrow Publishing Company. Opler, Morris Edward. 1936. A summary of Jicarilla Apache culture. American Anthropologist 38: 202–223. Opler, Morris Edward. 1961. Cultural evolution, southern Athapaskans, and chronology in theory. Southwestern Journal of Anthropology 17(1): 1–20. Opler, Morris Edward. 1994. Myths and Tales of the Jicarilla Apache Indians. Lincoln NE: University of Nebraska Press. Rodenwald, Amanda. 2003. The importance of land uses within the landscape matrix. Wildlife Society Bulletin 31(2): 586–592. Russell, Frank. 1898. Myths of the Jicarilla Apaches. The Journal of American Folklore 11(43): 253–271. Schroeder, Albert H. 1974. A Study of the Apache Indians. New York NY: Garland. Stanley, Francis. 1962. The Apaches of New Mexico 1540–1940. Pampa TX: Pampa Print Shop. Tiller, Veronica E. 1978. Jicarilla Apache. In Handbook of North American Indians, William C. Sturtevant (ed.), 440–461. Washington DC: Smithsonian Institution. Tiller, Veronica E. 2000. The Jicarilla Apache Tribe: A History. Revised edn. Alburquerque NM: BowArrow. Tuan, Yi-Fu. 1975. Images and mental maps. Annals of the Association of American Geographers 65(2): 205–213. Tuan, Yi-Fu. 1976. Literature, experience, and environmental knowing. In Environmental Knowing: Theories, Research, and Methods, Gary T. Moore & Reginald G. Golledge (eds), 260–272. Stroudsberg, Pennsylvania: Dowden, Hutchinson, & Ross, Inc. Tuan, Yi-Fu. 1977. Space and Place: The Perspective of Experience. Minneapolis MN: University of Minnesota Press.

chapter 18

Ontology of landscape in language Werner Kuhn

I make the case for ontology of landscape in language, addressing a series  of concerns that are hindering a broader take-up of ontology as a tool for intra- and cross-linguistic research. The bottom line of my argument is that ontologies, as formal specifications of vocabularies, address a core need of language studies and that the complications arising from different philosophical views on ontology are largely irrelevant for the practical task of studying landscape in language. I propose a view of ontologies as systems of constraints on interpretations of vocabularies, allowing language researchers to describe conceptualizations partially, but down to an arbitrarily fine level of detail. Foundational ontologies help to structure such specifications and to link them across languages and domains.

1.

Introduction

Language studies could benefit to a much larger degree from computational approaches to knowledge representation and reasoning than they currently do. Somewhat simplistic past attempts at formalizing semantics may have given linguists and anthropologists the impression that logic and algebra, together with the computational tools based on them, are inadequate to deal with the richness of natural languages and the distinctions these make. I argue here to the contrary that language studies need the formal, testable, and sharable models of language use that ontologies provide, and that current practice is far from reaching possible inherent limits of such formalisms. There are many ways to look at ontology (as a field) and ontologies (as an artifact), but they all share one trait: ontologies specify vocabularies. Tom Gruber, the author of the most often cited definition of an ontology as “an explicit specification of a conceptualization” (Gruber 1993:â•›08), summarizes the gist of ontology in a recent entry to an encyclopedia of database systems (Gruber 2008) as follows: “An ontology specifies a vocabulary with which to make assertions, which may be

370 Werner Kuhn

inputs or outputs of knowledge agents (such as a software program)”. Gruber’s audience, in this case, is the database community, but his point is general enough to apply to any endeavor studying “assertions”, including those in natural languages, by and for “knowledge agents”, including human beings. Note that the engineering idea of a specification implies incompleteness. Ontologies, thus, are partial representations of conceptualizations, constructed for particular purposes. Beyond this core trait of ontologies to specify vocabularies, one can make various claims about the significance of such specifications. For example, some ontologists claim that the specifications capture universals, denote categories, or define what exists in reality. As it is often unclear what such claims mean, they tend to lead to raging debates, generating more heat than light. Yet, the claims and debates do not preclude a pragmatic program of using ontologies as open, shared repositories for vocabulary specifications. My motivation for such a program is that language studies, particularly those of lexica, need this kind of tool, whether their object is landscape or something else, and whether they are intra- or crosslinguistic. They need it primarily for accessible documentation, where they can benefit from the precision and openness coming with it. The links that ontologies provide across languages as well as to other domains (such as agriculture or hydrology, if landscape is the subject) are a second incentive to use them. Finally, language studies concerned with landscape are likely to benefit to an even larger degree from the use of ontologies than those concerned with other domains, because conceptualizations of landscape vary so widely across cultures and regions (Mark et al. 2010). Consider the finding that even closely related languages weigh criteria differently when distinguishing water bodies, with size having more influence on lake vs. pond in English than on lac vs. étang in French (Mark 1993). How can we make sense of this finding and how can we document and share it? Are there patterns across language families, possibly related to geographic or cultural patterns? I will try to demonstrate in this chapter that ontology is the method of choice to address such questions and many others. I attempt to explain the idea of ontology of landscape in language, in the hope that it will help linguists, anthropologists, and geographers to benefit more broadly from ontology as a tool for research, understanding, cooperation, and documentation. However, the chapter provides no introduction to practical ontology engineering, as there is a growing number of tutorials and online courses with this goal freely available on the World Wide Web. The next section introduces the basic idea of ontology as vocabulary specification and shows how it applies to studies of landscape in language. Section€3 introduces the notion of a foundational ontology, using the example of DOLCE, and argues why choosing a foundational level is essential for language studies. From the perspective of vocabulary specifications resting on a foundational ontology,



Ontology of landscape in language 371

Section 4 then addresses some concerns that may stand in the way of a broader take-up of ontology as a tool for language studies. In conclusion, Section€5 summarizes what ontology of landscape in language can and cannot achieve.

2.

Ontologies and language use

Defining ‘meanings’ as objects of scientific study remains an unsolved and probably unsolvable problem. As Hilary Putnam (Putnam 2001:â•›iii) said, “thinking of ‘meanings’ (or ‘contents’) as ‘theoretical entities’ – as scientific objects, objects which can be isolated and which can play an explanatory role in a scientific theory€– is a mistake”. Trying to determine how the world really is (Smith 2004), and using the results to define ‘meanings’ may not get us very far then. I have proposed, instead, to use ontologies pragmatically as systems of constraints on interpretations (Kuhn 2009). Ontology engineers produce ontologies as records of language interpretation, in the sense expressed by the Gruber quotes above. They recommend striving for minimal ontological commitments (Gruber 1993), rather than completeness, in their specifications. This minimalism follows directly from the idea of an ontology as expressing a conceptualization shared by some community. The more detail it states, the fewer will share the conceptualization. Ontology engineering, thus, needs to find an optimum solution between stating enough to serve a useful purpose (of understanding, documenting, and reasoning), but not too much, so that the whole targeted community remains included. Ontologies do not need to, nor should they be used to, promote any kind of intellectual imperialism. A Yindjibarndi landscape ontology may, thus, include a warlu (metaphysical giant serpent) in every yinda (permanent pool) (Mark et al. 2007). Conceptualizations and the ontologies expressing them can include beliefs that disagree with western (or any other) science. This may make them harder to crosslink, but truer to what people mean. It reduces the dangers of imperialism (because no ‘truth’ is presupposed to define meaning), but also avoids idealism (because there is an accepted external world, conceptualized from different perspectives). There is nothing about ontology that precludes modeling the “great deal of plasticity in how language models the earth” (Burenhult & Levinson 2008)€– in fact, this is what ontology is about! In this spirit of regarding ontologies as systems of constraints on interpretations, valid in some communities (but not in others), I have recently proposed a pragmatic view of concepts for the purpose of ontological specifications (Kuhn 2009). It treats meaning as a process to be constrained, rather than as an object to be defined. As the saying “words don’t mean, people do” expresses, it is people

372 Werner Kuhn

express

Star

perceive

refer to

Figure 1.╇ The semiotic triangle enriched by processes

who mean something when they use a word, rather than the words having ‘a meaning’ on their own. This process of meaning by using words can be captured in the following threefold view of concepts, enriching the classic semiotic triangle (Figure 1), and involving – words (symbols, expressions), – which express ideas (thoughts) and – are used to refer to reality. If one adopts this view of concepts, ontologies do not need to specify ‘the meaning’ of words, much less ‘the existence’ of something – they simply guide interpretations. They provide (human or machine) language users with a set of constraints on how to use or interpret words. Applying these constraints supports reasoning, while normally not removing all ambiguities. For example, when Navajo people talk about eminences in their territory, they appear to use words expressing and evoking ideas of sacred animate beings (McPherson 1992). When English speakers talk about the same eminences, they may use words like ‘mountain’ or ‘hill,’ expressing and evoking ideas of elevated features of an inanimate earth. An approach to ontology that tries to fix what a mountain or hill ‘is’ either misses one of the two ideas (and forces the other onto its users) or it produces two separate ontologies which will be hard to connect in cross-cultural studies, because they commit to different ‘truths.’ By contrast, using ontologies to constrain interpretations, one can specify on which side of the animate/inanimate distinction Navajo and English words fall and inform the other side about the use of a word.



Ontology of landscape in language 373

Note that “ontologies as vocabulary specifications” implies that the concepts specified need to be in the language that is studied, rather than in, for example, English (where English is not the subject of study). This is an obvious request for language studies, but gets often overlooked in other ontology work. One commonly finds ‘domain ontologies’ (for example, for hydrology or land use) produced for concepts in (one form of) English. If these are to serve as backbones of information infrastructures spanning multiple languages, it is likely that the choice of English (or any other single language) hides important cultural differences, such as those between different conceptualizations of water bodies.

3.

Foundational ontologies and grounding

Ontologies have been widely adopted for information systems applications, as evidenced, for example, by the successful conference series on Formal Ontology in Information Systems, with IOS Press proceedings published since 1998. One of the main motivations to produce ontologies for information systems is the ‘silo problem’, i.e., the fact that data in science and society get collected at rapidly growing speeds, but are often based on different or even incompatible views of the world. Even seemingly unproblematic features of the landscape, such as water bodies or roads, get modeled in a wide variety of ways by mapping agencies all over the world. Ontologies hold the promise of making such worldviews explicit and, eventually, of mediating and translating between them. However, just by themselves, ontologies only add another layer of modeling on top of data collections, which may end up fortifying the silos rather than bridging them. The saying, attributed to Enrico Fermi (in a different context, of course), that “We’re still confused, but on a higher level” unfortunately characterizes the situation of ontologies documenting data collections in today’s semantic web. A proven (and so far the only known) way out of this higher-level confusion are foundational ontologies. At some point, the specifications of conceptualizations need to be shown to share some common traits, otherwise it will be impossible to bridge between them. Foundational ontologies are attempts to provide such common traits, based on human understanding rather than on syntactic and statistical properties. While each existing foundational ontology takes a slightly different philosophical stance, they all have in common that they identify some basic distinctions that people make in conceptualizations. The benefits of these distinctions for conceptual clarity exceed the small philosophical differences by orders of magnitudes.

374 Werner Kuhn

For example, the DOLCE foundational ontology (Masolo et al. 2004) proposes the four basic categories of – – – –

endurants (which endure over some time: objects and features); perdurants (which happen in time: events, processes, and states); qualities (which can be perceived and inhere in endurants or perdurants); abstracts (which do not fit any of the three other categories).

By using such categories (and their sub-categories) as a basic vocabulary in an ontology, one can constrain concepts using these more primitive notions. In addition, the fundamental ontological relations of inherence (between qualities and their carriers) and of participation (of an endurant in a perdurant) capture essential aspects of conceptualizations. For example, one can decide to specify the word ‘mountain’ as referring to a feature (a part of the earth’s surface) or to a physical object, both participating in their own way in processes such as orogenesis or erosion. A word that the Navajo apply to higher eminences in the landscape can, instead, be specified as referring to an ‘agentive physical object’ engaging some (animate or inanimate) participants in processes of its own. All three concepts are sub-categories of DOLCE endurants, capturing the fundamental (and possibly universal) ontological distinctions of animate vs. inanimate, and independent (object) vs. dependent (feature) existence. DOLCE, like any other foundational ontology, does not prescribe or preclude any of these choices. Instead, it makes them explicit and lets the ontologist commit to one or several in each case. Yet, even specifications tied to such foundational ontologies cannot solve the deeper problem of symbol grounding (Harnad 1990). Ontologies constrain interpretations of symbols (words) by other symbols (e.g., those taken from DOLCE) and this recursion needs to end at some point by tying some symbols to something outside a symbol system. This is the only way for meaning to emerge, as even scientists acknowledge who have been committed to a strongly symbolic view of meaning: “If it were not for the existence of such nonconceptual processes, our concepts would not be grounded in experience and thus would not have the meaning that they do” (Pylyshyn 2007). What exactly this ‘something outside’ or these ‘nonconceptual processes’ are remains an open research question. We have recently proposed to use J. J. Gibson’s elements of a meaningful environment, namely substances, media, and surfaces (Scheider et al. 2009). By anchoring concepts in elementary spatial processes, such as stepwise motion, and their observable properties (direction and length), we have outlined a promising approach to ground concepts.



4.

Ontology of landscape in language 375

Some questions and answers about ontologies

The debates about how to think about ontologies (whether as defining existence or as constraining interpretations) and the hype around the semantic web, together with its letter soup of acronyms and its jargon, may have harmed the takeup of ontologies to document vocabularies and to reason with their semantics in the humanities and social sciences. This is a great pity, as a unique opportunity to advance language studies toward more formal, open, and shareable models of languages remains underexploited. In an attempt to lower the hurdles, I now address a few concerns that have been voiced explicitly or may underlie implicitly the reluctance with which ontologies are met in practice. I paraphrase some objections related to work on landscape in language, followed by an attempt to respond to them.

4.1

“Ontologies standardize vocabularies, but linguists are interested  in understanding the differences, not eliminating them”

This is a common misunderstanding of the role of ontologies, which gets propagated even through the ontology literature. While ontologies may sometimes have the effect of standardizing vocabularies, this is not their primary goal. To the contrary, they specify differences between vocabularies and allow for understanding and reasoning about them. Ontologies can define any particular conceptualization (i.e. the use of some word) to an arbitrary degree of detail, showing deviations from other conceptualizations. In the context of landscape and language studies, ontologies can specify landscape words encountered in some language. They constrain the interpretation of these words by stating relationships to other words in the same language (such as hyponymy and synonymy) as well as to words in other languages. This allows for arbitrarily precise (or loose) statements about sameness or differences of interpretations across languages. Thereby, specific translations, as for example from Yindjibarndi yinda (roughly: permanent water) to English water body, can be explicitly excluded (because they miss an aspect, such as the spiritual component) or admitted (because they may be considered good enough in some context of mapping water bodies).

4.2

“Ontologies use formal logic, which is inadequate  to capture meaning”

If we can describe in natural language how a word is used, we can also do it in logic (first or higher order). But a formalization in logic is less ambiguous and

376 Werner Kuhn

supports automated reasoning. There are many kinds of (formal) logics, with varying degrees of expressivity and corresponding computational properties. If ontologies are used for automated reasoning about concepts (for example, to determine that a mountain is considered a feature, rather than an object, in some particular conceptualization), expressiveness needs to be traded against computability. However, when the primary purposes are documenting and understanding, the level of expressivity can be chosen to match these requirements. The so-called description logics that are commonly used in the semantic web may limit too much what we can say, though this needs to be assessed in each case. If we add capabilities (like those to refer to individuals or to declare numbers and functions), a lot more can be stated, and if we want to have real expressive power, higher order logics and algebra can provide elegant means of stating ontological constraints (see, for example, Kuhn 2007). The possibilities of logic languages and tools to express constraints are almost unlimited and nobody is forced to restrict themselves to description logics or even to set-theoretic model theory, where interpretations can only be given as sets of values. At the same time, the low levels of expressiveness offered by description logics can already go a long way to record important distinctions in vocabularies and has the advantage of coming with semantic web technologies (such as OWL, the Web Ontology Language, and automated reasoners).

4.3

“Ontologies sound like a useful approach for some of my problems,  but where and how do I start applying them?”

This is indeed a serious concern which has not been addressed adequately by ontologists. There is a tendency to write about how ontologies should be designed, as I do here, while producing little in terms of actual ontology design results. Not only does this raise some concerns about the advanced theories and their validation, but it also fails to address user needs. A pull from application communities like those doing landscape in language studies may change that situation, both at the individual researcher level and at that of funding programs.

4.4

“Stating relations between words in a vocabulary is hard  and foundational ontologies like DOLCE are difficult to understand”

Indeed, ontological terminology is sometimes intimidating (consider terms like endurant and perdurant) or confusing (consider the different uses of universal in linguistics and philosophy). Yet, inter- and transdisciplinary efforts face such challenges all the time. Language researchers will rarely produce ontologies on



Ontology of landscape in language 377

their own. Sharing the work between domain (geography), language (linguistics), cultural (anthropology), and knowledge representation (ontology) experts is the approach that is more likely to produce sustainable results. With the recent rapid growth of the knowledge representation field, the number of ontology engineers and their qualifications to work on subtle topic areas is increasing. One of these qualifications is to translate between the often opaque formalisms of ontology and some common sense account of their logical implications. A promising approach is Ordnance Survey’s project of a structured natural language interface to OWL, the Web Ontology Language (Hart et al. 2008).

4.5

“Ontologies may describe intended interpretations –  but languages change”

This is certainly true and reminds us that all theoretical accounts are subject to change, both in reality and in understanding. Language studies and ontological formalizations of their results are no exception. Ontologies bring this fact to light; they are not its cause or amplifier. Being specific about something and explicitly recognizing the potential for improvement and change is often more productive than avoiding to take a stance. Ontologies, as information artifacts produced in certain contexts, support being explicit about temporal and other constraints on interpretations.

5.

Conclusions

I have presented ontology as a tool of choice for studying conceptualizations of landscape and their expression in language. Linguists and anthropologists are too rarely found in projects designing ontologies of landscapes (or other geographic domains). At the same time, one could say that linguists studying the vocabulary of some language for some domain are the ontologists of the domain in that language. Given their often daunting task of recording, distinguishing, and relating, they would benefit from better tools for documentation and communication. In this chapter, I argued that such better tools are here to be used, while also pointing out their limitations. By taking a decidedly language-oriented view of ontologies and their roles (i.e., specifying conceptualizations expressed through vocabularies), I have tried to avoid the ill-defined separation of terminology on the one hand and objects or features on the other, which is too often attempted in the ontology and cognitive science literature. When landscapes are the subject of languages and ontologies,

378 Werner Kuhn

such a separation becomes very hard or impossible, as there are no obvious cuttings of the landscape into objects or features, which we could first define and then attach labels to them for different languages. My proposed view of ontologies as constraining interpretations intentionally takes a language-based view, where the ‘objects’ or ‘features’ in a landscape are carved out by references of words used in a particular language, rather than the other way around. Far from being relativist or even idealist, ontologies understood in this way need to be related to each other through a foundational level and grounded in experiences of reality outside symbol systems. Citing ontology work built on DOLCE and grounding work built on ecological psychology, I have indicated some promising directions of research to achieve these goals. The proposed understanding of ontology is also intended to limit the scope and aspiration to interpretations of vocabularies. No ontologists would claim that their work adequately captures ‘the culture’ of a community. There are meanings of ‘meaning’ that go far beyond what people mean when they use a word. For example, the ‘meaning’ of a mountain in a landscape may have a vast range of spiritual, historical, and other cultural dimensions. Some of these may be tied to linguistic meaning, others may not. Ontologies are not about ‘the meaning of life’ in any sense, and one should not expect from them more than a specification of concepts expressed in a language. At the same time, ontologists need to become more sensitive to the fact that these vaster aspects of meaning are often affecting the narrower, language-bound ones.

Acknowledgments Countless discussions with David Mark have contributed and sharpened many of the ideas in the chapter. His review and that of Andrew Turk helped to improve the chapter significantly. The work of my research group (http://musil. uni-muenster.de) keeps developing and testing some of the mentioned ideas. Funding from the U.S. National Science Foundation to participate in the Workshop on Language and Landscape is gratefully acknowledged.

References Burenhult, Niclas & Levinson, Stephen C. 2008. Language and landscape: a cross-linguistic perspective. Language Sciences 30(2–3): 135–150. Gruber, Thomas R. 1993. Toward principles for the design of ontologies used for knowledge sharing. International Journal of Human-Computer Studies 43(5–6): 907–928.



Ontology of landscape in language 379

Gruber, Thomas R. 2008. Ontology. In Encyclopedia of Database Systems, Ling Liu & M. Tamer Özsu (eds). Berlin: Springer. Harnad, Stevan. 1990. The symbol grounding problem. Physica D 42: 335–346. Hart, Glen, Johnson, Martina & Dolbear, Catherine. 2008. Rabbit: Developing a control natural language for authoring ontologies. In The Semantic Web: Research and Applications [Lecture Notes in Computer Science 5021], Sean Bechhofer, Manfred Hauswirth, Jörg Hoffmann & Manolis Koubarakis (eds), 348–360. Berlin: Springer. Kuhn, Werner. 2007. An image-schematic account of spatial categories. 8th Conference on Spatial Information Theory (COSIT) [Lecture Notes in Computer Science 4736], Stephan Winter, Matt Duckham, Lars Kulik & Benjamin Kuipers (eds), 152–168. Berlin: Springer. Kuhn, Werner. 2009. Semantic engineering. Research Trends in Geographic Information Science [Lecture Notes in Geoinformation and Cartography], Gerhard Navratil (ed.), 63–74. Berlin: Springer. Mark, David M. 1993. Toward a theoretical framework for geographic entity types. Spatial Information Theory: A Theoretical Basis for GIS [Lecture Notes in Computer Science 716], Andrew U. Frank & Irene Campari (eds), 270–283. Berlin: Springer. Mark, David M., Turk, Andrew G. & Stea, David. 2007. Progress on Yindjibarndi ethnophysiography. Proceedings of the 8th Conference on Spatial Information Theory (COSIT) [Lecture Notes in Computer Science 4736], Stephan Winter, Matt Duckham, Lars Kulik & Benjamin Kuipers (eds), 1–19. Berlin: Springer. Mark, David M., Turk, Andrew G. & Stea, David. 2010. Ethnophysiography of arid lands: categories for landscape features. In Landscape Ethnoecology, Concepts of Biotic Physical and Space, Leslie M. Johnson & Eugene S. Hunn (eds), 27–45. New York NY: Berghahn Books. Masolo, Claudio, Borgo, Stefano, Gangemi, Aldo, Guarino, Nicola & Oltramari, Alessandro. 2004. WonderWeb Deliverable D18. Trento, Italy: Laboratory for Applied Ontology (ISTCCNR). McPherson, Robert S. 1992. Sacred Land Sacred View: Navajo Perceptions of the Four Corners Region. Salt Lake City UT: Brigham Young University. Putnam, Hilary. 2001. Representation and Reality. Cambridge MA: MIT Press. Pylyshyn, Zenon W. 2007. Things and Places. How the Mind Connects with the World. Cambridge MA: The MIT Press. Scheider, Simon, Janowicz, Krzysztof & Kuhn, Werner. 2009. Grounding geographic categories in the meaningful environment. In Spatial Information Theory [Lecture Notes in Computer Science 5756], Kathleen S. Hornsby, Christophe Claramunt, Michel Denis & Gérart Ligozat (eds), 69–87. Berlin: Springer. Smith, Barry. 2004. Beyond concepts: Ontology as reality representation. In Formal Ontology in Information Systems, Achille C. Varzi & Laure Vieu (eds), 73–84. Amsterdam: IOS Press.

chapter 19

The role of geospatial technologies for integrating landscape in language Geographic Information Systems  and the Cree of northern Quebec Renée Sieber and Christopher Wellen

Advances in the semantic web allow indigenous peoples to seek a greater voice on the Internet that reflects how they structure their knowledge of landscape. Geographic Information Systems (GIS) – computerized mapping and spatial databases – is used to develop a culturally protected area for the Cree in northern Quebec, Canada. We first characterize GIS and then describe several GIS applications, many of which related to epistemologies and ontologies of the land. We focus on one specific application: geospatial ontologies for hydrography. We found that GIS could store and organize landscape related language and this storing and organizing allowed for mapping and modeling information. GIS was useful for ontology representation, but its utility depended on the stage of ontology development.

