Heterogeneity in Word Formation Patterns: A corpus-based analysis of suffixation with -ee and its productivity in English

Heterogeneity in Word–Formation Patterns

Studies in Language Companion Series (SLCS) This series has been established as a companion series to the periodical Studies in Language.

Editors Werner Abraham University of Vienna

Elly van Gelderen

Arizona State University

Editorial Board Bernard Comrie

Max Planck Institute, Leipzig and University of California, Santa Barbara

William Croft

University of New Mexico

Östen Dahl

University of Stockholm

Gerrit J. Dimmendaal University of Cologne

Ekkehard König

Free University of Berlin

Christian Lehmann University of Erfurt

Brian MacWhinney

Carnegie-Mellon University

Marianne Mithun

University of California, Santa Barbara

Heiko Narrog

Tohuku University

Johanna L. Wood

University of Aarhus

Volume 118 Heterogeneity in Word–Formation Patterns. A corpus-based analysis of suffixation with -ee and its productivity in English by Susanne Mühleisen

Heterogeneity in Word–Formation Patterns A corpus-based analysis of suffixation with -ee and its productivity in English

Susanne Mühleisen University of Bayreuth

John Benjamins Publishing Company Amsterdam / 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 Muehleisen, Susanne. Heterogeneity in word-formation patterns : a corpus-based analysis of suffixation with -ee and its productivity in English / Susanne Mühleisen. p. cm. (Studies in Language Companion Series, issn 0165-7763 ; v. 118) Includes bibliographical references and index. 1. English language--Suffixes and prefixes. 2. English language--Word formation. I. Title. PE1175.M84   2010 425’.92--dc22 2009050675 isbn 978 90 272 0585 8 (Hb ; alk. paper) isbn 978 90 272 8838 7 (Eb)

© 2010 – 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

Acknowledgments List of tables and figures List of abbreviations chapter 1 Introduction: Polysemy, heterogeneity and ambiguity in word-formation patterns 1.1 Theoretical and methodological approaches in this study  7 1.2 Outline of chapters  11 1.3 Aims and scope of this study: Wider implications  16

ix xi xiii

1

chapter 2 Phonological, syntactic and semantic constraints 19 on the formation of -ee words 2.1 Phonological limitations to -ee verb derivation   21 2.2 Syntactic limitations of -ee as a verb derivation  27 2.2.1 Verb derivation with -ee: Syntactic properties  33 2.2.2 Syntactic ambiguity in -ee verb derivation  37 2.2.3 Syntactic hypotheses of -ee word-formation: A critical review  41 2.3 Semantic explanations: -ee formation in a thematic role framework  44 2.3.1 Lexical conceptual structure and co-indexation of affixes  54 2.3.2 Ontological classification and prototypes: The question of categories  55 2.4 Conclusion: Heterogeneity, polysemy and ambiguity revisited  58 chapter 3 The career of -ee words: A diachronic analysis from medieval legal use to nineteenth-century ironic nonce words 3.1 Towards a contextualized history of -ee suffixation  62 3.2 Anglicized Law French beginnings  66 3.3 English indirect object formations in the sixteenth century  69

61

vi

Heterogeneity in Word-Formation Patterns

3.4 3.5 3.6 3.7

Growing diversity in the seventeenth century  72 Recession and continuations in the eighteenth century  76 Abundance in the nineteenth century: -ee words go popular  81 Changes across the centuries: Simultaneity and ambiguity  88

chapter 4 Morphology and the lexicon: On creativity and productivity of -ee words 4.1 Word-formation between morphology and the lexicon  92 4.2 Productivity and creativity in word-formation  93 4.3 Types of productivity: Synchronic and diachronic processes  100 4.4 Productive patterns and features of twentieth-century -ee words  108 4.5 Among the new words: Nonce words, neologisms and processes of establishment  114 4.6 Actual words and possible words: Towards an empirical study  117 chapter 5 A corpus-based analysis of 1,000 potential new -ee words 5.1 Web data for linguistic purposes: Searching corpora and searching the Web  124 5.2 Searching the Web for -ee words: On methods and procedures  127 5.3 Syntactic and semantic patterns of successful neologisms  135 5.4 On ‘failees’: Ineffective patterns of -ee words  147 5.5 Hapax Legomena and their production in various text types  149 5.6 New productivity? On data collection and productivity measurement  162

91

121

chapter 6 165 -ee words in varieties of English  6.1 Suffixation with -ee in American English: A comparison of historical and contemporary usages  169 6.2 Australian English usage: On hypocoristics and -ee words  175 6.3 A corpus-based analysis of some new -ee words in American, British, Australian, New Zealand and Indian Englishes  179 6.4 Global meets local: Variation and change of -ee word production in varieties of English  184 conclusion On the study of an individual word-formation pattern: General and particular implications

189



Table of contents vii

Works cited

193

appendix 1 Documentation of established -ee words with their citation sources: A comparison (in alphabetical order)