1.

Introduction

Geographic Information Systems (GISs) – computerized mapping and database management of geospatial information – provide a platform to digitally analyze and display the landscape. Indigenous peoples have been using GIS since the early 1990s for purposes ranging from natural resources management to land claims. With recent advances in what is called the semantic web, indigenous peoples are seeking a voice on the Internet that reflects how they structure their knowledge of landscape. To achieve the geospatial semantic web, we emphasize one aspect of the science behind GIS, known as geographic information science, which is geospatial ontologies for categorizing indigenous ways of knowing. GISs are contextualized in this project in applications to develop a culturally protected area for the Cree in northern Quebec, Canada. To promote cultural

382 Renée Sieber and Christopher Wellen

protection, an interdisciplinary team created numerous GIS applications and a spatial database. Many of these related to epistemologies and ontologies of the land. These include a multimedia atlas of indigenous cultural properties and a geospatial ontology of hydrography, which geospatially arranges toponymically related stories. Findings speak to the opportunities and challenges of applying technology in terms of elicitation and evolution of language, haziness of geospatial categories and visualization of knowledge. This research provides a benchmark for appropriateness of implementing GIS for integrating landscape and€language. The chapter begins with a brief description of GIS and how they offer more than just software or hardware tools. We briefly consider the existing use of these technologies by indigenous peoples and its critique. Then we will present in depth a particular application of applying GIS to landscape in language. This and other applications with the Cree illustrate the benefits and constraints of GIS.

2.

Use of GIS in landscape in language

A widely accepted definition holds that GIS is “a computer system capable of assembling, storing, manipulating, and displaying geographically referenced information, [that is,] data identified according to their locations” (US Geological Survey 2007). GIS incorporates data from a number of technologies, including Global Positioning Systems (GPS), and converges with other domains of knowledge such as remote sensing. Increasingly these tools are available on the Internet as digital earths (e.g., Google Earth). GIS is optimized for modeling the landscape. This predominant usage is because of the facility with which GIS organizes and manages data. GIS software divides geography into features and their attributes. Generally, this is done by organizing similar features, for example, all water bodies in a single database. Each of the features is stored as a structured set of x,y coordinates in separate files: for example, water bodies (a sequence of points joined by lines forming a closed area, or polygon), mountains (represented sometimes as a concentric set of discrete contour lines), or settlements (as individual points or polygons). These sets of coordinates can be augmented by descriptive data, called attributes, for example, the names of mountain peaks or their origin stories. These digital features can be overlaid geographically as

. Some landscape features are modeled as a continuous surface, for example, vegetation. Some can be modeled as continuous pixels: mountains can be modeled as lines or pixels.



The role of geospatial technologies for integrating landscape in language 383

desired. Relationships among these features also can be stored (e.g., movement down a slope or flow between sections of a river). At its base, this representation provides a very flexible way to do digital cartography. That is, GIS is a good method to make maps. It can be used to display the geographic extent of features, spatial distribution of features, topology (i.e., geographic relationships of adjacent features) and to show attribute values of different places such as total population, median income, or annual fish catch. Toponymic annotation constitutes a main application for GIS (Tobias 2009). One may map place names to explore why names are dense in some areas and sparse in others. To show topology, consider Basso’s 1996 recording of Western Apache place names like “Water Flows Down a Succession of Rocks”. GIS can visually relate database elements of stream, stream flow, and the geologic features. GIS can make maps but it is not a map, and mapmaking is not all it can do. GIS is a database, to which diverse information and media types can be attached (e.g., audio pronunciations of place names or features) (Elliot et al. 2008). Because it is a database and not a map, multiple versions of geolocated annotations and descriptions can be stored. For example, the database can store place names in different dialects or retain the changes in feature extents over time. The database allows for a flexible mapping tool. It also allows for analytic modeling, for example to calculate viewsheds (Brabyn & Mark, this volume), since landscapes are often viewed remotely and will look different dependent on perspective. The information stored in databases could be analyzed as well (e.g., geostatistical associations between land use and place or progressive erosion down slopes clearcut of trees). We have mentioned that geospatial concepts can be related. One important way to do this is through ontologies. Ontologies form the backbone of interoperability envisioned by the geospatial semantic web or Web 3.0 (Egenhofer 2002). The vision is to add natural language queries to the organization of the massive amounts of data on the web. In geospatial ontologies we look to categorize ways of knowing the landscape into sets of concepts, relationships, properties of those concepts and rules. Ontologies in philosophy refer to the essence of being or, more practically, the study of what exists. Computational scientists have adapted the philosophical idea of ontology as the way we digitally classify and relate data. In GIScience, language is about how people “cut” or subdivide features in geography. Take, for example, Smith and Mark’s (2003) questions about a mountain, which may be represented by a continuous surface of slope or of contour lines€– where is the feature called mountain in the database? Where does it begin and end? Is it a single feature or a set of related features (valley, foothills, mountain), and how are they related? This speaks to the potential need to move beyond modeling geospatial information as layers, as most concepts do not sit neatly in layers but possess

384 Renée Sieber and Christopher Wellen

a certain degree of vagueness. Hence GIS offers a role for landscape modeling that feeds into Web 3.0. For us, a GIS offers three approaches to integrating landscape and language. It is a mapping tool, and it is a spatial analytic modeling method. Lastly, it is a language unto itself producing a digital and abstracted landscape of the imagination that we can interrogate. To address this third approach, there is a small but vociferous literature that critiques the tool itself, from a social theory and ethics perspective (e.g., Schuurman 1999). It argues that we must appreciate the manner in which this technical language transforms natural language, in the same way that languages were transformed from oral traditions to written inscription, abstracting the physical landscape from how it is experienced to a set of x, y’s. We now examine how these three approaches can be employed for indigenous representations of the landscape.

3.

Use of GIS by indigenous peoples

GIS has been used extensively for and by indigenous people. The usage in indigenous communities tends towards land use planning and natural resources management (Bocco et al. 2001), for example siting of sewage treatment plants or management of minerals or forests. Emergent technologies such as Google Earth are being used to visualize threats (Butler 2006) or to advertise economic opportunities (Blangy 2006). We even see entries into geospatial software architectures, for example the Coeur d’Alene tribe’s use of web services to manage its geospatial resources (Zichichi 2010). However, GIS has a more normative goal with many indigenous peoples: to empower marginalized communities (Laituri 2002), influence state practices (Norwegian & Cizek 2004) and gain land rights (Poole 1995; Butler 2006). A common thread through this use of GIS is the geospatial representation of language. Through the spatial documentation of place names the knowledge of elders and hunters may be better protected (Tobias 2009). Many indigenous communities are in danger of losing their traditional knowledge as younger people lose their cultural roots (Laituri 2002) and, in the case of the Cree and other indigenous peoples in Canada, they become more sedentary and spend less time in the bush. Computational technology, and GIS specifically, has been identified as critical to sustaining cultural and linguistic diversity (Boast et al. 2007). As with standard GIS, significant GIS research vis-à-vis indigenous communities involves the use of geospatial ontologies and epistemologies (Mark & Turk 2003; Stea 2007; Wellen 2008; see also Turk, Mark & Stea (this volume) and Kuhn (this volume)). These authors illustrate the process of landscape language



The role of geospatial technologies for integrating landscape in language 385

elicitation, classification, and formalization to represent the underlying meaning of knowledge embedded in language about place. They demonstrate that we cannot take for granted a one-to-one dominant culture to indigenous culture translation of features such as water bodies (e.g., what classifies as a stream or a river). Concepts may or may not overlap; hierarchies about which subconcepts nest within higher order concepts are not necessarily universal (e.g., are ponds types of lakes in all languages?); and numerous spiritual or non-geophysical components may be required to complete our full understanding of place. The use of GIS to preserve indigenous culture is not absent criticisms. Aporta (2003) reports on how GPS has altered the way that Inuit navigate the sea ice. Much like the snowmobile before it and the radio before that, GPS is modifying the language of movement and contributes to a trend of people forgetting the meaning or names of landscape features, losing their ability to “read” the ice. Stewart, Jacobson and Draper (2008) caution indigenous communities about the resource requirements to implement a GIS, resources that they may find in short supply. Rundstrom (1995) is the most critical, comparing the transformation of indigenous place names into a form compatible with GIS technology to a form of “epistemic violence”. This includes truncating place names or adding generics (e.g., the words “river” or “lake”) to place names that already contain the generic in the indigenous name. This often conflicts with the strong desires of indigenous people, and the researchers who work with them, to use GIS to sustain ways of knowing the landscape, as opposed to merely creating a catalogue of indigenous knowledge in a written (or digital), static form. What we looked for in our project is not only the use of the technology for the preservation of language but a use that is sensitive to the cultural norms of the community and one that is participatory. In terms of language preservation, we looked at the use of GIS for maintaining traditional relationships on the land.

4.

Use of GIS to model landscape in language by the Cree

We worked with a Cree community in Northern Quebec, Canada. Cree is the largest indigenous language in Canada (Statistics Canada 2008). The community with which we worked is a fairly traditional one, where Cree is the primary language spoken by people of all ages in the community. The local dialect is Northern East Cree. It is a settled community, although over half of the adults still live part of the year on the land (i.e., in the bush), and nearly one quarter spend the equivalent of full time employment engaged in subsistence hunting activities (Berryman et al. 2004). On the land hunting occurs in family traplines, where hunting bosses (or tallymen) control access to the geography.

386 Renée Sieber and Christopher Wellen

Ours is part of a larger project to create a culturally protected terrestrial and marine area in the northern Cree territory. The project comprised a team of researchers, including anthropologists, philosophers, natural resources scientists and ecologists, as well as two geographers. To conduct this research, it was important to have a long-standing relationship with the community. The grant leader speaks Cree and has worked in the same community for 30 years. It also was essential to work collaboratively, so the Cree participated in the design and conduct of several of the research projects. We explored the use of GIS as a methodology to express the landscape of the Cree people of this specific town. Much of the GIS work was application building. We assisted them in converting data for their urban work. We made maps for the applications for the protected areas, which included GPS location of sites and trails. One group focused on creating a historical landscape through modeling changes in the coastline and conducting archaeological occupation mapping (Nielsen & Costopoulos 2005). Throughout the collaborative research process, we found it important to collect and georeference toponyms. A geospatial database was developed to store the names. Not only did it matter to the elders to see their place names on computer screens and output such as maps, it was important from a provenance standpoint. In Quebec only one official name is allowed per feature. French, English, and indigenous names compete for the same location or feature. Naming confers sovereignty and can lead to greater control over managing the land (e.g., the ability to determine who is allowed to hunt or extract minerals) (Nieminen 1998). As part of the larger project, we advocated for changing the official names of places. Achieving protected area status in part depended on Quebec accepting Cree toponyms as official names. Another way to show toponymy was a place-based story telling map (Stoncius 2007). Stoncius (2007) used Flash software to create a map (Flash is the underlying technology in Youtube). Places were identified by points, which were attached to multimedia. Each place on the map had a transcription of an associated story. Stoncius further illustrated the stories with static and animated images (the latter also using Flash). For some places she included the audio version of stories; for others she included audio pronunciations of specific words. So we can think of the use of GIS to identify places beyond the toponymy alone. We will focus on one application in greater depth: geospatial ontologies for hydrography. This example shows how GIS can store and organize landscaperelated language and how this storing and organizing can allow for mapping and modeling information. It also shows how the GIS can be interrogated as to its usefulness for the Cree.



5.

The role of geospatial technologies for integrating landscape in language 387

Examples of Cree language via GIS

We built a geospatial ontology of Cree language, focusing on hydrography (see Wellen 2008 for a complete description of the ontology project). Hydrography refers to the spatial distribution of surface waters (rivers, lakes, whether permanent or temporary, flowing or not). For us and for the Cree, the question was, could we utilize GIS to formalize their hydrographic knowledge (the words and concepts, and the place names) and give it a presence on the map and on the web? For our landscape in language projects, we conducted interviews with hunters and tallymen, and elders. To communicate with the elders and to validate the interpretation of the findings, we worked with a middle generation of bilingual people (Cree and English), which consisted of teachers, economic development specialists, Band council members, and GIS specialists. Secondary data collection relied on a wealth of existing stories on audiotape, lexicons and critiques (e.g., Denton 2006). Geospatial ontology development comprised a three-part process: conceptual, logical and implementation; and it involved GIS to varying degrees at each step. The conceptual most closely resembles the lexical phase, in that it elicits the underlying meaning of and relationships among words. Hydrographic words were put into the conceptual ontology, storing the underlying concepts and the relationships among concepts (Table 1). Table 2 shows how the concepts can be related geospatially; among these relations are emptiesInto and feedsFrom. Relationships can be typed: hierarchical (e.g., SwampLake is a subclassOf Lake), part-wise (e.g., bank is part of a river) and Table 1.╇ Lexical terms related to Cree hydrogeography (source Wellen 2008) Term

Definition

Shiipaashtikw Shiipiish Shikaapishii Siipii

A side channel of a river that breaks off and re-joins the main channel. A creek or a stream. One may be able to canoe down it. A very small creek. One cannot canoe down it. A large river that one can canoe down.

Table 2.╇ Relationships (source Wellen 2008) Cree term

Predominant relationships

Shiipaashtikw

– – – – –

Feeds from waterways. Empties into waterways. Connected to a waterway. Part of a waterway. Feeds from and empties into the same waterway.

388 Renée Sieber and Christopher Wellen

topological (e.g., emptiesInto). We also found aspatial relations that expressed social networks (e.g., told by elder). In an ontology, place names become instances (examples) of these concepts. We confirmed our understanding of the conceptual ontology with community residents, Cree hunters and trappers, and through participant observation of Cree as they lived and worked the land. At this stage GIS, in its basic definition, is not used. One cannot map a conceptual ontology, although the geospatial features and connections are explicit. This information with its generalized locations (e.g., channel and navigable stream) is stored in a standard non-spatial database. The same is done with the relations. This generalized data can be joined in a GIS to specific location data about places: their locations, names and particular attributes. After building a conceptual ontology, we proceeded to formalize it in a logical computer-readable language. Our logical ontology used the Web Ontology Language (OWL), which has emerged as a standard of the semantic web and formalizes ontological concepts. For example, aamaataamapiich ∃ connectsTo (siipii or shiipiish or shakaapishiish or shikapishii or shipaashitw …). ∃ means that for p(x) there is at least one of the following such that p(x) is true. Here aamaata‑ amapiich is the subject, connectsTo is the verb and the items in the parentheses are the objects. During this stage, GIS is also implicit because it still functions only at the instance level. We then implemented a geospatial application to model the Cree lands in and around the community. The ontology is used to relate features and show the stories associated with those features. For example, if one clicks on a specific stream, stories related to the stream and related to downstream features or part-wise components of the stream display on the map. Figures 1 and 2 show that GIS is used extensively at this stage. It is used to map the features and the geospatial engine is used in the background to draw out the topology of concepts. The hydrography project delivered several findings about conceptualizing landscape in language. The Cree ontology appears to be a flatter hierarchy than the Quebec government hierarchy, that is, it contained fewer nested levels and subclasses than the official ontology. We did not find upper level classes. We could have invented an all-encompassing class (waterbody) but we declined to impose it. We were sensitive to perceptions of acts of re-colonization in our handling of language. Additionally, mismatch in the number of classes as well as the content of classes complicates integration between the official and unofficial categorizations of concepts and illustrates the challenges of databases interoperability on the geospatial semantic Web. One designs ontologies so relationships occur among classes and not among instances of classes. We found that relationships could not be generalized for all instances in specific classes. Crees often pair their place names. These paired



The role of geospatial technologies for integrating landscape in language 389

Figure 1.╇ Cree Community Geospatial Ontology website (source: Wellen 2008). Cree Name

Cree Feature Type

English More Translation Information

Stories

Related Stories

LittleBrotherOf paakumisimuwaaw - Why Paakumshumwaau got its name. by Sinclair Matches About paakumshumwaaush and paakumshumwaau. by sinclair Matches paakamisimuwaas saakihiiikin

Little Old Factory lake

Lac McNab

Chiinistumchiaschuu story woth the waatinaakan. by Sinclair Matches

LittleBrotherOf paakumisimuwaaw - About Paakumshumwaaus and paakumshumwaau. by Sinclair Matches LittleBrotherOf paakumisimuwaaw - winter near old factory lake. by Sinclair Matches LittleBrotherOf paakumisimuwaaw - Atush at Old Factory by Sinclair Matches

Why Pakkumshumwaau got its name. by Sinclair Matches About paakumshumwaaush and paakumshumwaau. by Sinclair Matches paakamisimuwaas saakihiiikin

...

.

winter near old factory lake. by Sinclair Matches

BigBrother Of paakamisimuwaas - About paakumshumwaaush and pakkumshumwaau. by Sinclair Matches BigBrother Of paakamisimuwaas Chiinistumchiaschuu story with the waatinaakan. by Sinclair Matches

Atush at Old Factory by Sinclair Matches Describes how the river flows down to paakumshumwaau. by Sinclair Matches wiyaaschuunis

shiipiish

little... current

.

story with an atush. by Sinclair Matches

Little Brother Of wiyaaschuun Chiinistumchiaschu story with the waatinaakan. by Sinclair Matches

Atush at Old Factory by Sinclair Matches

Figure 2.╇ Clicking on a placename in Cree Community Geospatial Ontology website  displays Cree placename, feature type, English placename, and stories related to geolocation. Geospatial ontology brings up geolocated stories “downriver”, related through  the ontology.

390 Renée Sieber and Christopher Wellen

names occur in the same geographic category (e.g. siipii, a canoeable river), although one is considered the diminutive of the other. Paakumshumwaau is a large lake in the community. It has a smaller version called Paakumshumwaash. This relationship was classified by some as ‘Big Brother – Little Brother.’ It is possible to represent all variations from a class as subclasses (i.e., the brother subclass) of that class. However, there is a trade-off between completeness and generality. Is a concept or relationship common to the native band or specific to a family? One possible explanation is the trapline system, which might have isolated depictions of landscape to occupants of the trapline and out of which depictions might have diverged. Ideally, a concept/relationship would be created for the entire band but not the specific family. Working with relationships at the instance level forced us to “jury-rig” the ontology development platform to fit the language. There are likely many occasions in which the ontology platform, at the formalization and implementation stages, falls short. We also found several generational differences in defining concepts, with elders varying in their categorizations relative to middle-generation and younger Cree speakers. These variations are hardly unusual in a dynamic language. It is possible that the Cree terms about landscape taught in the grade school could contribute to the generational difference. Another possible reason is the residential school systems, which disrupted communication between parents and children. Also, Cree culture seems to prize individual learning and learning by experience (Preston 2002), likely contributing to a degree of ‘reinvention’ by each generation. GIS can store differences by generation with ease, although generating the geospatial ontology is a lengthy process that requires extensive time for elicitation and validation. There are challenges for using traditional GIS to model landscape. As mentioned above, GIS classifies data into similar features. Land features are stored in the same layer; water in another. These may be further divided into water linear features (e.g., rivers) in one layer and water polygons (e.g., lakes) in another. Because the Cree consider water an essential part of navigation, they categorize portages around river rapids as part of a single transportation network. A portage is land where you have to carry your canoe to avoid water hazards such as rapids. Table 1 shows that, for the Cree, water concepts are organized by their navigation potential. Newer versions of GIS may build these land-water linkages; the majority of legacy databases cannot. Ontology development is a challenging task. Cree language speakers were unaccustomed to thinking about meanings in their own language, particularly when they were asked to quantify those meanings. It was difficult to draw out concepts and relationships to realize the ontology. Cree respondents were



The role of geospatial technologies for integrating landscape in language 391

sometimes too polite to point out errors in the ontology. Concepts were in Cree but the relationships were not, and the ontology itself was not realized in the Cree language. This is unavoidable when building an ontology that will be shared among language€groups. Languages must be rendered sufficiently unambiguously so as to allow their formalization and then implementation as ontologies in software. Any ontology produces information loss because it reduces the richness of information. The concepts and relationships that are retained can achieve a fixedness in time and space. We cannot ignore that geographic concepts will change just as physical features change, evolve and adapt. Still Cree, especially Cree elders, loved seeing their place names on the computer as antidote to or countermapping of official names. The elders hope that the youth will be drawn to the technological version of their language. GIS provides a storehouse of their knowledge and interactivity to zoom and pan and identify specific place names and their connectivity across the landscape. Interrogation of the GIS as appropriate raises similar ethical questions to Rundstrom’s (1995): is it right to fit an indigenous language into a Western method? Conversely, to what extent are these representations universal human characterizations of landscape? What are the drawbacks of transforming an oral tradition into written text and then bytes? It is important to explicate assumptions when using geospatial technologies that “capture the cultural diversity of knowledge resources while still incorporating sufficient systematic information to enable effective retrieval” (Boast et al. 2007:â•›398).

6.

Conclusion

Our research speaks to the opportunities and challenges of applying GIS in terms of management, visualization and modeling of indigenous knowledge. We hope it provides a benchmark for appropriateness of implementing a GIS for integrating landscape and language. Our findings were about methods of representing the Cree language of landscape via GIS, among them elicitation and collection of language, formalization of concepts and visualization of information. GIS was useful for ontologies, but its utility varied dependent on the stage of ontology development. We are only at the beginning of finding ways to extend GIS, in practice as opposed to theory, with concepts, not just their properties but also the relationships among concepts. As the technology improves, GIS will offer a very productive method for comparing across classes and languages, and it holds opportunities for evolving the semantic web. Our research also points to the inverse: ontologies, through the formalization of semantics, will serve to evolve GIS.

392 Renée Sieber and Christopher Wellen

In addition to extending knowledge of Cree linguistics, our research was driven by normative goals: the creation of a terrestrial and marine culturally protected area and the furtherance of intergenerational transfer of knowledge. As with many indigenous communities, Cree worry that subsequent generations may lose their knowledge of the landscape. GIS can allow youth to digitally connect with aspects of their culture across time, to add new information and to socially network with others inside and outside the community. GIS offers a powerful tool but we must also reflect on its transformative prospects.

References Aporta, Claudio. 2003. New ways of mapping: Using GPS mapping software to plot place names and trails in Igloolik (Nunavut). Arctic 56(4): 321–327. Basso, Keith H. 1996. Wisdom Sits in Places: Landscape and Language Among the Western Apache. Albuquerque NM: University of New Mexico Press. Berryman, Nina, Burt, Jennifer, Fortier, Julie, Lotter, Elizabeth, Mackay, Briana, Paquin, Claude, Quock, Alanna, Stock, Meghan, Teitelbaum, Tamar & Wellen, Christopher. 2004. AaÂ�Wiishaautuwiihkw (Coming Together to Walk Together): Creating a Culturally Appropriate Protected Area in Paakumshumwaau (Old Factory) James Bay, Quebec: Year 2. Montreal: McGill University. Blangy, Sylvie. 2006. Le Guide des Destinations Indigènes. Montpellier: Indigène Editions. Boast, Robin, Bravo, Michael & Srinivasan, Ramesh. 2007. Return to Babel: Emergent diversity, digital resources, and local knowledge. The Information Society 23: 395–403. Bocco, Gerardo, Rosete, Fernando, Bettinger, Pete & Velazquez, Alejandro. 2001. Developing a GIS program in rural Mexico: community participation equals success. Journal of Forestry 99(6): 14–20. Butler, Rhett A. 2006. Amazon conservation team puts Indians on Google Earth to save the Amazon. Mongabay.  (31 May 2010). Denton, David. 2006. Cree culture resources – Cree placenames. (25 May 2010). Egenhofer, Max. 2002. Toward the semantic geospatial web. In The 10th ACM International Symposium on Advances in Geographic Information Systems (ACM-GIS), 1–4. New York NY: ACM. Elliot, Nancy, Hawley, Alex, Pokiak, Roslyn, Sherry, Erin & Koning, Aaron. 2008. Including aboriginal values in resource management through enhanced geospatial communication. In Proceedings of Spatial Knowledge and Information Canada Conference 2008, R.€E. Sieber (ed.), Vol. 1, 29–32. (20 February 2011). Laituri, Melinda. 2002. Ensuring access to GIS for marginal societies. In Community Participation and Geographic Information Systems, William Craig, Trevor Harris & Daniel Weiner (eds), 270–282. New York NY: Taylor and Francis.



The role of geospatial technologies for integrating landscape in language 393

Mark, David & Turk, Andrew. 2003. Landscape categories in Yindjibarndi: ontology, language, and environment. In Spatial Information Theory: Foundations of Geographic Information Science [Lecture Notes in Computer Science 2825], Werner Kuhn, Michael Worboys & Sabine Timpf (eds), 28–45. Berlin: Springer. Nielsen, Colin & Costopoulos, Andre. 2005. The impact of terrain severity on variation in viewshed generation: comparing Idrisi, ArcMap and GRASS. Archaeological Computing Newsletter 62: 17–26. Norwegian, Herb & Cizek, Petr. 2004. Using land use and occupancy mapping and GIS to establish a protected area network in the Deh Cho Territory. (23 May 2010). Nieminen, Anna. 1998. The Cultural Politics of Place Naming in Quebec: Toponymic Negotiation and Struggle in Aboriginal Territories. PhD dissertation, University of Ottawa. Poole, Peter. 1995. Indigenous peoples, mapping, and biodiversity conservation. Biodiversity Support Program Peoples and Forest Program Discussion Paper Series, Biodiversity Support Program. Preston, Richard J. 2002. Cree Narrative: Expressing the Personal Meanings of Events. Montréal & Kingston, Canada: McGill-Queen’s University Press. Rundstrom, Robert. 1995. GIS, indigenous peoples, and epistemological diversity. Cartography and Geographic Information Systems 22(1): 45–57. Schuurman, Nadine. 1999. Critical GIS: Theorizing an emerging discipline [Monograph 53]. Cartographica 36(4). Smith, Barry & Mark, David M. 2003. Do mountains exist? Towards an ontology of landforms. Environment and Planning B: Planning and Design 30(3): 411–427. Statistics Canada. 2008. Aboriginal peoples in Canada in 2006: Inuit, Métis and First Nations, 2006 census: findings. (15 June 2009). Ottawa, Canada: Government of Canada. Stea, David. 2007. Toward a theoretical basis for ethnophysiography. Paper presented at the Association of American Geographers, April 17, 2007, San Francisco CA. Stewart, Emma, Jacobson, Dan & Draper, Dianne. 2008. Public Participation Geographic Information Systems (PPGIS): Challenges of implementation in Churchill, Manitoba. The Canadian Geographer 52(3): 351–366. Stoncius, Jesslyn. 2007. Hypermapping and Narrative: Towards a Geovisualization of Wemindji Cree Stories. Undergraduate honors thesis, McGill University. Tobias, Terry N. 2009. Living Proof. The Essential Data-Collection Guide for Indigenous Useand-Occupancy and Map Surveys. Vancouver BC: Ecotrust Canada and Union of British Columbia Indian Chiefs. US Geological Survey. 2007. Geographic Information Systems. (15 Jan. 2009). Wellen, Christopher C. 2008. Ontologies of Cree Hydrography: Formalization and Realization. MA thesis, McGill University. Zichichi, Jessica. 2010. Geospatial Web services in government. Paper presented at the Association of American Geographers Annual Conference, April 14–18, 2010, in Washington, DC. See also .

chapter 20

Classifying landscape character Lars Brabyn and David M. Mark

This chapter discusses landscape character classification and provides an example of how Geographic Information Systems (GIS) can be used to produce landscape classifications. The first section examines the complexities of classifying landscapes; landscape has a range of meanings and can be conceptualized at a range of scales. The second section discusses the important role that landscape classification plays in communication and in determining how geographic space itself is conceptualized. The third section demonstrates how a landscape classification can be constructed with a GIS and uses the New Zealand Landscape Classification as an example. The last section reflects on how this classification can be improved through collaboration between landscape scientists and linguistic and cultural researchers.

1.

Introduction

Landscape often is conceptualized as being filled with, perhaps even composed of, features such as hills and valleys, rivers and lakes, each of which is delimited from its surroundings and classified into a category. But this is not necessarily how all people view landscape. An alternative approach would be to think of landscape as a continuum, filling space, and classified into types that indicate the character of regions of land. The perceptual nature of landscapes makes the formal classification of landscape character problematic, since there is no agreement about the appropriate levels of generalization of the classes and how these classes should be defined. Common language should provide the best basis for deriving appropriate classes of landscape within a GIS, but even within a single language such as English, landscape terminology may vary by a range of factors, including dialect, region, profession, activities, and social class. The two main conclusions of this chapter are that: (1) GIS provides the opportunity to build classifications, and there are an increasing number of new

396 Lars Brabyn and David M. Mark

classifications being produced; and (2) dialogue between geographical information scientists, linguists, landscape experts, and cultural specialists is important if GIS representations of landscape, considering cultural factors, are to be theoretically robust.

2.

What is landscape?

In this chapter, we define landscape as simply the appearance of the land and water from a distance. This understanding of landscape was conveyed by Granö in 1929 (translated in 1997), and it is consistent with everyday use of the term by the general public and with landscape’s association with scenery and aesthetics. This definition is also consistent with how many land use planners define landscape (Swaffield 1991): – Appleton (1980) defined landscape as “the environment perceived, especially visually perceived” (p. 14); – The Countryside Commission for Scotland (1970) defined it as “the spectacle presented by the countryside” (p. 2); and – Palka (1995) defines it as “the assemblage of human and natural phenomena contained within one’s field of view out-of-doors” (p. 71). Many values are derived from landscape. Landscape is an important tourism resource as well as contributing significantly to the quality of life of local residents. Not only can landscape have high aesthetic value but landscape is also associated with other important values such as heritage, biodiversity, sense of place, wilderness, science, and culture (Tuan 1974). Landscape, or “country,” appears to be even more interwoven with culture and language among indigenous people (cf.€Basso 1996). Although these values are derived from landscape, it is confusing to use them to define landscape. Landscape is not the biodiversity value but rather landscape includes the appearance of this biodiversity, as well as what else there may be in a view. In our usage, landscape consists of everything that can be seen in a distant view, which may consist of landform, water, landcover, and infrastructure. These are often referred to as landscape layers, but landscape is the combination of these components. Although it is necessary to break landscape down into layers

. We note that landscape has a rather different meaning in landscape ecology, where it appears to refer only to natural environments studied over large spatial extents, and that it is unfortunate that the same term is being used with different meanings in different research fields.



Classifying landscape character 397

in order to understand the complex nature of landscape, it is the combination of the layers that actually constitutes a landscape.

3.

Why classify landscapes?

“Classification is, perhaps, the basic procedure by which we impose some sort of order and coherence upon the vast inflow of information from the real world” (Harvey 1969:â•›326). Despite the importance of classification there has been very little acknowledgement of this importance in the geography literature since 1969. Harvey goes on to say, “Classification is, after all, basic to language since by every general name which we introduce we create a class. If language were restricted to proper names only, communication would become impossible” (p. 326). It appears that classification is so integrated into our thinking that people do not realize that they are doing it and need to be reminded of this fundamental aspect of human cognition (Langridge 1992). “We all classify objects, be they people, places, events or any other of a host of phenomena that enter our consciousness” (Shaw & Wheeler 1985:â•›255). Classification is important to academic knowledge because it provides a frame of reference that enables different researchers to communicate their results effectively. It also helps order and structure what is known (Haines-Young & Petch 1986). Given the range of values associated with landscape, it is not surprising that there is considerable communication about landscape and that people are searching for appropriate language to represent landscape. Landscape is part of everyday life, and is experienced positively and negatively through tourism and recreation. Professional planners are often engaged in contentious landscape debates regarding the construction of infrastructure, residential expansion, or forest clearing. Such debates require communication between different stakeholders, including local communities. Consultation with community interest groups is an important part of planning, and planners need methods for representing landscape character and values. Increasingly this consultation is assisted by GIS because maps and 3D visualizations are an important means of communication. Landscape assessment and decision-making are important applications of participatory GIS. A landscape character classification provides a frame of reference for communicating about landscapes as well as an inventory of the resource (Brabyn 1996, 1997, 2009). As an analogy, think of the benefits that plant classifications (scientific and common) have provided for communication about plants. Classification is important for communication because it provides a consistent frame of reference. All the major sciences work with classifications (plant classifications, soil classifications, chemical classifications, etc.) and it is difficult to see how these

398 Lars Brabyn and David M. Mark

disciplines could function without such classifications. Classification systems (taxonomies) are a core component of ontologies. When mapped, a classification system shows regions that are similar, thereby allowing inferences to be extrapolated from one region to another. In order to advance our theoretical understanding of landscapes and the management of landscapes, it is important to have a frame of reference for communicating about landscapes.

4.

Landscape classification

Given the fundamental role of classification in shaping how people construct knowledge, it is important to understand how classifications are constructed and the ways they vary. Kavouras and Kokla (2007) provide a useful summary of the fundamental drivers of classification (taxonomy) diversity. These drivers include the disciplinary training of the creators, the level of detail (granularity) required, the methods used, the cultural view being represented, and the cognitive diversity of individuals involved. The increasing use of GIS, and the ease by which classifications are being produced and/or represented using GIS, means that it is important that GIS professionals engage across disciplines and work with a range of perspectives and cultures. The classification of landscapes is complicated by the fact that it involves both human perception and physical reality, while many science classifications are based only on the physical reality. This sensual nature of landscapes makes classification particularly difficult because human perception generates a wide range of responses and meanings. Jones (1991) describes this complexity as the “elusive reality of landscape” (p. 229). He adds that, in the past, the lack of recognition that landscapes are both a physical reality and a social or cultural construct has led to an “academic battlefield,” with different disciplines and schools of thought frequently concentrating exclusively on either the physical landscapes or on the observer. Another important aspect of landscape classification is that it should be based on the perceptions of landscape held by the general public, rather than just the opinions of experts. This is because human perception is an important part of landscape. Classifying landscapes requires understanding of how people conceptualize landscapes – knowledge that can be gained through perception surveys and analysis of natural language associated with landscapes. The ultimate goal of the content category identification approach, which was popular in the 1970s and 1980s, was to eventually delineate a landscape classification system incorporating public perceptual assessments, rather than relying only on professional judgments of aspects such as scenic value (Amedeo et al. 1989). Content category identification studies generally determine attributes perceived



Classifying landscape character 399

as having negative impacts on quality, and those that have positive impacts. Such research typically involves using some form of statistical analysis, for example Q-sorting (Amedeo et al. 1989). For instance, preferences for different photographs of landscape are determined by asking samples of the public to rank them (see Kaplan & Kaplan 1989; Kaplan et al. 1989). Many attributes of the landscapes in the photographs can be identified and quantified, and then related to the preference for the photographs, using some form of regression analysis. For example, Pomeroy et al. (1989) used personal construct theory, the repertory grid, and a multidimensional scaling technique to ascertain these attributes. The Countryside Commission of Scotland (1988) reviewed studies involved with identifying those physical attributes of the landscape that determine quality. There are some attributes that are consistently identified as determining quality. Man-made structures have been identified in all the studies, especially as having a negative effect on perceived quality of landscape. This suggests that the degree of naturalness is an important component. Vegetation is identified in five of the six studies, and water features, such as sea, lakes and streams, have been identified in four out of the six studies. Landform is, surprisingly, only identified in two of the studies. Zube et al. (1982) identified relative relief, landuse diversity, water, and naturalness as determinants of quality in their review of this type of research. The aforementioned study by Amedeo et al. (1989) also identified aspects of vegetation, landuse, influence of water, and topography.

5.

The New Zealand Landscape Classification

The NZ Landscape Classification is an example of the type of landscape classification that can be produced using GIS (Brabyn 2009). It consists of four main components – landform, landcover, infrastructure, and water. The classification is built by combining (spatially overlaying) these landscape components. The unique combinations of the individual classes of the components provide the overall landscape classes. Landcover and infrastructure are the components of landscape that tend to have the most temporal variation, usually resulting from the clearing of forest or building new structures. Landform, water, and pre-existing landcover provide the context for this change. Tables 1–4 show the classes that have been used for these four components (Brabyn 2009). Each table has two levels of generalization, so that the user of the classification can determine the level of detail required. When the most detailed classes are used for each of the 4 components, there are 7,209 unique combination classes in New Zealand, whereas when the less detailed classes are used there are 264 classes. For localized applications, the user will most likely be interested in

400 Lars Brabyn and David M. Mark

a detailed classification, with perhaps only a subset of classes being instantiated within any particular local region. For national scale applications, too much detail may produce impractical classifications and maps. For most of the categories used in this classification, the information was already conceptualized as GIS data layers. For example, many of the infrastructure and water classes can be extracted from general purpose topographic data sets. As in most developed countries, landcover information in New Zealand is available as a GIS layer, and this information can be used in the landscape classification. The most detailed NZ landcover layer has 70 classes, while the NZ Landscape Classification uses only 20 classes. GIS was used to aggregate classes, and the 20 classes were based on major changes in the form of the vegetation and whether the plants are indigenous or exotic to New Zealand. A similar GIS-based landscapes map for the United States recently was published by Sayre et al. (2009). In the New Zealand case, landform was the most difficult layer to develop, because in NZ a suitable conceptualized landform layer did not already exist. Instead, this information was derived from 20 m elevation contours extracted from a topographic database. A GIS process was used to derive this information using terrain parameters such as height, slope, relative relief (height difference over a defined radius), and a profile index. This process was an adaptation of a method that was manually applied by Hammond (1954) in the U.S. Hammond’s method is based on explicit, quantitative procedures that use slope, relative relief, and profile to define different landforms. Dikau et al. (1991) developed automated processes that simulate Hammond’s manual method. Brabyn (1997) applied the algorithms developed by Dikau et al. to NZ and made improvements that identified open valleys and adapted Hammond’s landform categories for the New Zealand setting. A problem with the automated landform classification technique is quantifying the distinction among hills, high hills, mountains, and high mountains. This classification cut-off when using an automated GIS process is based on a radius from the highest point in a given area, because a neighborhood range function (using a moving circle of 3000 m radius) is used to calculate relative relief. This results in classification boundaries being located on the side of mountains or hills, and thus not corresponding to how people conceptualize the spatial distribution of these landform classes. Intuitively, a mountain should extend to a valley floor or in some cases to a nearby plain, unless there are substantial foothills. Even low saddles can be important natural boundaries that demarcate a hill from a mountain. The demarcation of the boundary between flat areas and hills or mountains is possible because a slope threshold can be used to identify this boundary. However, the demarcation between low undulating hills and flat areas is not that easy because of the accuracy of the elevation data, and the change in slope is sometimes too subtle.



Classifying landscape character 401

As can be seen from Tables 1–4, there are many different labels used for the different landscape components and levels of generalization. Even though they are just labels, careful consideration is required in choosing them because of the connotations these labels have for people. There is also overlap between the different Table 1.╇ Landform classes (Brabyn 2009) Landform detailed Mostly flat Low hill Low plateau Hill High hill Open valley with hill Plateau Mountain High mountain Very high mountain Open valley with mountain High plateau Very high plateau

Landform generalized Mostly flat or low hill Hill Open valley with hill Plateau Mountain High mountain Open valley with mountain High plateau

Table 2.╇ Landcover classes (Brabyn 2009) Landcover detailed

Landcover generalized

Urban Airport Mine or dump Coastal sand Horticulture High producing grassland Low producing grassland Exotic forest Exotic scrub Indigenous forest Indigenous scrub Permanent snow Alpine rock Tussock Sub-alpine scrub Freshwater wetland Saltwater wetland

Urban Airport Mine or dump Coastal sand Agriculture Exotic forest and scrub Indigenous forest and scrub Permanent snow Alpine Wetland

402 Lars Brabyn and David M. Mark

Table 3.╇ Water classes (Brabyn 2009) Water detailed Estuarine Enclosed sea Open ocean Small coastal island Large coastal island Large lake Medium size lake Small lake Island in lake River

Water generalized Enclosed sea Open ocean Coastal island Lake Island in lake River

Table 4.╇ Infrastructure classes (Brabyn 2009) Infrastructure detailed Transmission line Wind turbine Railway Highway Natural with sealed road Natural with unsealed road Natural with vehicle track Natural with mast Natural with overhead cable Natural with ski lift Marine farm

Infrastructure generalized Major utility

Minor utility

components. ‘Urban’ and ‘airport’ are listed as landcovers, but they could also be part of infrastructure. A GIS dataset of a landscape classification for New Zealand has many applications, which are described in Brabyn (2009). Landscapes that are similar can be identified, and the landscapes of regions can be described. The classification provides an inventory of landscape classes, allowing the area (hectars) of different classes to be calculated. This enables rare landscapes to be identified, and also allows to determine what landscapes make regions distinctive. A landscape classification combined with viewshed analysis provides opportunities for a range of landscape experience analysis, such as representing the landscape experience of a walking track. This analysis helps represent the values associated with landscape and provides tools for landscape planning.



6.

Classifying landscape character 403

The importance of language for developing landscape classification

The need to capture the general public’s perception of landscape classes within a classification distinguishes landscape classification from other physical geography classifications. The general public’s perception is particularly important because visual landscapes are perception-based and their essence is “in the eye of the beholder”. This provides a difficult challenge for visual landscape classification, as insight is required as to how the general public perceive and conceptualize landscapes. There are studies that identify landscape classes that are important to the public. This includes content category research which identifies landscape components that affect landscape quality, as discussed previously. There is also specific wilderness perception-based research that explores the concept of wilderness (Stankey & Schreyer 1987; Kliskey & Kearsley 1993). Also cognitive categorization research, which identifies categories that humans use to conceptualize space, is helpful (cf. Tversky & Hemenway 1983). Mark et al. (1999) summarized this research and also reported on experiments for identifying category norms for landform, landcover, water, human settlements and other human-made infrastructure. It is clear that basic common language terms, such as ‘hill’, ‘mountain’, ‘lake’ and ‘ocean’, are the categories that laypeople are using for conceptualizing many different landscape components. These classes are further substantiated within the field of “Naïve Geography”, which researches how non-experts (the general public) conceptualize spatial phenomena (Egenhofer & Mark 1995). The NZ Landscape Classification uses common language classes. Although words do vary in meaning between people and cultures, within a particular language (e.g. English) they nevertheless have a degree of common (shared) meaning that is sufficient to support communication of intended meanings in most circumstances. Even just using common language classes, the combination of different layers in the landscape produces over 7000 landscape classes. Thus there is more that could be learnt from common language that would provide insight into landscape perception and classification. Mark (1999) makes the point that studying both the structure and use of natural language for describing spatial situations is a good way to study how people think about such situations. Thus studies of landscape feature types in other languages and cultures could lay the groundwork for relevant landscape classifications for other nations and communities. Recently, Mark and Turk (2003) introduced the term ethnophysiography to refer to an ethnoscience of landscape and how people relate to landscape. For examples of such studies, see Mark and Turk (2003), Mark et al. (2007, 2010), the papers in the special issue of Language Sciences on landscape, edited by Burenhult (2008), and many of the chapters in this book.

404 Lars Brabyn and David M. Mark

The internet is now also an important source of insight into how people conceptualize landscapes. There is an almost exponential growth of the numbers of images in photo-sharing sites such as Flickr, Picasa, and Panoramio. Last year, Flickr passed the 4 billion mark in photographs uploaded (Wikipedia 2009a). More than three million of these Flickr photos have been geotagged or placed on maps by their users. The annotation or ‘tagging’ of photographs with text and other metadata, regarding both location and semantics of the content of the photograph, is now an important source for information on how landscapes are being conceptualized. In the future, as enhanced tools become available, there will be an increasing number of tagged photographs on the internet, which can be analyzed for landscape content. The formation of folksonomies is closely related to tagging. “A folksonomy is a system of classification derived from the practice and method of collaboratively creating and managing tags to annotate and categorize content” (Wikipedia 2009b). ‘Folksonomy’ is an emerging label that refers to a “bottom-up categorical structure development” (Guszlev & Lukács 2007:â•›194). Typically, texts and tags on the internet form the foundation of different folksonomies and provide an alternative to more institutionally supported taxonomies or “controlled vocabularies” (Guszlev & Lukács 2007:â•›194). Although internet tagging will produce variation in the use of semantic categories, the sheer volume of tagging submitted on the internet over a period of time results in the formation of “tag clouds”. Each tag cloud will have a central core of common meaning, but also contain peripheral meanings that are less common. Landscape folksonomies could play an important role in making links between tags and landscape image content (Veres 2006; Vander Wal 2007), and ultimately assist in identifying landscape categories that are meaningful to the general public and cultural subgroups (Edwardes & Purves 2007a, b). Given the socially constructed nature of landscape categories and the fact that language provides a useful insight into what categories are used, a landscape folksonomy and associated tag clouds may prove to be useful for developing more formal landscape classifications.

7.

Conclusions and implications

The intent of this chapter has been to emphasize the importance of researching language associated with landscape so that the classification of landscape is theoretically informed. Landscape classification is necessary to improve communication; however, landscape classification is different from most scientific classifications, which are generally based on the physical environment such as



Classifying landscape character 405

soil, plants, and geology. The perceptual nature of landscapes adds complexity and requires an understanding of how people conceptualize landscape. This chapter has briefly described the process used for developing the NZ Landscape Classification (Brabyn 2009). This classification has been developed using the most current data available and is based on the assumption that it should use common language landscape classes. From the tables provided for the different landscape attributes, it can be seen that there are many different landscape classes and terms involved. It has not been practical or possible to substantiate every one of these classes using research on category norms. More research on landscape and language is required. Classifications evolve with improved understandings and are never fixed. The classes used for classifying NZ’s landscapes will doubtlessly change over time. The perceptual nature of landscapes means that classes will evolve with changing perceptions, and these changing perceptions will be reflected in changes in landscape language. GIS has been proven to be very powerful for representing the environment through maps and data visualization techniques, including 3D views and the use of spatially located symbols and images. The use of computer graphics and maps is an influential form of communication which often has more persuasion than the written text. The ease by which GIS can be used by people is resulting in an increasing number of classifications being developed. What will be missing from many of these classifications is an underlying theoretical base. A particular challenge is providing substantive evidence associated with the perception of landscapes, as such perceptions are influenced by culture and context. It seems obvious that different cultures place different values on landscape in general and on its components. Holistic or analytical views of landscape are likely to vary across cultures. Ethnophysiographic research findings can support the adjustment of landscape character definitions and classifications to make them culturally sensitive (cf. Mark & Turk 2003; Mark et al. 2007, 2010). This could support mapping of landscape character for the same region of geographic space, perhaps even from the same data layers, but reflecting the values, interests, and languages of different cultures. Many people utilize more than one classification system for a single domain. With plants in New Zealand, there is the scientific classification, the common English names, and the Māori names. It is often the case that scientific classifications are useful not because they closely match language specific folk (common) classifications but because they provide a common framework by which to understand folk classification. For example, the question “Is a ‘whale’ a ‘fish’?” can only be articulated on the basis of a standard (e.g., Linnaean) classification system with which to frame it. An advantage of a universal classification is that it assists with the comparison across languages. A classification doesn’t make the languages the

406 Lars Brabyn and David M. Mark

same, as they will still have words which do not share common meanings. However, a universal classification does allow researchers to discuss the differences between languages, using a common frame of reference. Language provides insight into how people conceptualize landscapes. GIS researchers and practitioners, landscape linguists and cultural scholars need to work together to develop theoretically robust representations of landscape. This will be a challenge because there are differences among these researchers. GIS scientists combine theory with available data sets and analysis functions to produce operational definitions that are often a compromise. These will be premised by the belief that the benefits of a workable model may outweigh the issues associated with conceptual biases. Cultural scholars may tend to point out these issues in classifications and generalizations. Dialogue between these groups will improve understandings associated with representation issues such as conceptual scales (granularity), perceptual variation (fuzzy sets), and category norms (classes). Although these issues are prevalent in all classifications, the classification of landscapes brings these issues to a wider audience. This is because landscape is highly valued, multi-disciplinary, and is in the conscience of many people, including the general public in their everyday life. Landscape classification systems such as Brabyn’s (2009) provide useful input for landscape planning and environmental decisions. But they also provide a model of how people in some cultures might conceptualize landscape as continuous fields rather than as collections of features. As Mark and Turk (2003) noted: “Even … the propensity to transform a continuous landscape into objects … may vary across cultures” (pp. 30–31).

Acknowledgments We would like thank the University of Waikato and the U.S. National Science Foundation (grants BCS-0423075 and BCS-0753737) for supporting this research. We are also grateful to Andrew Turk and Gary Holton for their comments on an earlier version of this chapter.

References Amedeo, Douglas, Pitt, David G. & Zube, Ervin H. 1989. Landscape feature classification as a determinant of perceived scenic value. Landscape Journal 8(1): 36–50. Appleton, James. 1980. Landscape in the Arts and the Sciences. Hull: University of Hull.



Classifying landscape character 407

Basso, Keith H. 1996. Wisdom Sits in Places: Landscape and Language Among the Western Apache. Albuquerque NM: University of New Mexico Press. Brabyn, Lars K. 1996. Landscape classification using GIS and national digital databases. Landscape Research 21(3): 277–300. Brabyn, Lars K. 1997. Classification of macro landforms using GIS. ITC Journal 1997(1): 26– 40. Brabyn, Lars K. 2009. Classifying landscape character. Landscape Research 34(3): 299–321. Burenhult, Niclas (ed). 2008. Language and Landscape: Geographical Ontology in Cross-linguistic Perspective [Language Sciences 30(2/3), special issue)] Amsterdam: Elsevier. Countryside Commission for Scotland. 1970. A Planning Classification of Scottish Landscape Resources. Perth: Countryside Commission for Scotland. Countryside Commission for Scotland. 1988. A Review of Recent Practice and Research in Landscape Assessment. Perth: Countryside Commission for Scotland. Dikau, Richard, Brabb, Earl E. & Mark, Robert M. 1991. Landform Classification of New Mexico by Computer [open-file report 91-634]. U.S. Department of the Interior, U.S. Geological Survey. Edwardes, Alistair J. & Purves, Ross S. 2007a. Eliciting concepts of place for text-based image retrieval. In Proceedings of the Workshop on Geographic Information Retrieval (GIR), Ross Purves & Christopher Jones (eds), 15–17. New York NY: ACM Press. Edwardes, Alistair J. & Purves, Ross S. 2007b. A theoretical grounding for semantic descriptions of place. In Web and Wireless Geographical Information Systems, 7th International Symposium, J. Mark Ware & George E. Taylor (eds), 106–120. Berlin: Springer. Egenhofer, Max & Mark, David. 1995. Naive geography. In Spatial Information Theory: A Theoretical Basis for GIS [Lecture Notes in Computer Science 988], Andrew Frank & Werner Kuhn (eds), 1–16. Berlin: Springer. Granö, Johannes Gabriel. 1997. Pure geography. Edited by Olavi Granö and Ansi Paasi, translated by Malcolm Hicks. Baltimore MD: Johns Hopkins University Press. Guszlev, Antal & Lukács, Lilla. 2007. Folksonomy and landscape regions. In GI-Days 2007 – Young Researchers Forum [IfGIprints 30], Florian Probst & Carsten Keßler (eds), 193–197. Muenster: IfGIprints. Haines-Young, Roy H. & Petch, James R. 1986. Physical Geography: Its Nature and Methods. London: Paul Chapman. Hammond, Edwin H. 1954. Small scale continental landform maps. Annals of Association of American Geographers 44: 32–42. Harvey, David. 1969. Explanation in Geography. London: Edward Arnold. Jones, Michael. 1991. The elusive reality of landscape. Concepts and approaches in landscape research. Norsk Geografisk Tidsskrift 45: 229–244. Kaplan, Rachel & Kaplan, Stephen. 1989. The Experience of Nature. A Psychological Perspective. Cambridge: CUP. Kaplan, Rachel, Kaplan, Stephen & Brown, Terry. 1989. Environmental preference: a comparison of four domains of predictors. Environment and Behavior 21(5): 509–530. Kavouras, Marinos & Kokla, Margarita. 2007. Theories of Geographic Concepts: Ontological Approaches to Semantic Integration. New York NY: Taylor and Francis. Kliskey, Andrew D. & Kearsley, Geoffrey. 1993. Mapping multiple perceptions of wilderness in southern New Zealand. Applied Geography 13(3): 203–223.

408 Lars Brabyn and David M. Mark

Langridge, Derek W. 1992. Classification: Its Kinds, Systems, Elements and Application. London: Bowker Saur. Mark, David M. 1999. Spatial representation: a cognitive view. In Geographical Information Systems: Principles and Applications, Volume 1 (2nd edition), David J. Maguire, Michael F. Goodchild, David W. Rhind & Paul Longley (eds), 81–89. Harlow: Longman Scientific and Technical. Mark, David, Smith, Barry & Tversky, Barbara. 1999. Ontology and geographic objects: an empirical study of cognitive categorization. In Spatial Information Theory: A Theoretical Basis for GIS [Lecture Notes in Computer Science 1661], Christian Freksa & David M. Mark (eds), 283–298. Berlin: Springer. Mark, David M. & Turk, Andrew G. 2003. Landscape categories in Yindjibarndi: ontology, environment, and language. In Spatial Information Theory: Foundations of Geographic Information Science [Lecture Notes in Computer Science 2825], Werner Kuhn, Michael Worboys & Sabine Timpf (eds), 31–49. Berlin: Springer. Mark, David M., Turk, Andrew G. & Stea, David. 2007. Progress on Yindjibarndi ethnophysiography. In Proceedings of the 8th International Conference on Spatial Information Theory [Lecture Notes in Computer Science 4736], Stephan Winter, Matt Duckham, Lars Kulik & Benjamin Kuipers (eds), 1–19. Berlin: Springer. Mark, David M., Turk, Andrew G. & Stea, David. 2010. Ethnophysiography of arid lands: categories for landscape features. In Landscape Ethnoecology, Concepts of Biotic Physical and Space, Leslie M. Johnson & Eugene S. Hunn (eds), 27–45. New York NY: Berghahn Books. Palka, Eugene J. 1995. Coming to grips with the concept of landscape. Landscape Journal 14(1): 63–73. Pomeroy, J., Fitzgibbon, J. & Green, M. 1989. The use of personal construct theory in evaluating perceptions of landscape aesthetics. In Landscape Evaluation: Approaches and Applications [Western Geographical Series 25], Philip Dearden & Barry Sadler (eds), 151–176. Victoria BC: University of Victoria. Sayre, Roger, Comer, Patrick, Warner, Harumi & Cress, Jill. 2009. A new map of standardized terrestrial ecosystems of the conterminous United States [Professional Paper 1768]. U.S. Department of the Interior, U.S. Geological Survey. Shaw, Gareth & Wheeler, Dennis. 1985. Statistical Techniques in Geographical Analysis. New York NY: Wiley. Stankey, George H. & Schreyer, Richard. 1987. Attitudes towards wilderness and factors effecting visitor behaviour: a state-of-knowledge review. In Proceedings National Wilderness Conference: Issues, State-of-Knowledge, and Future Directions [Gen. Tech. Rep. INT-220], 246–293. Ogden UT: USDA Forest Service, Intermountain Research Station. Swaffield, Simon R. 1991. Roles and Meanings of ‘Landscape’. PhD dissertation, Lincoln University. Tuan, Yi-Fu. 1974. Topophilia: A Study of Environmental Perception, Attitudes, and Values. Englewood Cliffs NJ: Prentice Hall. Tversky, Barbara & Hemenway, Kathleen. 1983. Categories of environmental scenes. Cognitive Psychology 15: 121–149. Vander Wal, Thomas. 2007. Folksonomy: Coinage and definition. Vanderwal.net, 2 February 2007,



Classifying landscape character 409

Veres, Csaba. 2006. The language of folksonomies: What tags reveal about user classification. Natural Language Processing and Information Systems [Lecture Notes in Computer Science 3999], Christian Kop, Günter Fliedl, Heinrich C. Mayr & Elisabeth Métais (eds), 58–69. Berlin: Springer. Wikipedia. 2009a. Flickr, (19 October 2009). Wikipedia. 2009b. Folksonomy, (19 October 2009). Zube, Ervin, Sell, James & Taylor, Jonathan. 1982. Landscape perception: Research, application and theory. Landscape Planning 9: 1–33.

chapter 21

Perspectives on the ethical conduct of landscape in language research Andrew G. Turk and David M. Mark

This chapter provides a transcript of the panel session held late in the Landscape in Language Workshop to discuss ethical issues, especially in the context of research by, or with, Indigenous peoples. The session was chaired by David Stea and the panel members were Renee Louis, Carmelita Topaha, Andrew Turk and Renée Sieber. Eight other workshop participants also contributed to the discussion. The transcript of the panel session has been slightly edited to remove comments about the process and to improve the coherence and flow of the dialogue. No significant content was removed.

1.

Introduction

A panel discussion was organized at the Landscape in Language workshop in October 2008 (Mark et al. this volume) to foreground ethical issues relevant to landscape in language research, especially in cases focusing on Indigenous people, languages, and cultures. Indigenous and other participants raised ethical issues relevant to landscape in language research from both the point of view of Indigenous research participants/collaborators and as Indigenist researchers and research facilitators themselves. The panel session provided an opportunity for dialogue between these Indigenous people (researchers and others) and non-indigenous participants who explained how these issues have impacted on their research projects and the ways they have tried to deal with them. The issues raised in the Ethics Panel discussion are not necessarily specific to landscape in language research, and there is not much mention of hills, valleys . The term Indigenist refers to Indigenous scholarship which adopts such methodological approaches as conducting research in a culturally safe and respectful manner and adhering to three principles: resistance as an emancipatory imperative; political integrity in indigenous research; and privileging indigenous voices (Rigney 1997).

412 Andrew G. Turk and David M. Mark

and other landscape features; however, the issues discussed are motivated by the central role of ‘place’ in all societies. The research subject (landscape in language) is what makes these issues so important for participants/collaborators and researchers alike, especially the links to cultural and spiritual aspects of place. This chapter is not intended to provide an exhaustive or structured examination of all the ethical issues relevant to landscape in language research; this is attempted more fully elsewhere, including Turk et al. (in press). Rather, this discussion is a kind of coda to the preceding chapters that deal with specific research projects. It is intended to provide readers with a flavor of the type of dialogue that occurred at the Landscape in Language workshop. Discussion during this panel session confirmed the importance of landscape in language researchers being carefully attuned to the rights of participants/collaborators and language communities (Smith 1999; Turk & Trees 1999; Whiteley 1998). Rice (in press) provides a large number of references regarding ethical issues for linguistic fieldwork. Ethical issues that are especially important include: recognition of the validity of Indigenous as well as ‘Western’ scientific forms of knowledge; respect for Indigenous cultural and spiritual practices; obtaining proper permission for all fieldwork; making appropriate payments to participants/collaborators; ensuring that gender-sensitive issues are properly handled; allowing the maximum possible community control of project timing, direction and conclusion; not eliciting, recording, storing or publishing any secret/sacred information not directly relevant to the study; obtaining clearance from representatives of the language community for all publications; inclusion of culturally appropriate acknowledgment of participants/collaborators; providing feedback to the community about the project in a culturally appropriate manner; and ensuring that copies of all data and resulting publications are provided to the language community. This requires a respectful and reflective approach to research practice and regular discussion of ethical issues with colleagues and collaborators.

2.

The discussion

David Stea introduced the panel and the theme: ethical issues in landscape and language research, especially in the context of Indigenous peoples. He asked Renee Louis to speak first.

. Phillip Tuwaletstiwa kindly audio-recorded the session; initial transcripts were produced by Production Transcripts, Glendale, California.



Perspectives on the ethical conduct of landscape in language research 413

Renee Louis: In an article I wrote for Geographical Research I kind of started off this way, and for those of you who are not familiar with Hawaiian fiery attitudes, I apologize ahead of time. I am a Hawaiian by birth, a cartographer by training, and an academic by choice. I have been conditioned through colonial education systems to believe that Indigenous ways of understanding reality are considered subordinate to Western science and Indigenous ways of sharing knowledge, mostly through one-to-one oral or performative communication modes, are hearsay and inferior to the written texts that record a superior intelligence. Decolonizing an Indigenous mind is a beautiful thing and knowing that I’m not alone is a powerful aphrodisiac. All around the world, Indigenous aboriginal native peoples are responding to Western researchers’ needs to further develop Western scientific knowledge systems. Our voices may have started out as a little murmur from the margin but they have now become a distinct and unified cacophony of resistance and distrust. The doors previously opened for doing research on Indigenous communities in the name of science are closing and very soon these doors will be shut for good. “Why?” you ask, “We only had the best intentions for you. We’ve only tried to help you”. It’s because we’ve had enough of your conspicuous innocence. We’ve been pathologized by Western research methods that have found us deficient, either genetically inferior or culturally deviant, for generations. We’ve been dismembered, objectified, problematized via Western scientific rationality and reason. Then we have been politically, socially and economically dominated by colonial forces and marginalized through an armed struggle, biased legislation, and educational initiatives and policies that promote Western knowledge systems at the expense of our own. We know better now. Have I got your attention yet? I hope so because it’s really not my intent to preach here either. What I go on to say in the article I mentioned is that it’s really more about looking at various research methods around the world in various disciplines to find out the commonality about what it means to do Indigenous research. And I found that there are different approaches. Every place is different, really localized, but what I found was that there are some distinct similarities. And four of the distinct similarities that I found have to do with responsibility or relational accountability, respectful representation, reciprocal appropriation, and rights and regulations. There are people talking about these categories all over the world in various disciplines. But what I think I wanted to end my little introduction with is that from an Indigenous research perspective, the search for knowledge is considered a spiritual journey. It is in a Hawai’ian sense and I am generalizing from that.

414 Andrew G. Turk and David M. Mark

And in the article I mentioned, I do go into the concept of calling everybody Indigenous and what that means and how that plays out and works against us in some places; but I’m using it as a generalized term for different peoples around the world. In Indigenous epistemologies the greatest mysteries lie within; and this is why research, the search for knowledge, is considered a spiritual journey. And a lot of times the knowledge that we have within us is accessed through ceremony and ritual. The last thing I want to say is remember when you’re going out into these communities and you’re working with these people, with Indigenous people, that not everything they share with you is meant to be shared with everybody else, that sometimes that knowledge is for you and has something to do with your own evolution of learning and it isn’t necessarily something that can be, or should be, shared with all of us here. I have a few discussion points to make later but I want others to also have a chance to start this talking and I’m hoping this ends up being a discussion.

Carmelita Topaha: Good morning. Yá ’át’ééh, Diné. [Hello, people.] You need to know why you need to know. Because you’re researchers? You’re products of universities? You’re products of getting your Master’s? You’re getting your thesis or going for your dissertation? That’s part of the academic world, but that’s not our world. Your world comes from text, images, videos, photography, GIS mapping. Our world is all oral history. There is a great difference between the two. We have a lot of variability, and as I spoke about yesterday, one of the biggest issues for researchers, scientists, colleagues is that they think linear and I keep expressing this to people that I work with. They always say: the need to know. And I say to myself: [CT speaks Navajo and then translates] Why do you need to know? Thinking from a female point of view, I have male researchers working with me. I don’t know if that’s easy in the Western world. Do Western males work easier with male consultants? I have no idea, but coming from being a member of the Diné Nation, we have protocols, and I’m glad that Ron is here [Ron Maldonado of the Historic Preservation Office, Navajo Nation] and we have the IRB and all these departments that have these research approval steps that we have to go through. We didn’t have this, you know, in the beginning when research was coming into play. We have to look back at historical contexts, what the non-natives did to native people in the early historical years, and that still hurts, especially ‘The Long Walk.’ When you’re forced to do something you don’t want to do, it’s not by choice. Religious people came in, forced things on us we didn’t like, but we conformed. Boarding schools came in. We had to learn English, we had to conform. What



Perspectives on the ethical conduct of landscape in language research 415

else next? Taking the knowledge that has kept us very, very strong, very dignified€– what else, what else do you want? What else? There are so many books written on Navajo people; how we live, social behaviors, landscape. I understand now, but it took me a long time to understand why outsiders need to know. I banged my head many, many times and wondered why, you know, because of the influence of my grandfather and my grandmother, they always say, “Why do they want to know? Why? Because they don’t have a culture or because they don’t do this”. I kept thinking, “Well there’s got to be more to this. There’s got to be”. Well, simply, it is, you know, patrilineal versus matrilineal societies, the way of being, the way we think, the way our culture is intact with Earth Mother. We are intact with her. Every part of our landscape has some assemblage of body parts, of rivers, ceremonial practices, color, stone, shells, you name it; it’s all intact. We cannot separate just landscape, in terms of Western geography, as a dictionary, and just have that representing the Navajo world. We can’t do that. We are connected to the earth, and a lot of outside researchers do not understand. They want to put us into something like the GIS models, but you only can do so much with technology with certain aspects of landscape, certain attributes. I forget which gentleman mentioned [earlier in the workshop] “Where are we going to put the cultural part? How do you put that?” And I don’t know. I have no idea. The other thing I want to mention is Phillip Tuwaletstiwa. I’ve never met Phillip before and I’m very grateful to him. We always, I always, my family always had respect for the Hopi people, and other families, Navajo families did not respect the Hopi people, but we’ve talked, we’ve come a long way through our academic, you know, background, our fights with our own people, our own relatives, our own medicine people, and we’re struggling. We have struggled, but for a good cause. I struggle and I look at my past. He interviewed me last night, and I was thinking about – as I was speaking with him – “Gosh, that’s all in the past. Now I’m like moving forward, you know, and leaving all these things behind.” Thank goodness for the November 1990 Native American Graves Protection and Repatriation Act (NAGPRA). It was a good thing at the beginning, but what it did was it also made us fight with our neighbors. On the other side of the coin, I welcome research. I welcome it because if the anthropologists and geologists and psychologists, you name it, all the science fields, if they did not come into our reservation and document and take narratives and draw maps and record video, information would be lost. The early anthropologists, they came on the reservation and did this. If they did not do this, we wouldn’t have an overwhelming library of information, and I try to tell my people this, you know. If they didn’t do that, who would have? We already know our culture, so why would we want to go out there and record these things, you know? It’s

416 Andrew G. Turk and David M. Mark

just that our history is oral. It’s my job to tell the people orally, not with this thing here [she gestured to the microphone]. So that’s the huge difference between science and culture, and we clash and we’re still going to clash. I don’t want it to integrate, because once it integrates then someone has control, someone has territorial rights, someone wants to have that – how do you say that – that prestige, you know, and we don’t want that. We can integrate or combine or connect to a certain degree and then there has to be a cutoff. What I do is collect data in a way where it’s going to be preserved because when I go out, when I was doing a lot of archaeology on the reservation, I could see a lot of the young kids not speaking Navajo, not knowing the landscape, not knowing a lot of things and I thought “Well, you know what? By doing this project and other projects, at least I will contribute back”. It’s a very small, very small contribution back to my people and hopefully somebody else will continue it on but that’s the way our process works. You give information – you pass it on orally€– and it works a lot better when you do it in your own language. It works better because the people understand what you’re doing. And just a last thought, I’ve learned so much in this Landscape and Language workshop. When we were in Australia thinking about this workshop I didn’t realize it would come to be, with all of you experts in your own academic fields. As I told you the other day, I had no conception that this would ever happen but it did happen and I hope it happens again. I was just writing notes this morning and I thought, this is very, very unique, what was said throughout the workshop. Each one of you, with your own specialty and your research project or projects, would you change anything at all if we came back and met again? How would you do it differently? If you were to use your specific project or research and gear it to my reservation, what would be there that will please me and my people? But that’s just a thought. But I thank each and every one of you for coming to this workshop. I’ve learned a lot and you really, really have a long way to go because research never stops. Thank you.

Andrew Turk: I’d like to use a metaphor from my own culture – a metaphor from cricket – but that will be lost on almost all of you. So I will use a baseball metaphor because of the country we’re in. Batting for the team of research and seeking knowledge, my first problem is that I’m not Indigenous. That’s strike one. I’m male. That’s two strikes. And I’m an academic, so it’s three strikes. I could give up at that point. But one has to find some ways to at least put bat on ball, occasionally, even if sometimes it’s a field hit rather than a home run and maybe you’ll get a walk to first base if you’re lucky.



Perspectives on the ethical conduct of landscape in language research 417

So there are lots of detailed things we could talk about but I’d like to mention just some of the key things. For me, being non-indigenous, working through Indigenous collaborators is the obvious thing to do. Being male, working together with females is necessary, not only to account for my maleness, but also to assist in accessing different sorts of information. And as an academic it’s about overcoming that problem via relationships. I think this is difficult for young researchers in particular because there’s different levels of relationships. In some cases my research collaborator, who is my wife, and I have very deep relationships [with Indigenous collaborators]which require large commitments. In some cases, we’re on to the second generation of community people who spend some time living with us. And you can’t really expect every researcher to do that. So there are ways that work for some people but are not going to work for everybody. It’s about the depth of relationship. I guess it’s about being honest in those relationships and understanding the responsibilities that come with that, and being serious in your estimation of those responsibilities, and setting some limits for yourself as to what commitments you can make and communicating those limits. I could go to a funeral [in an Indigenous community] every week, but as an academic I can’t go every week thousands of kilometers to a funeral, so you have to put some restrictions on it. That’s just some introductory thoughts.

Renée Sieber: Okay, so I’m going to primarily speak from a GIS standpoint, not a language standpoint, and it’s just a series of questions and when I use the word “our” I’m speaking mostly for us non-indigenous people. So the first question is: what is our moral obligation to goals that are distant from our original goals? So for instance, one of the things that happened to us when we went into the community is they said “We want a museum.” Well a museum sidetracked us. I mean, it could be viewed as sidetracking us, but it actually turned out to be an incredibly rich experience – but it doesn’t fit the original plan. Our sense of moral obligation is the very normative sense and it doesn’t fit with a lot of research academies, and, certainly, working with Indigenous communities doesn’t fit into our academic time scales. That is a problem with a lot of new researchers and old researchers, or wise researchers, as well. You have to have a lot more patience and you can’t parachute in. You definitely can’t parachute in to help people because that’s just helping yourself. The second question is: how do we communicate technical knowledge? And I think that Phillip Tuwaletstiwa went a long way in talking about that. Technical knowledge is a third language in our discussion of English or other Western languages and Indigenous languages. Technology is an example of the additional languages that we have to think about because they are not transparent. Niclas

418 Andrew G. Turk and David M. Mark

Burenhult had mentioned early on that we haven’t fed back the information we collected because we don’t have language for that. The question would be: should there be a language of linguistics that is popularized in some way so it can be fed back to the community?

Niclas Burenhult: I think what I was referring to was that if you work from scratch on a language there is no consistent way to develop a writing system, which is part of the linguistic endeavor. It’s the first thing we do. Renée Sieber: Okay, so sorry for misrepresenting – but I do wonder, though, if there is a way. We also have to write these detailed languages of the disciplines and we get so unbelievably comfortable in our disciplines. One of the things you discover quite quickly when you work inter-, trans-, multidisciplinary, postdisciplinary is how disciplinarily bound you are, not only to your colleagues but with the groups that you work with. The third issue is: what do we do about the transformation or the inscription? We have a lot of assumptions about the transparency of the inscription, that it’s identical to the orality that we’re transcribing, but obviously, as Carmelita Topaha points out, they’re not the same thing, not the least of which is that our whole culture gets shaped when we’re oral as opposed to textual. So we think, “Oh, yeah, it’s just transcribing what people say” and it is not at all. And I was very struck by Carmelita, and I hope I don’t misrepresent her. She mentioned this phrase really early on [in the workshop] called ‘information depletion,’ which I thought was really fascinating because we work in this model, this economic model of information, that once you inscribe it it’s never depleted, right? You can share it multiple times – but in fact that may not be the case at all. And Phillip talked about ‘information leakage’. We work in a model, a very egalitarian model. That’s what Web 2.0 is all about. The more information sharing, the better, but you’re really talking about information haves. Like we have tons of information that we can build on already but this is not knowledge. I’m not saying there are knowledge ‘haves’ and ‘have nots’ but there are information or data ‘haves’ and ‘have nots’, so we take for granted that we can use these incremental bits. We use all this data. We sort of build on top and people can lose out on that in the sharing mechanism because the native layer, or the native database, becomes just one set of data points . This exchange refers in part to a much earlier conversation between Sieber and Burenhult about differences between basic and applied research with indigenous communities, and on appropriate feedback of results to the communities. Burenhult pointed out that linguist’s scientific representation of a language may not be of practical utility to the community.



Perspectives on the ethical conduct of landscape in language research 419

on top, instead of thinking about the knowledge behind this transformation of information. Lastly, how do we fit the technology to the Indigenous science and the Indigenous knowledge? And if we don’t, then what do we do? I’m really torn. Phillip talks about not always being participatory. I’m torn about that because I’m very cognizant of the need to get things done at the end of the day and perhaps you don’t always need to consult with the public. Certain framework data, such as the Federal Geographic Data Committee [in the United States] have these framework data layers, and one is geodetic control, and you don’t want or need participation, because if you have participation at all levels then things kind of grind to a halt. It’s the same thing as if you practice social deconstruction; you can deconstruct all the way down and just critique, critique, critique instead of being normative. But I don’t know the balance between that and I guess to some extent we have to, we non-indigenous people, have to create spaces for people, for Indigenous peoples to tell us when they need to get things done, and we have to step back from our critique. And then there’s the whole issue, which we could go on and on about, of building technological sustainability and capacity. Some of these [software systems] are not simple applications, and the next version of a browser comes along and everything falls to pieces. So building capacity is really tough, although I’m heartened by digital applications like Google Earth and Google Maps. It takes this deep learning curve out of it, but then our responsibility becomes, well okay the technology is easy but what about the geography? So we assume that, with the technological understanding, you have the geographic literacy of what buffers are, and what topology is, and that can get lost as well. I think I’ll leave it there.

David Stea: Thank you. We have lots of time for everybody’s contributions. If there’s anybody in the panel (or audience) that you specifically wish to respond to your contribution, then please feel free to call on that person. Phillip Tuwaletstiwa: Well in so many cases every situation is unique and consequently you have to customize what you’re going to do. I’m going to speak about Hopi and it’s going to be different than what you would experience elsewhere. And I want to discuss first Hopi and Navajo. We’re ‘recognized tribes’ so we have an overarching structure that interfaces between us and the outside world. And in our particular case, Ron Maldonado would be one person that interfaces for Navajo with the outside community when it comes to cultural issues and, of course, we Hopi have that same counterpart. On the other hand, for a ‘non-recognized tribe’, and I use that term in the strict legal sense, such as the Native Hawai’ians, that’s an unjust situation in

420 Andrew G. Turk and David M. Mark

my opinion, but when they have to interact with the outside world they have to interact on a community basis, as opposed to a government to an outside entity basis, so their situation is different. I don’t know how it works in New Guinea but my guess is there’s some overarching thing that you’ve got to do. Having said that, then today in this world, in this present time, Hopi has total control, as do the Navajo. Actually you never have total, but you have fairly complete control over what kind of study might take place on our reservation. Consequently, we have the responsibility, and no one else, for making the determination of whether this is in our interest, or not in our interest, to do. We do have the responsibility, as Ron [Maldonado] described the other day when we were sitting over in Window Rock [at the Navajo Nation headquarters]. He described the particular areas of interest that you’re going to have to have answers for if you are going to work on Navajo. We have the same thing. Our job then is, from a Hopi perspective, our job is to educate you, to describe to you, to make you aware and to work cooperatively with you so that you can properly form a request that we will find accessible. Now that’s highly idealized. I want to mention one other area. Up until the early 1990s when NAGPRA was passed, tribes were not empowered to protect what they saw as their cultural interest. In other words, if over at Chaco Canyon [which is not on the Hopi reservation] there was going to be a Federal Government action that might affect our interest at Hopi, we had no mechanism to insert ourselves in that decision-making process. After NAGPRA passed, suddenly throughout the south – I can only speak about the southwestern archaeological community – there was a monumental change because it prevented unilateral actions. Whether it was an academic institution or whatever it was, anything that used federal funding or on federal lands, if you did not then engage the participation of the tribal peoples, who suggested that they had a cultural interest, you were stopped and it would not happen. That changed everything in Indian country. It went completely different than it had ever been. Now, of course, NAGPRA has proven to be a two-edged sword, and I’m not going to dwell on it at length, but right now Hopi and Navajo are at odds with each other over cultural affiliation with some of the human remains found in Chaco Canyon. We’ll sort that out. That will get sorted out one of these days. It’s not without its problems but it really changed the whole situation with the Indian community. Last night when I was working with Carmelita she expressed to me some of her sadness and her disappointment in working with non-Navajo people when they were handling human remains; I won’t repeat her exact words but it was very sad, the attitude towards these human remains. This is something that had been in place since Wetherill began his excavations up in southeastern Utah at a ‘basket maker’ site. It continued, and in fact some of this attitude towards Native American



Perspectives on the ethical conduct of landscape in language research 421

remains was institutionalized within the southwestern academic system, and it was not a pretty sight. Today, though, it’s a better situation because we now are in the position to absolutely demand respect for our ancestors, and, through that NAGPRA process, remains that were in the Field Museum or in the Peabody Museum, all of these come back [to the tribes], or you have the opportunity to bring back these remains and associated grave goods, and it has raised the level of respect. And we will no longer tolerate blasphemy and casual, cavalier treatment of our remains. NAGPRA has done that for us and that’s that, so that’s all I have to say right now.

David Stea: Any other inputs? Anyone in the panel you’d like especially to hear from? It’s a self selection process Gunter Senft: I would like to put the whole discussion in the context of decisions that an individual experienced field researcher makes when he or she comes to another culture. I just speak from my experience, first doing field research in a factory in my hometown and now I’m doing field research in a village in the Pacific. Well, in the Pacific setting I didn’t know the language. In my factory, I knew exactly the same dialect. I was socialized in that dialect. But nevertheless they thought I came seeking a participant or servant. I didn’t. I was the one who was the servant and it was only after a week that the first people came standing by me at the machine talking with me, trying to integrate me into that kind of community, realizing that I really was interested in what they were doing. And that was exactly the same experience that I had then when I was on the Trobriand Islands in the Pacific. There I didn’t know the language but I knew that there were approaches by the people after they had observed me. I was not seeking a participant or servant. I was the one who served and it took us about six weeks until the people in Tauwema [a village on Kaile’una Island] decided, “Okay, they can stay. We build a house”. And that was the official sign of agreement: “Well you can stay. You have shown that you really seem to be interested. Now we will see how you work in the next five or six months and then you will see what we give you”. And I must say the data that I got, either from the workers of my hometown or from the Trobriand islanders with whom I worked, were things that I would never have gotten if they wouldn’t have been willing to share that opportunity with me. And this was not only a one-way thing when I got knowledge from the Trobriand Islands. They would always ask me, “Why are you doing it? Why do you want to know that? Why do you use these very strange signs to write down our language? This is not how you write English? What are you doing here?” I said, “Well

422 Andrew G. Turk and David M. Mark

I do not know how to come up with an orthography yet. I just have to know what kind of sounds you have and all the sounds have a specific meaning and later maybe I can come up with an orthography”. “Ah, that’s interesting”. And then you have one beautiful expression on the Trobriand Islands that can be used in all kinds of shades. You can use this expression and you’re really admiring what they’re stating. It means ‘white person, what he’s doing.’ And I don’t need to explain it. But that brings me to another point. It is so real what we do in the field as individual researchers. It is something that is a process that is based on communication, but it’s also based on trust. For example, I collected a lot of songs that are sung in a variety of ways. It’s called ‘The Songs of the Spirits of the Dead’ and they describe the life of the spirits of the dead in paradise. I always knew certain verses. There are always certain verses that belong to certain song cycles. And I collected about 200 or 300 of these verses and then I went back home and I wanted to make my wife a beautiful birthday present and I ordered these songs in a kind of poem and translated the poems from Kilivila into German. And then I went back to the Trobriand Islands and said to my consultants there, “Well I tried to order your verses that you gave me into a story, is that correct? Can I do that?” And they said, OK. That was some knowledge they were not ready then to give to me before€– that they really were stories that were told – and once they knew that I found it out, that I really showed enough interest in dealing with the songs, they told to me for the first time these song cycles as they should be sung. So this is also something where you just have to rely on the cooperation of the people you are working with. The question is, if I accomplish that, it can very well be the case that the missionaries who have destroyed quite a lot already may use these bits and pieces of Trobriand knowledge to really destroy the whole system, but then this is not my responsibility. I think what I’m responsible for is that I can transmit and codify and write down that kind of knowledge that is almost gone now on the Trobriand Islands, we just have five elders that know what they are singing. There are other people that sing the songs in the same way as, for example, German Catholics in a peasant village singing in Latin, and they didn’t even know what they meant, but they just knew that this was the ritual and they had to say it. It didn’t mean “let’s bend our knees and stand up again”. But I think the knowledge that is qualified in these beautiful song cycles is so important for the identity of the people because they know their forefathers believed in a very specific life cycle that’s completely different to what the missionaries, and what we, tell them. Just to give you one problem that they had to face while they were so-called Christianized. What the hell is the relationship between Jesus Christ, God, and Trobriand society? This is a matrilineal society in which a father is not related whatsoever with his son (or his daughter)!



Perspectives on the ethical conduct of landscape in language research 423

One other point that I would like to make is that one of the problems that Renée Sieber mentioned is writing down things. That’s a big problem because we are of course aware of this fact if we write down a beautiful story. We may write it down but we just give you one idea. We cut down a whole lot of levels of semiosis. If we do not tape-record it, if we do not video-record it, then we cannot give any kind of idea of the situational impact in which these stories are told, in the way of how people use gesture, how they use their body parts or they use a posture, how they use facial expressions, how they use prosody in order to come up with what they want to transform in such a story, for example. This is completely cut out. It is just dried up and we are aware of that fact, but it was only recently that we tried to do a better job because we now have the technology to do it.

Renee Louis: Can I just add to that last thing you just said, that it’s not only cutting out all of those other nuances but also, even in the videotaping of it, which includes all of the other things, you’re not seeing the greater picture of, okay, who’s in the audience? In Hawai’i, depending on who I’m talking to, I’ll give you a different story; the speaker is noticing the audience. It’s a cumulative knowledge learning thing for you. You just learned something a few weeks ago, the same thing. The closer you are to the person telling the story and their feeling that you’re ready now for this knowledge, you get it, which I think brings about a very different way of looking at knowledge. In a Western sense, you can go out, you can get your professional degree and you become the expert, where for Hawaiians not everybody gets to know all the knowledge. Fishermen have specific knowledge. Farmers have specific knowledge that they don’t necessarily share with each other, but a farmer has to know, “Oh this is a good time for this fish to run” so he knows to go down and grab his specific type of agricultural product to share with the fishermen. So he knows, “I want that fish and I know this guy likes this” so there’s some knowledge sharing, but there isn’t [sharing of] the specific knowledge. And someone in a hula clan doesn’t necessarily know the knowledge of the person who is in the feather gathering clan. (I’m using the term ‘clans’ even though we don’t have clans.) And so the knowledge is more specific and you don’t get to know it just because you want to know it, in our society, so it’s a very different way of looking at how knowledge is produced and shared. Renée Sieber: We know an awful lot about written stuff and there’s a certain exactitude about written stuff. For example, we put quote marks around things, so the exactitude matters. I don’t know about what exactitude means in orality because even when we record things we fix a story and we don’t know how the story can vary and still

424 Andrew G. Turk and David M. Mark

be the story. And there’s an analog with location too, because being in GIS, we’re very concerned about the accuracy of places and we don’t always have the notion of the fuzziness of places.

Bruce Janz: This is a session on ethics and that, of course, is one of the areas of philosophy that we’ve had a longstanding interest in. I just want to mention a couple of things that I think are relevant here. One of them is that the word ‘ethics’ comes from ethos, which is a term of specificity. It’s not in some sense a kind of generalizable term. And what that leads to is a kind of common distinction that’s made in ethics, which I think is relevant here, between theories of the right and theories of the good. The difference is the theories of the right are very much what you find in the Enlightenment in the West and after that; which is just a question of what is the right thing to do at this time. And so it’s very much about actions. How do I decide whether this action is the right thing to do, or not? So we have various ways of getting at that. We have Kant’s theory, which is all about beauty, and we have consequentialist ethics, which is all about what an action brings about. The reason I bring this up is that the word ‘ethics’ doesn’t translate very well into other cultures, or other languages, quite often. Because that kind of talk about ethics, that is, the notion of the theory of the right, doesn’t necessarily show up in cultural schemas that are outside of the West. Sometimes the theory of the good really is more like what we’re after. For that, you have to go to the West with respect to Aristotle, the notion that the question is “What is the good life?” What is the life that should be lived among these people here? And so the question is not what action should I do right now but what constitutes the proper way of living in a society. And for Aristotle the example is of the statesman, who is the person who not only knows the rules of the land but then also knows how to work the system, as it were, all the unwritten stuff, all the stuff about who to talk to now and how to phrase things and all of those little things that are required to bring the community to a wholeness or whatever. And so if you ask what you should do, well, it’s very context-specific. It’s very specific to being able to understand those particularities of the time, and that’s the kind of ethics that I find kind of resonates more with cultures, certainly those that I have witnessed in Africa, knowing the specificity of things like that. And so the reason I bring all of that up is that when we think about things like interests – I mean, ‘interest’ is a very consequentialist word, well interest for what? That’s almost utilitarianism, right? That’s putting things in an ethical framework that may itself actually be foreign to the framework of whatever group we’re dealing with.



Perspectives on the ethical conduct of landscape in language research 425

So I think there’s going to be an incongruity for asking about ethics within the research areas that we’re working with here, because we want to know what to do. We want to know the right thing to do at this location. Here I am faced with this issue. What do I do now? What you should do maybe is not answered by that question, “How do I work out a calculus or how do I figure out the best interest for who, and whose stake is in it?” and all those sorts of things. Maybe what we should be doing is in some sense understanding the right thing to do at the moment, which is much more getting into a kind of system of knowledge that requires more commitment, I suppose. I just wanted to make that distinction because the word ‘ethics’ gets used a bunch of different ways and it actually doesn’t work very well across cultural boundaries sometimes.

Renée Sieber: If I can just respond to that briefly – somebody’s written a paper on that, I don’t know if it’s published, with regards to GIS. That distinction is really hard for GIS people because it’s much easier to talk about the theories of right. Because part of what is ‘doing good’ is good science, so you can distance the moral obligation from what you do, and it’s very difficult to distance the normativeness when you’re talking about the theories of good and proper life. Andrew Turk: I think one of the problems is, particularly if you’ve been working with people for a long time, then you absorb these things, but how do you start? And I think the key to that is reflective practice. It’s about looking at the consequences of your actions and having a structure for reflective practice, particularly having the time for it. We’re always into busy work, particularly if you have field visits. You are flat out trying to get the data, but you should program time in the day for reflection. At the individual level, I found in early parts of my work that a reflective diary was extremely useful, even if I never read it again. But to sit down each evening and write some reflections about what were consequences of my actions, or even things I didn’t understand. Why was it this woman wouldn’t accept a lift to the shops? It was actually because they had assigned a ‘skin’ to me, a particular formal moiety; a kinship relationship. In that system I was ‘wrong way’ for her, which meant that, as people of opposite sexes and wrong ‘skin’ for each other, we couldn’t be alone together. So that reflective practice helped me to investigate and understand what was going on. So at an individual level diaries are a useful thing. I think what was also beneficial was that my research partner, Kathryn Trees, and I were driving long distances to the field site; two days drive. On the drive home we would discuss these ethical and practical issues with each other. So reflecting within the team is also important, and making time for that reflective activity. Also, one of

426 Andrew G. Turk and David M. Mark

my backgrounds is in psychology and I think that clinical psychologists have a really good approach to supervision. As a clinical psychologist you’re working with some client, but then you have to report back to a supervisor who is not only checking that you’re behaving appropriately to the client, but is also checking on your own well-being. So I think that that’s another important structural thing that we can put in place – particularly older researchers doing that for young researchers, at least initially in a formal way, which could become less formal over time. Those are some approaches I think which allow us to reflect upon good practice in a particular circumstance, rather than just having some prescriptions for the right things to do.

Brian Murton: There are many things that could be said but I’m just reflecting on how this resonates as someone who understands what happens in New Zealand in Māori research. Māori themselves have developed research protocols now. In Māori research, for example, there’s a sense that researchers really need to be concerned with the ‘goodness’ rather than the ‘rightness’ of their research, and that involves lots of things and there are probably a dozen aphorisms that can be used. The ‘seen face’ is one of the very important first steps, and most of the people here are ‘seen faces’ in their communities – and you probably know what I mean. That means that you need to be there, popping in and out. And of course respect, responsibility, the willingness to just sit down; and in New Zealand sitting down involves kitchen-table type discourse. Sitting down and having a cup of tea and just letting things flow. Not trying to force your way into things. It also means making sure you bring a few boxes of biscuits or something to share. Another thing that seems to be very important in working with the Māori research agenda is to go with the flow. You go where the community takes you. You don’t force yourself onto your agenda, but you go and you may go to places with people who you never thought you’d go with, and most of you are well aware of how that works. It can be formalized as networking or whatever. These are all things that have come out of the Māori experience of having been researched, and then the development of a group of Māori researchers in the universities. And it’s written up there quite articulately by people like Linda Smith. She’s only one of many that are involved in developing what is called Indigenous inquiry, Kaupapa Māori research. And you won’t get very far, even if you are Māori, if you don’t follow that Kaupapa Māori research. And that was another question I was going to ask of the group, and we heard a bit with Phillip this morning – the difference between going in as a total outsider, and coming in, coming back in, as, shall I say, a previous insider? Does that make a difference? That’s my question. I’ll leave it at that.



Perspectives on the ethical conduct of landscape in language research 427

Ron Maldonado: Well I’m glad you brought up respect because I think that’s often the way ethics is translated. I see respect as process-based, and that’s why we see in Native American communities a lot of protocols being developed which are fairly specific to their cultures. And I think that’s a good thing and the way it got generalized in NAGPRA has been a good thing – but in some ways the problem is it’s not specific to each culture, NAGPRA, and that’s often where the problem is. And the other point I wanted to make was regarding your question of: what are you going to do with the information? I think it’s useful to think about the distinction between the information and knowledge, and I think, you know, knowledge is a critical perspective or interpretation of information. I think that often the concern is that if you have the information, you might not have the knowledge, and then what are you going to do with that information? It might be to create a new knowledge, or a knowledge that’s contradictory to Indigenous knowledge, or maybe even disrespectful to it. And that tension is very hard to get around. I think it’s always going to be there at some level but it may be useful to think about that from the tribal perspective. Brian Murton: Just on that, can I just say one thing very briefly? I think there’s a huge difference between having information and knowledge. Isn’t developing knowledge another phase? There is a huge difference between knowledge and wisdom. And how knowledge should be used appropriately. And that’s something we should all be aware of when we’re trying to do work. Phillip Tuwaletstiwa: I want to comment on what Andrew said, and I want to emphasize it because I think he was leading into something which I would consider critically important when you’re acting not only with an outside culture, but with your own culture. He suggested that you write this journal, or this diary, or these notes and put down what you were thinking about and what you were doing on that day, and then be as candid as you can be. And I think you need to reread them, what was happening, and here’s why. I’ve noticed when people come to research Hopi, oftentimes the people that benefit the most from being there, and watch, say, a dance, I think are those people who know themselves already pretty well, who are already pretty centered within themselves. They can then absorb it. Sometimes a person comes there, and they’re like – they’re so needy they’re like an octopus or a squid with ten arms. And they’re just grabbing. And those are the ones I run away from, right? Because you can never fill them up, because they have not reached a point within themselves where they’re centered enough to be able to take information and process it in such a way that it’s meaningful. Wisdom, which I hope we all want – well

428 Andrew G. Turk and David M. Mark

I know everyone in this room would strive for wisdom. A young person can be wise, and I think most of you in this room are wise. But ten years from now with experience, you’ll be a wiser person, and for me wisdom is the increased ability to be able to anticipate the results of an action that you’re going to take now. If you can foresee the consequences of what you’re doing now, that is wisdom, and this would particularly apply to a leader in a village. I’m making a decision today. What’s this going to mean five years, ten years from now? And that really requires that self-examination of soul before you can begin to have that kind of experience; and the younger we are, the harder it is. I know I was a foolish young man, and probably I’m a foolish old man, but I’m not as foolish as I used to be.

Jim Kari: I’ve always described what I do as “language work” and that communicates with people who never heard of linguistics or anthropology. This conference is also about geography and that’s where most of my interest lies. I consider that to be the most important specialization of my language work. Many of the speakers of dialects, speakers of the last dialects, are gone now. And there’s a shortage of linguists. It’s harder than ever to do this now. I do my best to mentor young people. I try to encourage colleagues. I work with people in fisheries and archaeology. All aspects about spatial knowledge are wonderful and they’re underappreciated, and also another major theme is that most of the people I work with can barely read English and they don’t understand what I write. Only Susan Paskvan’s mother, who’s a great Indigenous linguist in Alaska, can read everything perfectly at that level. There are very few people like that in any of the Athabaskan languages. I’ve been retired for some years and I’m trying to back away from some of this stuff. I mean, it is so much about how to schedule, how to budget when things are so expensive, and practically nobody even does any fieldwork with the Athabaskan in Alaska and knows how to do it, and when money is tight, and I do know. Quite often I find myself on a long drive with someone and I’m giving them, you know, all the advice I can and who they should meet that they haven’t met and the Athabaskan speakers welcome this research. And there’s hardly anybody qualified to ask them the questions at their level, whether it’s in English or their own language. I’m very interested in the narrative. And I’ve worked with the key speakers and most of that stuff is still untranscribed. Just small percentages of it are transcribed. And I also don’t believe in publishing as much as getting things out in the grey literature or publishing and scanning copies of things, because that’s more costeffective now, and you get more done that way. Using little ephemeral reports and things, perhaps only 20 copies. You get them in libraries or recopy them and so on. Because publishing costs are another thing. You know, large expensive books



Perspectives on the ethical conduct of landscape in language research 429

that take quite a bit of time, and that the press loses money on, are hardly worth the trouble. So those are a few comments anyway.

Susan Paskvan: Well as somebody who studies our own language – you know, much like what most of you have, you develop relationships with your informants and what they tell you may be for you. But for me, as being somebody that studies my language, I have to ask them “Are you giving me this recording” – because people will bring me these old reel-to-reel tapes – “Are you giving it to me for my knowledge or do you want me to share it?” And oftentimes they want me to share it but they say “We don’t want it for commercial use,” which is a very big distinction. And so when I do share our songs – you know, because within our community, I put the songs on iTunes and then I share that with my fellow teachers, who share that with their students. But we always say that, right up front, it’s for educational use, make a big distinction about that. And you know, what they do tell me sometimes, because I am female and sometimes I can sense, you know, that what they’re sharing with me is something that’s just shared with the men, so I have to ask them specifically, “Okay, when you want me to share this, do you want me to share it with everybody or specifically with certain people? Are there any restrictions? Can young girls hear this story?” And a lot of times I can sense that and they’ll say “No, it’s not for that”. But I have to ask just to make sure and put a context on the knowledge, you know, like who is the audience. I’m the audience in that situation, but I’m also somebody that’s going to pass it on. Also, like what Jim was talking about, there aren’t enough people out there [documenting language], and I do feel that pressure, and a lot of people tell me what they study out there and then they say “When I pass on, I’m going to give you my material”. And that’s a heavy obligation – when you give material to an individual, versus giving it to, like, the Rasmuson Library in Fairbanks. They store material but only if they can share it. You know they have a great archive, but then I’ve had to go back into the archives and listen to some of the material when it’s only appropriate to listen to it. And so I’ve had to go back to them, to the people, and we couldn’t listen to it until it was the right time to listen to it. So we’d tell them, okay, this is material that’s restricted, and the only time that you can listen to it is, you know, before the ceremony. So when people, you know, like people from our villages (or even not from our villages) say they’re going to give me material, I have to think about whether they’re giving it to me as a safe place for it to be, and what context they want it to be used in. So there are different things to think about.

430 Andrew G. Turk and David M. Mark

Andrew Turk: One of the problems is understanding what the taboos are. For instance, in Australian Indigenous society there’s what’s sometimes called the ‘mortuary taboo,’ not showing pictures or using the names of people who have passed away. There are communities I’ve worked with where I can’t be called “Andrew” because there was another Andrew who passed away, so there has to be another name used for me. As a researcher you get a fix on this and you say “Right, that’s the taboo”, but as I’ve worked with Yindjibarndi people I’ve understood that there are many layers within this topic. It depends, for instance, on the particular circumstances of how the person passed away. Also, when did they pass away? What’s the circumstance of sharing the image? Local people will show images of people who passed away in particular circumstances, and so it’s hard for us outsiders to get all that right. For instance, do you use the names of project participants who have passed away? Because we are working with a local cultural organization, an answer is to follow what they say to do today, because that situation is changing all the time. So if they say to us “Yes, publish those people’s names” then that’s what we do, rather than trying to second-guess the complexities of the taboo. Lars Brabyn: I’ve got a specific question. I guess Phillip was talking about total control of information on the reservation; but we happen to live in an era where we’ve got access in the public domain to high resolution satellite imagery, you know, it might be one meter resolution scale, and I guess you could argue that the capturing of the information by the satellite is an infringement of rights. So, before we use these images, should we be asking the local people whether it’s okay that we use this information? Carmelita Topaha: I want to comment on Lars’ comment. You’re talking about the GIS based on aerial photography or satellite imagery. I was involved with a NASA project with the University of New Mexico years ago and it involved lots of money. We were looking for pebble mulch fields in gardens in Pueblo country and also Navajo country. And they hired myself and three Pueblo individuals. The project was starting in New Mexico and then going into Arizona and then Southern California. I thought, wow, this was a five-year project. Great, you know, research and travel and get to talk to people and look at the land and all these pluses. Until the technology process became clear. We looked at Jemez and Zia Pueblo and then there was one at Pojoaque. And of course we had to get permission from the governors. And the only reason why we were able to take photos and do research on their land was because we had asked first the Isleta Pueblo governor, a female governor, and she had a lot of influence. And she speaks highly for Pueblo people



Perspectives on the ethical conduct of landscape in language research 431

and their knowledge of keeping, you know, their culture intact. So we got the OK to start this project. But when we went to the other governors I asked them, “What do you not want us to do?” We had the written text about the project in all this technical jargon, technical language that maybe they understood or did not understand. So I said, “What photos can we use? What are we not to do?” One governor just took me aside. He says, “I want you not to take photos of the north”. I didn’t ask any questions because I already knew what they meant by “of the north”. So I told my colleagues, “We can’t take photos of the north. Whatever you do, please do not”. And I said we had told the governors everything we do, you know, papers, photographs, videotapes, whatever. They were all to get, you know, a copy. The University of New Mexico, being an EDAC (Earth Data Analysis Center), they have such wonderful, beautiful imagery of everything in the world. I mean, you can download those things – whatever – to look at people’s license plates. Whatever, I mean, that was cool. And so we took all these beautiful pictures up in Jemez Pueblo, and also I had five Native American students, researchers. Four of them were going into anthropology; one was going into business. One of the students was Jemez and the other one was Zia. So that also was very easy, to get into the Pueblo. So when we did our research up on the hills, we’re looking at Pueblo mulch fields and great gardens, it was just amazing with the knowledge they have on that, and so, well, we took photos and then we took it back to EDAC and we looked at it and of course we said specifically in our meetings, you know, “They don’t want nothing shot to the north”. Well the next day I go back and the guy, you know, the techie, we had our little website, and he had these photos on the computer, and I was astounded, like, “What are you doing?” He goes, “Man, these are great photos!”, and I said, “Well we can’t release these photos yet”. I said, “The project’s not over yet. We haven’t done this”. And he goes, “Who cares?” And I’m like, “Oh my gosh”. So I had to go back to my colleague and I told him what was going on. So the issue came up when we met with the Pueblo governors again and some of the elders. And this was very interesting, when you talk about GIS and aerial photography and imagery, and all the good stuff that we can see as we’re up in the air – so someone said, “I’m a very spiritual leader. I believe what’s above. Somebody goes with all these different satellites all around the earth, the Earth Mother, daily, and I’m standing there giving homage, paying homage to the people above or to the side or wherever they’re at. You are violating my spiritual process. I do not agree with it”. So I said, “Whoa, you know, this is very powerful stuff ”. But he had a right. So from where he was standing, from what his consciousness, his memory was saying, the photographs up there that were being taken were violating his process. So then we had to go back and rethink this again, and it had just gotten to confusion, complexity, so we had to put aside the old approach and follow their restrictions, and then we started again. But some of the

432 Andrew G. Turk and David M. Mark

things, like the technologies, you know, science, hard science, people don’t really understand that process. That’s all.

Ron Maldonado: First of all I want to say thank you for allowing me to sit in. I asked David Mark if I could be a fly on the wall and I think a bit more than a fly on the wall. But just kind of listening to everyone speak. My perspective is [that of] someone who can give you permission to come and do research on the reservation. And one of the questions that were asked is, what if you’re Navajo and you come to the reservation. You live on a reservation, you go to school, now you’re going to do a research project. Yes, you have to get a permit from us. It’s tribal code. You know, everybody does, whether you’re Navajo or not Navajo, you come on the reservation and you get a permit. Most of the discussion has been about: if I give you something what do you do with that? What do you do with that knowledge? And that’s kind of where I come into it. When you’re out there, you give me a proposal, you go do your research and you turn in your final product or your draft product to me and I read it. And I have to make a decision based on my knowledge of Navajo culture, based on my knowledge of the landscape, whatever is out there, and my colleagues that I can go ask questions to, and then I make that decision and say “Okay, yeah, go ahead and publish it or change this or change that”. And I can give you a couple of really quick examples. There was a gentleman who was doing research on Navajo traditional ceremonies and Western medicine and how knowledge combined the two. He wasn’t specifically talking about the ceremony itself. He was asking questions about what type of ceremony you could have. Do you have particular ceremony in your house? And he was in a very small community and he was using community people to input the data. And I told him when he was asking the questions, I could tell you who had that ceremony, because it’s such a small community. And it took me many hours of discussion on the phone with him to make him understand that. That in a small community, if I ask you “Have you had a ceremony? When did you have it?” and someone else is transcribing this data and putting it in the report, they’ll know it’s you because everybody knows what everybody else is doing in small communities. We all know that. So it was hard to explain to him. He finally got it and changed his questions and we allowed him to do the research. He needed to modify it just enough that they wouldn’t be able to identify people. The other one was a published book. It’s based on information from John Holiday and it came out a few years ago. They sent me a copy of the rough manuscript and they said “Read it”. And so I read it. John Holiday had also written a note to me personally. Before I read the letter that he wrote, I read the manuscript. And it was an interesting manuscript. There wasn’t a whole lot that we



Perspectives on the ethical conduct of landscape in language research 433

could consider as difficult. There was some stuff about ceremonies and we probably wouldn’t have allowed it to be published, but I finished reading it. I had my notes, I had my comments. I opened up Mr. Holiday’s letter and I read it and it said “I’m giving this man this information freely. I want him to know that. And I have given him permission to publish this. I ask that you respect my wishes”. And that was that. I told him to go ahead and do it. And that’s basically it. If you look at a lot of the early publications of the Navajo ceremonies, there’s always something in there from the Medicine Man who says, “I’m giving you this knowledge in the hopes that future generations will use it”. And in my mind that gives you permission to use that knowledge in any way. And that’s kind of a gift. As I mentioned earlier when I first spoke to you, I love to read. When someone hands me a book I tell them “Thank you for the knowledge” because they’re giving me a gift. They’re giving me something I don’t have. I learned something a long time ago. I came out to the reservation 33 years ago. I came out here to work on the reservation and I remember I was out recording a Navajo site and I wasn’t paying attention. I was recording information when this very elderly Navajo lady snuck up behind me and hit me in the back of the knee with a stick. She clubbed me good. She hit me on the back of the leg, almost knocked me over. And she’s sitting there, shaking the stick at me and yelling at me in Navajo and I don’t speak Navajo, did not understand Navajo. She pointed at my truck and got in my truck and so I gave her a ride down to her house. And her granddaughter was there and she was telling me that grandma was upset that I was violating her space. Well what I was doing, I was doing my job. I’m an archaeologist. This is what I do. I see something, I record it. And it made me realize that I was violating her space. I should have asked her permission first because this belonged to her. I saw it as a resource, but it was her home. It was a part of her property. I should have asked permission. So over the years it’s kind of – what I tell people is that if somebody invites you into their home, invites you into their space, they’re giving you permission to violate their space in a sense, and when they give you something you have to ask them “What am I supposed to do with this?” A lot of times they give you stuff and they say, as Susan pointed out, “This is for you”. And how you use that gift is really up to you. And so it’s kind of that. So it kind of puts the onus back on your role. What are you going to do with all this knowledge that you’re gathering? And how is it going to benefit not only yourself, but the people you’re studying in the communities that you’re studying?

Phillip Tuwaletstiwa: I’m more interested in the pure practical part. You go to a community. What are the practical ways of dealing with people? When you have an objective and you’re going to have to interact with that community and with individuals. What are

434 Andrew G. Turk and David M. Mark

the ways that have worked for you? All of you have done it, so what has worked for you very well? Did you seek advice of people that worked in the community before? Did you have an introduction? Had you studied the culture? It can be very small things, but what have you done that will improve your chances of success? And then for kind of human interest, what didn’t work? I remember one time I went into a longhouse up in Iroquois and I was with a lady friend of mine and I went over and sat with her. Well I should have been over there sitting with the men, and they politely got me onto the seat where I should be, but I had no way of knowing that. Maybe I should have asked, “Where do I sit?” but I didn’t. So that was to me kind of funny and embarrassing. Anyway, so my interest is a little different. If it’s not practical, if it’s not going to help me with what I would like to see happen, it is not so interesting.

David Stea: Unfortunately, we have run out of time in this session. I would like to thank the panelists and everyone else here for their participation in a very interesting session. References Rice, Keren. In press. Ethical issues in linguistic fieldwork. In Oxford Handbook of Linguistic Fieldwork, Nicholas Thieberger (ed). Oxford: OUP. Rigney, Lester I. 1997. Internationalism of an Aboriginal or Torres Strait Islander anti-colonial cultural critique of research methodologies: a guide to Indigenist research methodology and its principles. Research and Development in Higher Education: Advancing International Perspectives. Higher Education Research and Development Society of Australia, Annual International Conference Proceedings 20: 629–636. Smith, Linda T. 1999. Decolonizing Methodologies: Research and Indigenous Peoples. Dunedin NZ: University of Otago Press. Turk, Andrew G., Mark, David M., O’Meara, Carolyn & Stea, David. In press. Geography€– documenting terms for landscape features. In Oxford Handbook of Linguistic Fieldwork, Nicholas Thieberger (ed.). Oxford: OUP. Turk, Andrew G. & Trees, Kathryn A. 1999. Ethical issues concerning the use of Geographic Information Systems technology with indigenous communities. Proceedings of the Australian Institute of Computer Ethics Conference (AICEC99), Melbourne, Australia, 385–398. Whiteley, Peter M. 1998. Rethinking Hopi Ethnography. Washington DC: Smithsonian Institute Press.

Notes on contributors

Lars Brabyn is a senior lecturer in the School of Social Science at the University of Waikato, New Zealand. He has researched methods for representing landscape values, which includes completing a PhD on landscape classification in the mid-nineties. He has developed two versions of the New Zealand Landscape Character Classification. Integral to this research is the application of GIS and the wide range of relevant geospatial data sets now available. Lars’ core area of teaching is GIS, but he is also teaching about tourism environments and associated planning issues. Niclas Burenhult is Research Staff at the Max Planck Institute for Psycholinguistics in Nijmegen, Netherlands, and Associate Professor of Linguistics at Lund University, Sweden. Since his PhD in General Linguistics from Lund University in 2002, he has been a member of MPI’s Language and Cognition Group, carrying out intensive research on spatial representation in little-known language settings. He coordinated the group’s cross-linguistic project on landscape categorization and was the editor of the ensuing Language and Landscape: Geographical Ontology in Cross-linguistic Perspective (a special issue of Language Sciences, 2008). He has a general interest in the relationship between language, cognition and culture, with a focus on spatial categories in endangered languages, language description and documentation, and linguistic prehistory. He is a leading expert on the Aslian languages, spoken in the Malay Peninsula. His publications on spatial topics have appeared in leading journals such as Language, Language Sciences, Linguistic Typology, and Studies in Language. Chris S. Duvall is an Assistant Professor in the Department of Geography at the University of New Mexico. His research interests range from biogeography to cultural ecology, and he has published in the Journal of Biogeography, Landscape Ecology, The Geographical Journal, and elsewhere. His field research in Mali, where he has traveled and worked since 1994, explores the Indigenous knowledge of traditional land-use practices, and the role of these practices in ecosystem structure and function.

436 Landscape in Language: Transdisciplinary Perspectives

Scott Heyes is an Assistant Professor of Landscape Architecture at the University of Canberra. Recently, he was the 2010 Roberta Bondar Postdoctoral Fellow in Canadian Studies and Indigenous Studies at Trent University, Canada and is an External Fellow at the Arctic Studies Center, Smithsonian Institution. His research and teaching interests are concerned with Indigenous conceptions and knowledge of the land and sea. Scott holds a PhD in Geography from McGill University. He is an avid outdoorsman with a penchant for fishing. Gary Holton is a documentary linguist specializing in the Athabaskan languages of Alaska and the Papuan outlier languages of Eastern Indonesia. His publications include a grammatical sketch of Tobelo, a pedagogical dictionary of Tanacross Athabaskan, and a dictionary of Western Pantar (Lamma). His current research includes unraveling the linguistic prehistory of the Alor-Pantar languages and the cross-linguistic comparison of landscape categorization along the Athabaskan-Eskimo boundary. Dr. Holton is currently Associate Professor of Linguistics at the University of Alaska Fairbanks and Director of the Alaska Native Language Archive. Bruce Janz is Chair and Associate Professor of Humanities in the Department of Philosophy, and the Director of the Center for Humanities and Digital Research at the University of Central Florida in Orlando, Florida. His PhD, in Philosophy, is from the University of Waterloo (1992). He writes on concepts of place across multiple disciplines. He also writes on African philosophy, contemporary European philosophy, cultural philosophy and theory, interdisciplinarity, and the history of mysticism. He is the author of, among others, Philosophy in an African Place (Lexington, 2009); ‘Making a scene and dwelling in place: exhaustion at the edges of modes of place-making,’ Rhizomes; ‘The terror of the place: anxieties of place and the cultural narrative of terrorism,’ Ethics, Place and Environment; and ‘Thinking like a mountain: ethics and place as travelling concepts,’ Drenthan, Keulartz & Proctor, New Visions of Nature, Springer. Stephen C. Jett holds an A.B. in Geology (Princeton 1960) and a PhD in Geography (Johns Hopkins 1964). He taught geography at Ohio State University and, from 1964, at the University of California, Davis, serving as chair; he became emeritus in 2000. From 1996, he was also Professor of Textiles and Clothing. He serves on the boards of various research organizations. Jett’s research specialties include Navajo history and culture. He wrote or co-authored: Tourism in the Navajo Country: Resources and Planning (1966); award-winning Navajo Wildlands (1967); House of Three Turkeys: Anasazi Redoubt (1977); award-winning Navajo Architecture: Forms, History, Distributions (1981); Navajo Placenames and Trails of the Canyon de Chelly System, Arizona (2002); and France (2004); plus scores of



Notes on contributors 437

articles. He has also published widely on pre-Columbian transoceanic influences, and he founded and edits Pre-Columbiana: A Journal of Long-Distance Contacts. He has curated or co-curated four textiles exhibits. Leslie Main Johnson teaches anthropology at Athabasca University, Canada. She lived in northwestern British Columbia in Gitksan territory for 12 years before returning to graduate school at the University of Alberta in the 1990s, where she earned her MA and PhD in Anthropology and held the Grant-Notley Postdoctoral Fellowship from 1997–2000. Her research interests include ethnoecology, ethnobiology, traditional textile production, and health and healing among Canadian First Nations. She has recently published Landscape Ethnoecology, Concepts of Physical and Biotic Space (co-edited with Eugene Hunn), and Trail of Story, Traveller’s Path based on her landscape research in northwestern Canada. Her latest work is a chapter on landscape written with Iain Davidson-Hunt in the forthcoming text Ethnobiology (Wiley Interscience, co-edited by E. N. Anderson, Karen Adams, Deborah Pearsall, Eugene Hunn and Nancy Turner). She has also published a number of articles and several book chapters on ethnobiology and landscape. James Kari is Professor Emeritus of Linguistics at the University of Alaska Fairbanks. In 1997 he retired from the Alaska Native Language Center, University of Alaska Fairbanks. He has specialized in the Alaska Athabaskan languages since 1972, and he continues to work on several Alaskan Athabaskan language projects, especially Lower Tanana, Ahtna and Dena’ina. His specializations include lexicography, texts, grammar and place names research. His publications include Shem Pete’s Alaska (with James A. Fall, 2003), Koyukon Athabascan Dictionary (editor, 2000), and The Dene-Yeniseian Connection (editor, with Ben A. Potter, 2010). Werner Kuhn is a Professor of Geoinformatics at the Institute for Geoinformatics, University of Münster (Germany). He leads MUSIL (the MUenster Semantic Interoperability Lab, http://musil.uni-muenster.de), a research lab working on semantic interoperability problems in the semantic, social, and sensor web. Since 2002, this group has been establishing the theoretical and practical foundations for Semantic Reference Systems, including geospatial ontologies and approaches to their grounding. Dr. Kuhn has published on topics ranging from humancomputer interaction through interoperability standards to geospatial semantics and ontology. In his work, he has been applying ideas from cognitive linguistics (image schemas, metaphors, blendings) and ecological psychology (affordances, meaningful environments) to the design of computational approaches to semantic translation.

438 Landscape in Language: Transdisciplinary Perspectives

Renee Pualani Louis is a Hawaiian woman and graduate of the University of Hawai‘i Geography Department. She is an Indigenous cartographer passionate about Hawaiian storied place names, cross-cultural ethical research standards, and advocating the integration of Indigenous spatial knowledge systems with Western geosciences. She volunteers with the Hawai‘i Board on Geographic Names, is co-chair of the Indigenous Peoples Specialty Group of the Association of American Geographers, is secretary of the Indigenous Peoples Knowledges and Rights Commission of the International Geographic Union, and treasurer of a Hawai‘i Island based non-profit that supports quality education and sustainable communities. She is also owner of a small business, Pacific Data Digitizing, specializing in merging Indigenous wisdom of place with technological innovation. Elizabeth Lynch is currently pursuing her doctorate in archaeology at the University of Wyoming. She earned a Bachelor’s degree in French from the University of Texas at Austin and a Master’s degree in Anthropology from New Mexico State University. Her research interests center on building a conceptual framework for understanding how prehistoric peoples of southeastern Colorado and northeastern New Mexico used bedrock grinding areas as social or communal gathering spaces within the broader, regional landscape. David M. Mark is a SUNY Distinguished Professor in the Department of Geography at the University at Buffalo (UB), State University of New York, where he is the director of the Buffalo site of the National Center for Geographic Information and Analysis (NCGIA). Mark also is Project Director of the Integrative Graduate Education and Research Traineeship (IGERT) in Geographic Information Science, and is a member of UB’s Center for Cognitive Science and the National Center for Ontological Research. Mark completed his PhD in Geography at Simon Fraser University (Burnaby, Canada) in 1977, and joined the University at Buffalo in 1981. He has written or co-authored more than 230 publications, including numerous articles, book chapters, and technical reports, and has edited five books. His research interests include ontology of the geospatial domain, geographic cognition, cultural differences in geographic concepts, geographic information science, and digital elevation models. Brian Murton received his BA (1960) and MA (1961) from the University of Canterbury, and his PhD (1970) from the University of Minnesota. A New Zealander, he is of Irish, Scottish, English and Māori descent. After teaching for a year at York University, in 1969 he joined the Geography Department at the University of Hawai‘i, where he is now Professor Emeritus. Until 1995 his research focused on aspects of the agrarian historical geography of southern India, as well as on famine and food vulnerability. From the mid-1990s he became involved in



Notes on contributors 439

Waitangi Tribunal claims in New Zealand, producing four extensive reports and participating in a number of Tribunal hearings as an expert witness. He has also been a member of a collaborative research project (non-Māori and Māori) focusing on landscape transformation in the Bay of Islands, New Zealand, between 1769 and 1840. Susan Paskvan be’ooze’ dehoon Denaakk’e hełde K’etsoo beeznee. Bedełnekkaa Benedict yeł Eliza Jones heelaanh. Meneelghaadze T’oh huts’enh ts’aadaanlet. Denaakk’e hedohudeleeh dehoon hedok’uhudeł’eehenh. Yukon-Koyukuk School District okko kk’oneedeneeyh. Her name is Susan Paskvan while in Native they call her K’etsoo. Her parents are Benedict and Eliza Jones. She is from Koyukuk, Alaska. She is learning Denaakk’e while teaching. She works for Yukon-Koyukuk School District. Following in her parent’s footsteps, her passion is teaching language and culture. She is working on her thesis Native Place Names near Kaltag, Alaska for her interdisciplinary Master of Arts degree in Alaska Native Languages and Linguistics at the University of Alaska Fairbanks. Renée E. Sieber received her PhD from Rutgers University and is currently an associate professor at McGill University in Montreal, Canada. Her prime research focus is Public Participation Geographic Information Systems (PPGIS), which is methods by which those who are marginalized from public policy can use computational mapping and spatial databases to better participate in policymaking. These individuals may be inner-city or Indigenous people. She brings to this a background as a community organizer and activist as well as a computer programmer. She increasingly researches PPGIS on the geospatial Web 2.0 (Geoweb). Her research areas are diverse. She leads a team of 10 researchers in the use of the participatory Geoweb for global environmental and climate change. She also conducts research into the digital humanities with Chinese historical databases. She organized the first public participation GIS conference, co-founded the GIS specialty group of the Canadian Association of Geographers and co-organized Spatial Knowledge and Information Canada, the first academic GIS conference in Canada. David Stea received a BS (hons) in Mechanical/Aeronautical Engineering from what is now Carnegie-Mellon University, an MS in Psychology from the University of New Mexico, and a PhD in Psychology from Stanford University. He is Professor Emeritus of Geography and International Studies, Texas State University, San Marcos, Research Associate, Center for Global Justice (Mexico), and Enrique Aragon Professor (emeritus), Universidad Nacionál Autonoma de México. A co-founder of the field of environmental psychology, his research interests have included spatial cognition, map learning in young children, techniques of

440 Landscape in Language: Transdisciplinary Perspectives

effective participatory planning with Indigenous peoples worldwide, and sustainable development. His books, including Image and Environment, Maps in Minds, Environmental Mapping, and Placemaking, have been translated into various languages. Thomas F. Thornton is an environmental anthropologist and Senior Research Fellow at the Environmental Change Institute, School of Geography and Environment, University of Oxford, where he also directs an MSc course in Environmental Change and Management. His primary research interests are in landscape perception, use, and conservation among Indigenous peoples of the North, especially the Pacific Coast of North America. His most recent books are Being and Place among the Tlingit (University of Washington Press, 2008) and Haa LÈelk’w Has AanÌ Saax’˙: Our Grandparents’ Names on the Land (Sealaska Heritage Institute and University of Washington Press, 2010). Carmelita Topaha is a member of the Navajo Nation, Newcomb Chapter. She has an Associates degree in Anthropology from San Juan College in Farmington, New Mexico, and a BA in Anthropology from Fort Lewis College, Durango, Colorado. She has worked as a consulting anthropologist, archaeologist, or ethnographer on a variety of projects. She has conducted many archaeological clearances for homes, utility lines, waterlines, and power lines, and has been the primary Navajo consultant for Mark, Stea, and Turk’s Ethnophysiography project. She also is a weaver, a potter, and a writer. Carmelita teaches courses at San Juan College. Andrew G. Turk has degrees in Surveying, Applied Science (Cartography) and Arts (Psychology Honours and Philosophy major) and a PhD. In the 1970s and early 1980s he worked for the Australian Government producing topographic maps. In 1983 he commenced research at Melbourne University into design of tactual maps and graphics for visually impaired persons. From 1993 he worked at Murdoch University. His research and consultancy projects covered: design and evaluation of user interface, websites and interactive television; socio-technical methodologies for developing information systems; geographic information systems; and cultural and ethical aspects of ICT. He has worked with Indigenous communities in the Pilbara and Ngaanyatjarra Lands in Western Australia. His current research concentrates on ethnophysiography (cultural/linguistic aspects of conceptions of landscape). In 2007 he retired, but continues as an Adjunct Associate Professor at Murdoch University. He is working on a second PhD regarding philosophical approaches to concepts of place.



Notes on contributors 441

Christopher C. Wellen’s background is varied, but its central theme is water. He received a BSc in Water Ecosystems and Environments in 2005 from the McGill School of Environment. His senior research project centered on a landcover change study which integrated satellite imagery and interviews with Cree hunters to demonstrate their response to landcover changes associated with coastal uplift. He received an MSc in Geographic Information Science from McGill’s Department of Geography in 2008. For his thesis, Wellen developed an ontology of waterbodies using concepts from the Cree language, and implemented it in a digital gazetteer. This work demonstrated the importance of developing ontologies with Indigenous peoples and presented basic techniques for doing so. He currently is working on a PhD in Ecohydrologic Modeling at the Geography Department of the University of Toronto. His research interests include hydrology, modeling, Bayesian statistics, human-environment interaction, and cyberinfrastructure.

Index

A aestheticâ•… 3, 60, 280, 396 affordancesâ•… 7, 16, 40, 143, 158, 160, 235, 276, 279–282, 284, 286–287, 292, 321 Alaskaâ•… 225–236, 239–260, 261–274, 275, 280, 284–286, 322, 328, 331–333, 336, 355, 428 alpineâ•… 296–299, 304, 307–310, 317, 319, 401 ancestorsâ•… 59, 62, 64, 74, 78, 85, 87–95, 172–173, 220, 272, 331, 421 ancestralâ•… 84, 90, 168, 173, 191, 193, 207, 220, 276, 284, 287, 328, 331, 355, 357–358 ancestral beingsâ•… 59 anthropologistâ•… 1, 95, 108, 122, 264, 328, 332–333, 350, 369, 370, 377, 386, 415 anthropologyâ•… 1–2, 4, 10, 26, 32, 51, 55–56, 101, 103–104, 106, 109, 118, 264, 287, 297, 348, 350, 377, 428, 431 appropriationâ•… 413 archaeology â•… 4, 32, 52, 103, 132, 343–344, 347–350, 358, 366, 386, 416, 420, 428, 433 Arcticâ•… 187–191, 219, 225 Aristotleâ•… 110, 424 Athabascanâ•… 239–259, 261–272  see also Athabaskan Athabaskanâ•… 32, 95, 225–236, 280, 284, 287, 327–337, 355, 428 see also Athabascan B Basso, Keithâ•… 1–2, 11, 55, 63, 79–80, 264, 267–268, 322, 327–329, 354, 357, 383, 396

berriesâ•… 263, 280–282, 296, 307–309, 317–318, 320 berry patchâ•… 283, 294–295, 309–310, 312, 318–321 biodiversityâ•… 287, 396 biophysical environmentâ•… 14, 121–123, 126, 134–135, 137, 321–322 biospiritualâ•… 126–127, 276, 282–284, 287 borealâ•… 187, 296, 299, 307, 322, 333 boundary, boundariesâ•… 41, 59, 91, 94–95, 177, 179, 190–191, 205, 218, 221, 225–226, 233, 235, 240, 251, 259, 262, 287, 294–296, 308, 322, 356–357, 359, 400, 425 C calendrical sites â•… 320 campâ•… 30, 40, 61, 189, 250, 261– 264, 270, 273, 278, 292, 306, 310–312, 318–319, 357–366 Canyon de Chellyâ•… 259, 327– 330, 334–337 cardinal directionsâ•… 40, 313, 333, 335, 347 caribouâ•… 296, 304, 307–308, 319, 321 cartographyâ•… 88, 168–170, 174, 383, 393 case studiesâ•… 4, 7–9, 25–28, 33, 36–38, 42, 48, 51, 56, 66, 239 Casey, Edwardâ•… 1, 48, 58, 60, 62, 65, 75, 79–80, 83, 91 categories, categorizationâ•… 1, 4–10, 34–35, 64, 76–77, 125– 136, 143–144, 149, 160, 165, 195, 225–226, 228–229, 232, 235–236, 254, 264–265, 298,

370, 374, 381–383, 388, 390, 395, 398, 400, 403–406, 413 ceremoniesâ•… 7, 15, 28, 31, 41, 146, 161, 269–270, 332–336, 347, 356–357, 414–415, 429, 432–433 chantsâ•… 84, 169, 180, 182, 185 chantwayâ•… 332, 336 classificationâ•… 1, 10, 38, 47, 118, 125–126, 132–134, 137, 144, 149, 158, 245, 265, 275–276, 279, 287, 385, 395–406 climateâ•… 40, 122, 129, 134–135, 147, 241, 264 cognitiveâ•… 1–2, 4, 6, 11, 51, 83, 168–169, 174, 191, 195, 264, 287, 328, 338, 398, 403 cognitive mapâ•… 153, 266, 269, 332, 336, 353–354, 356–357, 362, 366 collaboratorsâ•… 411–412, 417 colonialâ•… 12, 60, 81, 167–168, 174, 179, 413 colonizationâ•… 41, 63, 388 colonizedâ•… 56, 179 communicationâ•… 18, 85, 107, 121, 147, 179, 219, 268, 377, 390, 397, 403–405, 413, 422 communityâ•… 12–13, 26–28, 31–32, 34, 53–55, 57, 64–65, 82, 106, 161–162, 170–171, 179, 181, 184–185, 187, 189, 193, 197–198, 200, 209, 218, 231, 264, 270–272, 277, 281, 307–308, 344, 349–350, 371, 378, 384–386, 388–390, 392, 397, 403, 412–414, 417–421, 424, 426, 429–430, 432–433 see also speech community conceptsâ•… 7–8, 13, 26, 47–49, 51, 53–57, 60–61, 64–66, 69,

444 Landscape in Language: Transdisciplinary Perspectives

73–76, 79–81, 83, 86, 94, 108–109, 113–114, 118, 123, 125–126, 128, 130, 132, 134–136, 170, 173, 176, 192, 205, 231– 232, 243, 291–292, 295, 298, 308, 319, 321, 350, 371–374, 376, 378, 383, 385, 387–388, 390–391, 414 conceptualizationâ•… 11–13, 25, 36, 47, 54, 85, 149, 165, 214, 219, 225–226, 231, 233–235, 275, 282, 369–371, 373–377 conceptual ontologyâ•… 387–388 consultantsâ•… 28, 33, 125–126, 128, 132, 134, 137, 155, 266–269, 357, 359, 414, 422 cosmologyâ•… 73, 81, 94, 192, 276, 280 countryâ•… 26–28, 30, 33, 40–41, 49, 58–63, 90–91, 248, 283– 287, 298–299, 301, 304, 306– 307, 309, 316–317, 319, 328–333, 335, 362, 396, 420, 430 creation, creation storiesâ•… 7, 31, 41, 61, 82, 84, 92, 127, 191, 330, 332, 353, 359 cross-culturalâ•… 1, 4, 7, 56, 106, 109, 113, 121–123, 372 cross-linguisticâ•… 1, 4, 7, 77, 128, 134, 143, 225, 239, 369–370 culturalâ•… 1–4, 6–7, 10–13, 25, 28, 37, 41, 47–48, 54–57, 60–63, 67–69, 75, 95, 103–117, 121, 144, 146, 149, 156, 164, 167– 172, 178–180, 182, 184–185, 220, 225, 248, 264–265, 275– 276, 278, 280–284, 286–287, 292, 295, 297–298, 310–312, 328–329, 333, 336–338, 354–355, 366, 370, 372–373, 377–378, 381–382, 384–386, 395–396, 398, 404–406, 412, 415, 419–420, 424–425, 430 cultural diversityâ•… 287, 391 cultureâ•… 1, 3, 8–10, 12, 15–16, 25, 35–36, 41, 47, 56, 58–61, 63–64, 68–69, 76–77, 84, 102, 104–117, 119, 122–123, 146, 161, 164, 168, 170, 174, 179, 185, 225, 241, 248, 263–265, 276, 280, 284–285, 293–294, 319, 322,

332, 337–338, 343, 348–350, 370, 378, 385, 390, 392, 396, 398, 403, 405–406, 411, 415–416, 418, 421, 424, 427, 430, 432–433 customsâ•… 7, 47, 81, 248 D decolonizingâ•… 413 deconstructionâ•… 419 descriptive model â•… 8, 13, 25–26, 36–38, 42, 50, 66, 68 dialect, dialectsâ•… 36, 74, 124–125, 151, 230, 234, 240, 259, 261– 263, 267, 276, 296, 304, 323, 338, 383, 385, 395, 421, 428 dictionaryâ•… 28–29, 34–35, 42, 194, 229, 235, 243, 249, 297, 297, 348–350, 415 directional, directionalsâ•… 6, 16, 39, 233, 235, 239–240, 245, 254–259, 265–266, 272, 276, 285, 313–315, 323, 327, 332, 334–335, 347, 357 documentationâ•… 37, 144, 165, 225, 228, 235, 251, 264–265, 297, 370, 377, 384 DOLCEâ•… 17, 370, 374, 376, 378 dwellingâ•… 2, 41, 55, 57–58, 64, 68, 79–80, 83–84, 94–95, 190, 275, 322 E ecologyâ•… 27, 33, 52, 297, 318, 321 see also human ecology ecosystemâ•… 121, 147, 275, 283 ecotopeâ•… 276–277, 281–283, 285–286, 292, 304, 310, 314, 318, 321 elevationâ•… 27, 126, 129–130, 143, 147–149, 153–156, 164–165, 229–230, 235, 296, 334 elicitation â•… 255, 297, 382, 385, 390–391 elicitingâ•… 10, 125, 412 embodied, embodimentâ•… 55, 61, 76, 80, 83, 102, 113, 134, 156, 192, 278–279 emicâ•… 51, 90, 107–108, 113 emic/eticâ•… 107, 113 empiricalâ•… 51, 53, 108, 114, 267

empiricismâ•… 105, 108, 112–113 empiricistâ•… 14, 66 environment, environmentalâ•… 1–4, 7–8, 13, 26, 38–40, 55, 57–60, 63, 77, 79–80, 89, 91, 94–95, 105, 121–123, 126–127, 134–137, 144, 161, 164, 167–170, 172, 174, 187, 189–191, 193–196, 208, 214, 217, 219–221, 251– 252, 264, 267, 272, 279–281, 286–287, 292, 316, 327–328, 330, 353–354, 357, 365, 374, 396, 404–406 epistemology, epistemologicalâ•… 74, 81, 84, 168, 173, 179, 185, 220, 381–382, 384, 414 ethics, ethicalâ•… 13, 55–6, 111, 384, 391, 411–412, 424–427 ethnocentrismâ•… 56 ethnoecologyâ•… 10, 121–123, 134, 136–137, 291–292, 316, 321 ethnogeographyâ•… 118, 121–122, 136, 240, 276 ethnography, ethnographer, ethnographicâ•… 11, 36–37, 48, 56, 63, 67–68, 77–78, 106–109, 118, 121, 125, 165, 174, 182, 191, 251, 264, 267, 271–272, 354 ethnomethodologyâ•… 54 ethnophysiographyâ•… 7–10, 12–15, 17, 25–43, 47–48, 50–51, 54–56, 60, 62–69, 77, 101–103, 105–109, 111–119, 279, 283, 349, 403, 405 eticâ•… 34, 51, 107–108, 113, 156, 163 F featureâ•… 2, 5–6, 8, 10–11, 29–31, 33–37, 39–40, 49, 53–54, 59, 63–64, 66, 77, 83, 89–91, 101, 104, 109, 110–111, 114, 121–123, 125–130, 132, 134–137, 143, 149, 151–152, 154, 158–159, 162–164, 168, 170, 173, 178, 187, 190, 193–195, 197–198, 208–209, 214, 217, 228–230, 233, 235, 240, 242–243, 246–247, 250–251, 254, 257, 259, 278–279, 281, 284–287, 291–292, 297–298, 300–



301, 304, 306, 308, 310–313, 315, 318–322, 327, 329–330, 332–335, 337–338, 343, 357, 365, 372–374, 376–378, 382–383, 385–386, 388–391, 395, 399, 403, 406 feature, geographicâ•… 1, 9–10, 121, 126, 128, 134, 136–137, 144, 190–191, 193, 208, 265, 279, 297 feature, landscapeâ•… 2, 5–10, 16, 25–26, 30–31, 35–37, 39, 41, 47–49, 75, 89, 123, 128, 144, 163, 225, 235, 276–279, 297, 318–319, 327, 330, 357, 382, 385, 403, 412 field interviewsâ•… 28, 33 fieldworkâ•… 5, 28, 34, 42, 104, 106–107, 118, 187, 193, 304, 412, 428 folksonomyâ•… 404 folk taxonomyâ•… 122–123 formalizationâ•… 117, 375, 377, 385, 390–391 formal ontologyâ•… 106, 109–110, 115, 117–118 foundational ontologyâ•… 369– 370, 373–374, 376 G genealogyâ•… 75, 78, 82–84, 86–87, 94–95, 109, 167–169, 180, 182, 328 generic landscape termsâ•… 9, 11, 35, 37–39, 47, 63, 116, 143, 165, 281–282 geographicâ•… 1, 3–4, 6–10, 12, 55, 63, 80, 85–86, 89, 102, 110, 114, 121, 123, 125–126, 128, 134, 136–137, 144, 147, 154, 156, 164, 174, 179, 181, 184, 187, 190, 193, 195–198, 208–209, 218–219, 233, 239–248, 251–252, 254, 257–259, 264, 278–279, 287, 297–298, 304, 315, 321, 323, 329, 332–333, 335, 339, 370, 377, 381–383, 390–391, 395, 405, 419 geographic information systemsâ•… 8, 12, 381, 395 see also GIS

Index 445

geographyâ•… 1, 3–4, 26, 36, 51, 55, 57, 101, 106, 110–111, 118, 123, 147, 159, 167, 193, 220, 239, 242, 245, 250–251, 254, 259, 264–265, 285, 298, 321–322, 332, 349–350, 377, 382–383, 385, 397, 403, 415, 419, 428 Gibson, James J.â•… 2, 7, 16, 279– 280, 292, 374 GISâ•… 8, 10, 12–13, 115–116, 118, 250, 271–272, 349, 381–392, 395–402, 405–406, 414– 415, 417, 424–425, 430–431 see also geographic information systems Googleâ•… 13, 148, 159, 266, 272, 382, 384, 419 grammarâ•… 4, 6, 8, 41, 104, 134, 254, 257 Grounded Theoryâ•… 37, 66 H habitatâ•… 122, 265, 291–292, 296–297, 306, 308, 317–318, 328, 336 habitationâ•… 63, 91, 143–144, 191, 310, 312, 319 habitusâ•… 80 Heidegger, Martin â•… 52–53, 55, 59, 62, 65–67, 76, 79–81, 85, 94, 112–113 hermeneuticsâ•… 66, 107, 113 Historic Preservationâ•… 414 historyâ•… 3, 11–12, 27–28, 32, 60, 73, 75, 88, 90, 94, 109, 114, 118, 121, 149–150, 161, 172, 174, 191, 228, 264, 267, 269, 271–272, 295–296, 312, 319–320, 322, 327–328, 334–336, 338, 347, 349–350, 366, 414, 416 holisticâ•… 52, 82, 86, 134, 405 Holy Beingâ•… 344 homelandâ•… 74, 90, 92, 291–294, 296, 317, 322, 332, 356 Hopiâ•… 10, 56, 415, 419–420, 427 human ecologyâ•… 3, 318 hunterâ•… 32, 40, 132, 193, 196–197, 205–208, 213–214, 217–218, 220–221, 312, 321, 332–333, 337, 363, 366, 384, 387–388

hunter-gathererâ•… 27, 353–354, 356 Husserl, Edmundâ•… 52–55, 65–67, 112–113 hydrography, hydrographicâ•… 291, 298, 310, 317, 322, 381–382, 386–388 hydrologyâ•… 30, 128, 132, 134–135, 219, 248, 370, 373 I iceâ•… 15, 189–191, 193–221, 228, 230–231, 248, 284, 292, 298, 304–306, 316, 321, 323–324, 385 indigenousâ•… 7, 11–13, 26, 31, 41, 47, 50, 56–59, 61, 63–64, 67–69, 73, 76, 79, 86, 121–122, 144, 146, 165, 173, 178, 219– 220, 225–226, 228, 240, 261, 264, 270, 272–273, 280, 292, 322, 366, 381–382, 384–386, 391–392, 411–414, 416–419, 426–429 indigenous knowledge, knowledgesâ•… 20, 47, 56–57, 76, 79, 122, 366, 385, 391, 419, 427 informationâ•… 12, 51–52, 61, 109, 111, 150, 170, 178–179, 181–182, 184, 191, 195, 208, 217–218, 221, 242, 250, 264–265, 271–272, 297, 327, 333, 338, 350, 354, 359, 366, 373, 377, 381–383, 386–388, 391–392, 395–396 information systemsâ•… 8, 12–13, 17–18, 51, 272, 373, 381, 395 Ingold, Timâ•… 1–2, 76, 80, 83, 91, 109, 277, 280, 287 inscriptionâ•… 109, 174, 384, 418 interdisciplinaryâ•… ix, x, 9, 52, 67, 101–102, 117, 382 intersubjectivityâ•… 54, 79, 115 Inuitâ•… 187, 189–195, 197–198, 200–202, 209, 219, 226, 385 Inuktitutâ•… 193–194, 197–198, 203–204, 206, 209, 211–213, 215–219 itinerariesâ•… 327, 332

446 Landscape in Language: Transdisciplinary Perspectives

J Jicarillaâ•… 353–366 journeysâ•… 90, 152, 205, 207–208, 213–214, 217–218, 220, 284, 333–335, 345, 359–361 K Kant, Immanuelâ•… 58, 112, 424 kinshipâ•… 39, 77, 87, 425 knowledgeâ•… 7, 11–12, 36, 42, 47, 51–52, 54, 56–59, 61, 63, 67, 73–76, 79, 81–84, 86–89, 93, 102–103, 107–108, 113, 115–118, 121–123, 126–128, 130, 135–137, 156, 163–164, 167–170, 172–173, 179–180, 185, 189–191, 193, 195, 217, 219, 239, 251–252, 258–259, 261, 264, 267–268, 270– 273, 280, 287, 291–292, 297, 304, 312, 315, 322, 328, 332, 334, 338, 344, 346–347, 349, 353–354, 357, 359, 363, 365–366, 369–370, 381–382, 384–385, 387, 391–392, 397–398, 412–418, 421–423, 425, 427–433 see also indigenous knowledge, knowledges knowledge representationâ•… 369, 377 L land, landsâ•… 1, 3, 9–10, 12, 13, 25, 28, 32–33, 41, 47, 58–62, 64, 69, 75, 77–79, 82, 84–86, 88, 90–91, 105, 110, 115–116, 121, 123, 132, 134–136, 149–150, 153, 157, 167–169, 171–173, 175–177, 185, 187, 189–193, 194, 195–202, 204–209, 214, 217–221, 222–223, 240–241, 247, 257, 261, 263–264, 267– 272, 275–276, 280–282, 287, 291–293, 297–301, 310, 312, 317, 319–322, 327, 330–332, 334, 338, 344, 346, 348–350, 366, 381–386, 388, 390, 395–396, 420, 424, 430 landformâ•… 4–10, 34, 47, 61, 91–92, 101, 104–106, 108–112,

114, 116, 118, 125–126, 129, 143, 150, 153–154, 156, 160, 164, 191, 229–230, 246, 248, 280–283, 285–287, 345–346, 396, 399–401, 403 landscapeâ•… 1–11, 13, 18, 25–31, 33–42, 47–52, 54, 56, 58–65, 68–69, 73–83, 86–89, 92, 94–95, 102–103, 105–106, 109, 111, 115, 121–123, 128, 130, 132–135, 137, 143–145, 149–150, 152–157, 159, 163–165, 167–168, 170–173, 178, 187, 190–191, 194, 208, 225–233, 235–236, 239– 240, 244–245, 250, 257–258, 267, 272, 275–279, 280–285, 287, 291–301, 306–307, 310– 315, 317–322, 327–330, 333, 336, 338, 343–345, 347–350, 353–359, 361, 363, 365–367, 369–371, 373–378, 381–388, 390–392, 395–406, 411–412, 415–416, 432 landscape categoriesâ•… 8, 165, 232, 235, 325, 404 landscape classificationâ•… 18, 144, 275–276, 287, 395, 398– 400, 402–406 landscape featureâ•… 2, 5–10, 25–26, 30–31, 35–37, 39, 41, 47–49, 75, 89, 123, 128, 144, 163, 225, 235, 276–279, 297, 318–319, 327, 330, 357, 382, 385, 403, 412 Landscape in Language Workshopâ•… 3, 26, 42, 411 landscape termsâ•… 4–6, 8–11, 25, 28–29, 33–35, 37–39, 41–42, 47–49, 54, 58, 63, 89, 109, 154, 165, 225, 232, 244, 250, 278, 281, 297–298, 300–301, 317 319, 348, 350 landscape terminologyâ•… 33–34, 164, 228–229, 240, 276, 301, 321–322, 394 landform terms, terminologyâ•… 5, 14, 101, 105, 108–112, 114, 116, 118, 142, 150, 165, 224, 229, 281–282, 286, 290 language, languagesâ•… 1–11, 25–28, 30–32, 34–42, 48,

50–51, 53, 56–57, 59–60, 62, 68–69, 73–75, 77–78, 81, 85, 89, 94–95, 101–102, 106, 109, 111, 114–115, 119, 121, 124–125, 134, 136, 138–139, 143–146, 148, 149, 151–152, 158, 161, 164–165, 180, 194, 225–229, 231–232, 234–236, 239–244, 248, 250–255, 257, 259, 261, 263–264, 270–273, 275–278, 280–281, 291–294, 296–297, 301, 304, 312–313, 316–317, 320–323, 327–329, 331–333, 336, 338, 349–350, 355–356, 367, 369–373, 375–378, 381–388, 390–391, 395–398, 403–406, 411–412, 416–418, 421, 428–429, 431 language area, areasâ•… 124, 145, 148, 239–242, 250–251 language group, groupsâ•… 9, 16–17, 26, 28, 36, 39–40, 59, 225, 254, 328, 391 language documentation and preservationâ•… 37 lebensweltâ•… 53–54 Levinson, Stephenâ•… ix–x, 4–6, 8, 38–39, 77, 143, 165, 220, 225, 229, 235, 239, 245, 254, 287, 327, 336, 371 lexicalâ•… 6, 10, 164, 225, 235, 243, 257, 331, 387 lexicon(s)â•… 4, 6–7, 159, 187, 193–194, 235, 244, 249, 254, 257, 279, 296, 387 lifestyleâ•… 8, 27, 32, 40–41, 270 lifeworldâ•… 53, 57, 60–62, 67, 69, 79, 113 linguisticâ•… 1, 4, 6–8, 11, 13, 26, 41, 48, 55, 57, 63, 66, 78, 85, 107, 114–115, 118, 122–123, 128, 134–136, 143–144, 161, 165, 195, 225–228, 239, 277, 295, 297, 301, 304, 328, 333, 338, 369– 370, 378, 384, 395, 412, 418 linguisticsâ•… 3–5, 10, 26, 50, 101, 106, 114, 272, 287, 338, 376–377, 392, 418, 428 livelihoodâ•… 2, 321



M Maliâ•… 121, 123 Malpas, Jeffâ•… 12, 55, 60, 62, 80 mapâ•… 124, 145, 163, 168, 170, 174–182, 252–253, 263, 271– 272, 332–333, 353–368, 383, 386–388, 397, 400, 404–405 mappingâ•… 12–13, 15, 88, 168, 176–177, 221, 250, 271, 353– 354, 359, 365–366, 381–386, 405, 414 matrilinealâ•… 32 Max Planck Institute for Psycholinguisticsâ•… xi–xii, 4, 8, 26, 42, 77 mental mapâ•… 332–333, 355, 365 mereologyâ•… 106–107, 110–111, 117 metaphorâ•… 5, 39, 41, 75, 84, 86, 88, 153, 172–173, 190, 230, 233, 278, 416 metaphysicsâ•… 76, 85–86, 105, 113–114, 168–170, 172, 276–277 methodologyâ•… 28, 33–34, 51–52, 54–56, 65–66, 101–119, 125– 126, 165, 180, 195, 297, 386, 400, 411, 413 moralâ•… 267–269, 280, 283, 360 mountainâ•… 9, 29–30, 35, 38–40, 49, 57, 59, 62, 64, 89–90, 144, 154–156, 199–201, 228–230, 246–247, 253–254, 265, 291, 298–301, 317, 320, 322, 335, 343–346, 360–363, 372, 374, 376, 378, 382–383, 400–401 multidisciplinaryâ•… 9, 52 multimediaâ•… 180, 382, 386 mythâ•… 50, 91, 151–152, 248–249, 258, 265, 284, 288, 319–320, 332–336, 356, 368 N Na-Deneâ•… 240, 264, 276, 328–329 NAGPRAâ•… 415, 420–421, 427 names, namingâ•… see place name and proper name narrativeâ•… 31, 60, 62, 64, 78, 83, 84, 86, 87, 88–89, 91–92, 107, 109, 114, 116–117, 152, 220, 240,

Index 447

242, 245, 251, 257–258, 268, 297, 306, 320, 353–366, 428 Native Americanâ•… 79, 329, 350, 415, 420, 426–427, 431 natureâ•… 59–60, 76, 173, Navajoâ•… xi–xii, 8, 26, 31–36, 244, 252, 327–338, 343–350, 364–365, 366, 372, 374, 414– 416, 419–420, 430, 432–433 navigationâ•… 170, 191, 242, 279, 301, 316, 322, 353–355, 385, 390 non-indigenous scholarsâ•… 61, 79, 411, 417, 419 O ontologyâ•… 4–7, 9, 13, 17–18, 53, 67, 81, 108–111, 114–115, 117–119, 369–378, 381–392, 398 oralityâ•… 78, 84, 88, 174, 185, 195, 264, 267, 271, 280, 295, 336, 353–358, 384, 391, 413, 414, 416, 418, 423 orientationâ•… 190–191, 243, 254–257, 276–77, 279, 284– 285, 312–313, 332–333 P Pantarâ•… 143–165 Papuanâ•… 143–165 participantâ•… 28–29, 33–34, 51, 107–108, 125, 185, 193–198, 209, 217–218, 272, 388, 412, 421, 430 participationâ•… 79, 95, 170, 173, 277, 374, 419–420 patchâ•… 283, 294–295, 309–310, 318–321 pathâ•… 38, 90, 243, 255, 276, 287, 321, 344, 353–357 perceptionâ•… 2, 53, 55, 58, 79, 94, 121, 123, 168, 170, 259, 275–280, 284, 287, 313, 322, 327–329, 334, 338, 353, 357, 388, 398, 403, 405 performanceâ•… 88, 168–169, 172, 174 performativeâ•… 413 phenomenaâ•… 6–7, 53, 67–68, 82–84, 88, 94, 108, 112, 123, 168–169, 190–191, 195, 205, 248, 322, 328, 396–397, 403

phenomenology, phenomenologicalâ•… 1, 3, 12–14, 47–48, 51–55, 58–59, 61–69, 73–75, 79, 82, 86, 95, 103–105, 108, 112–113, 277, 287 philosophicalâ•… 26, 51, 56, 59, 63–66, 86, 101, 116, 119, 174, 369, 373, 383 philosophyâ•… 1, 4, 10, 12, 47–72, 76, 81, 101–119, 376, 383, 424 photographsâ•… 33–34, 79, 189, 247, 349, 399, 404, 414, 30–431 photosâ•… 28–29, 33, 49, 180, 184, 297, 404, 430–431 physiography, physiographicâ•… 7, 14, 33, 123, 125, 128, 157, 227, 233, 276–277, 278–279, 282– 285, 287, 310, 317 place, placesâ•… 1–3, 5–8, 11, 26, 30, 31, 35–36, 39–41, 47–50, 55, 57–65, 68–69, 73–95, 101, 104–105, 107–108, 110, 112, 114, 116–118, 123, 133–134, 136–137, 143–144, 149–152, 156, 160, 162, 167–170, 172, 179–180, 182, 184–185, 190, 192, 205, 225, 228–229, 231, 233, 242, 244, 247–250, 252, 255, 257, 265–273, 275–282, 284, 286–287, 292, 297–299, 302–304, 306–315, 317–322, 327–338, 344–349, 353–366, 383, 385–386, 388, 396–397, 412–414, 420, 423, 426 place attachmentâ•… 7, 47 place kindâ•… 292, 298, 314–315, 319–321 place names, placenamesâ•… 3–6, 8, 10–11, 31, 35, 39, 47, 63, 74–75, 78, 85, 88–90, 92, 94, 114, 118, 150, 152, 167–172, 178–185, 190–192, 225, 228–232, 239–259, 261–273, 275, 282, 297, 327–338, 345–346, 357–359, 361–363, 383–389, 391 see also toponyms place/space, space/placeâ•… 76, 174, 179 positivismâ•… 14, 51–52, 73, 112–113

448 Landscape in Language: Transdisciplinary Perspectives

postcolonialismâ•… 57 postdisciplinaryâ•… 9, 51–52, 418 practiceâ•… 7, 11, 41, 48, 60–61, 80–81, 87, 95, 105, 107, 113, 122, 136, 156, 168–169, 171–174, 179, 185, 225, 276–277, 281–282, 285, 287, 337–338, 343, 412, 415, 425–426 praxis (social action)â•… 60 premodernâ•… 48 presence-at-handâ•… 60 protocolâ•… 33, 171, 348, 414, 426 psychology, ecologicalâ•… 279, 287, 378 publish, publishingâ•… 347, 349– 350, 412, 428, 430, 432–433 R ready-to-handâ•… 59 recordingâ•… 423, 429 representationâ•… 1, 3–4, 6, 8, 10, 13, 30, 37, 51, 53, 57, 64, 68, 73–76, 77–79, 83, 86–88, 94, 102, 109–111, 115, 168, 170, 173–174, 176, 185, 245, 271, 277, 354–355, 363, 369–370, 377, 381–385, 390–291, 396–398, 402, 405–406, 413, 415 research, researcherâ•… 9, 50–52, 55–56, 61, 63, 65–68, 74, 78, 81, 100–109, 111–116, 118–119, 122–125, 297, 350, 386, 403, 411–417, 421, 424–428, 430–432 reservationâ•… 17, 26, 31–34, 267, 331, 343, 348–349, 415–416, 420, 430, 432–433 ritualsâ•… 64, 84, 86–87, 151, 167–169, 172, 333–334, 343, 414, 422 riverâ•… 29, 31, 33, 35, 40–41, 49–50, 57, 59, 64, 89–90, 92, 95, 130, 149, 164–165, 218–219, 228–229, 231–234, 239–242, 244–245, 247, 249–252, 254–259, 261–266, 270, 275, 278–281, 283–287, 292, 294– 296, 298–299, 301–305, 308, 310, 313, 316, 318–319, 321–322, 345, 356, 358, 366, 383, 385, 387, 389–390, 395, 402, 415

S sacredâ•… 32, 35, 41, 55, 60, 84, 86, 114, 173, 248, 300, 312, 319–322, 330, 332–333, 335–337, 344–345, 347–349, 356, 358, 372, 412 sacred place, sacred siteâ•… 30, 41, 44, 58, 91, 272, 298, 365 salience, salientâ•… 58, 126, 128, 130–135, 137, 144, 149, 155–156, 251–252, 259, 275–276, 284– 287, 307, 329 scienceâ•… 4, 53, 55–56, 65, 67, 76, 179, 264, 371, 373, 396–398, 413, 415–416, 418, 425, 431 secret/sacredâ•… 412 semantic, semanticsâ•… 4–6, 10, 38–39, 53, 77, 82, 109–110, 118, 126, 151, 157, 229–230, 232–233, 235, 251, 254–255, 257, 261, 264–265, 272, 301, 320, 369, 373, 375–376, 381, 383, 388, 391, 404 semigenericsâ•… 329 semplatesâ•… 6, 39, 254, 257 sense, sensesâ•… 2, 77, 179, 185, 218, 248, 271, 279, 281–282 sense of placeâ•… 7, 11, 55, 58, 63, 68, 80, 137, 267, 272, 357, 396 sensoryâ•… 2, 112, 168, 170, 279 Smith, Lindaâ•… 56, 81, 426 social structureâ•… 104, 291 song, songs â•… 61, 78, 84, 169, 180, 182, 261, 264, 269–272, 333, 335, 347, 422, 429 space, spacesâ•… 2–3, 11, 30, 40, 48, 55, 57, 60–62, 76, 80, 82–84, 87, 104–107, 115–116, 123, 168, 174, 179, 184, 190–193, 196–197, 199, 205, 208, 210, 214, 218–221, 243, 254, 258, 275, 278–279, 287, 328, 353, 355, 357–358, 391, 395, 403, 405, 419, 433 speech communityâ•… 8–9, 54, 56 spirit, spiritsâ•… 11, 31, 39, 41, 49, 61, 74, 82, 87, 92, 126, 132, 173, 192, 196, 220, 267, 270, 276, 280, 283–284, 287, 345, 422 spiritual, spiritualityâ•… 11, 25, 31, 35–36, 41, 49–50, 54, 57, 63,

67–68, 86, 95, 105, 114, 122, 126–127, 132, 137, 172–174, 192, 220–221, 267, 281–283, 287, 311, 319, 375, 378, 385, 412–414, 431 story, storiesâ•… 7, 32, 35–36, 61–62, 64, 87, 91–92, 97, 109, 150–152, 171–172, 180, 185, 191, 193, 196–197, 207, 217, 258, 261, 264–265, 267–273, 319, 321, 323, 327, 331, 333–336, 344–347, 349–350, 353–360, 362–366, 382, 386–389, 422–423, 429 storytellers, storytellingâ•… 185, 190, 267–269, 271, 345, 355, 360, 366 subjectâ•… 54, 56, 58, 60, 76, 80, 86, 111, 134, 227, 280, 334, 370, 373, 388, 412 subjectivityâ•… 54, 113 T temporalâ•… 57, 80, 94, 114, 377, 399 temporalityâ•… 157 timeâ•… 26, 50, 61–62, 69, 75, 82–83, 87, 95, 114, 130, 152, 167, 170–172, 175–176, 184, 192, 207, 217–218, 220, 242, 252, 267, 270–271, 279, 284, 328, 330, 357–359, 374, 391–392, 433 tjukurrpaâ•… 47, 61–63, 69 topography, topographicâ•… 9–11, 31, 39, 49, 54, 62, 65, 123, 128–129, 132–135, 173, 177, 179, 228, 231, 299–301, 315, 333, 399–400 toponymsâ•… 1, 3–4, 8, 10, 25, 35, 39, 41–42, 47, 109–110, 114, 118, 144, 149–155, 160–165, 192, 232, 235, 246, 275, 314, 319, 322, 329, 333, 336, 338, 382–383, 386 see also place names topophiliaâ•… 1, 11, 58 traditionâ•… 1, 3–4, 12, 28, 58–59, 64, 66, 74, 82–85, 89, 92, 137, 146, 169–170, 174, 248, 269, 343–344, 353–356, 365, 384, 391



traditionalâ•… 3, 7, 12, 28, 30, 32, 35–36, 40–41, 47–48, 56, 87, 91, 124, 161, 164, 191, 219, 240, 250, 252, 257, 265, 270, 272, 275, 287, 291, 294, 296, 301, 311, 314, 320–323, 330–332, 338, 343, 355–358, 384–385, 432 transcribing, transcriptionâ•… 28, 182, 194–195, 336, 386, 418, 428, 432 transculturalâ•… 11, 63 transdisciplinaryâ•… 9, 13, 48, 51–52, 54–55, 65, 67, 376 U utilitarianismâ•… 424

Index 449

V vegetationâ•… 2, 10, 27, 30, 33, 40, 124, 127–129, 132–135, 157, 160, 195, 228, 248, 292, 297–298, 307–308, 310, 315–321, 382, 399–400 videoâ•… 180, 182, 184–185, 272, 349, 414–415, 423 viewscapeâ•… 291 W waterâ•… 2, 10, 27, 29–31, 34, 40, 49, 58–59, 61, 89, 92–93, 124, 130–131, 133, 135, 143–146, 148–151, 158–165, 170–171, 191, 195–196, 199–202, 205, 207–214, 217–219, 231, 234, 245–249, 252, 255, 258, 267,

276, 283, 291–292, 297–298, 301–305, 310, 314, 316–317, 320, 329, 361–362, 387, 390, 396, 399–400, 402–403 water bodiesâ•… 92, 128, 130–131, 161, 370, 373, 375, 382, 385, 388 watercoursesâ•… 11, 27, 33, 40, 89, 158–160, 231 wayfindingâ•… 35, 105, 109, 332–333 wisdomâ•… 117, 427–428 worldviewâ•… 8, 12, 54, 56–58, 60, 64, 82, 85, 95, 174, 373 Y Yindjibarndiâ•… 7–8, 11–12, 25–31, 34–35, 38–42, 48–50, 57, 61, 69, 371, 375, 430