201

appendix 2 Quantitative analysis of 1,000 potential -ee words (Web-search, February–June 2005)

215

Name index Subject index

241 243

Acknowledgments

This study started out as a rather small and playful project, but has gradually become much larger and more fundamental than I had originally intended it to be. In the course of the development of this book, there are a number of people without whom its progress and completion would not have been possible. My first and foremost thanks goes to Edgar W. Schneider who accompanied this project supportively and encouraged its completion as a postdoctoral thesis (Habilitationsschrift) at the University of Regensburg. I also thank my mentors Roswitha Fischer and Björn Hansen (University of Regensburg) for their critical guidance and encouragement. In the initial stage of the study, my colleagues Bettina Migge and Raimund Schiess were patient and encouraging listeners to my preliminary ideas about the subject. In this phase and many years before, I had the good fortune of being part of Marlis Hellinger’s team at the Johann Wolfgang Goethe University Frankfurt and I would like to thank her wholeheartedly for her support. I am very grateful to Ingo Plag for a thorough and critical reading and many helpful and muchneeded suggestions to help improve the study in the closing phase of the project. During a conference discussion, Dieter Kastovsky provided me with encouraging comments and very useful material. I am indebted to the late Michael Noonan for taking an interest in my study, and to Elly van Gelderen and Werner Abraham for their highly cooperative communication as editors of this series. An anonymous reader provided many helpful comments. All errors and shortcomings are, of course, my own. As always, Kees Vaes has been a pleasure to work with and I would like to thank him for his kind and professional support. I am grateful to Richard Edge (University College Dublin) for his meticulous reading of the manuscript. Members of my team at the University of Bayreuth, notably Eric Anchimbe and Hanna Strass, provided me with a lot of help in the final stages of preparing the manuscript. This study would not have been possible without the enduring encouragement and support even in times of doubt from Tobias Döring.

List of tables and figures

Table 1. New -ee words formed with verb base ending in -ate Table 2. Unsuccessful test words with verb base ending in -ate Table 3. New -ee words with vowel end limitation Table 4. Word list – fourteenth and fifteenth century (OED entries) Table 5. Word list – sixteenth century (OED entries) Table 6. Word list – seventeenth century (OED entries) Table 7. Word list – eighteenth century (OED entries) Table 8. Word list – nineteenth century (OED entries) Table 9. Summary of the development of Direct Object nouns in -ee formation Table 10. The growth of productivity of affixation with -ee, based on entries in the OED (14th to 20th century) & various sources for the 20th century Table 11. Comparison of 19th and 20th century syntactic patterns in -ee words according to Bauer (1994: 41, 46) Table 12. Twentieth century -ee words from various sources (according to date) Table 13. Twentieth century -ee words from various sources (not dated) Table 14. Test word analysis according to frequency categories Table 15. New -er words (as a result of the collocation test) Table 16. 120 ‘frequent’ new -ee words Table 17. Patterns in ‘frequent’ new -ee words Table 18. 252 unsuccessful -ee test words Table 19. 214 Hapax Legomena Table 20. Domain-specific Web search 10/2002 Table 21. Domain-specific Web search 03/2006 Table 22. Development of trainee and retiree in % (.com = 100) Table 23. Development of mentee in % (.com = 100) Table 24. Development of refugee in % (.com = 100) Table 25. Distribution of retiree, advisee and trainee on Australian, New Zealand and Irish websites (.com = 100 %) Table 26. Selected new -ee words in varieties of English

xii Heterogeneity in Word-Formation Patterns

Figure 1. Continuum of thematic relations (Portero Muñoz 2003: 139) Figure 2. Prototypical and marginal patterns in -ee formation Figure 3. English dialects, Latin, French and Standard English in varieties of written and spoken English (Görlach 1999: 462) Figure 4. Diachronic increase in number of lexemes (Nevalainen 1999: 339) Figure 5. Direct object formations in new -ee words between the sixteenth and the nineteenth century (in %) Figure 6. Increase of -ee word-formations (14th–20th century) Figure 7. Proportions of test words according to frequency patterns Figure 8. Collocations versus no collocations with -er words Figure 9. Collocations with -er according to frequency categories

List of abbreviations

App. BNC C

= Appendix = British National Corpus = Corpus item, i.e., one of the words tested in my corpus study (see Appendix 2). The corpus items are numbered in alphabetical order, e.g.:

C1 C2 C3 C4 C5 C6 C7 C8 C9 C 10 …

Abhorree Abidee Abjectee Abolishee Abrogatee Absorbee Acceptee Acclaimee Accommodatee Accursee …

C 991 C 992 C 993 C 994 C 995 C 996 C 997 C 998 C 999 C 1000

Workee Woundee Wrappee Wreckee Wrestlee Wrongee Yammeree Yearnee Yieldee Yodellee

EModE = Early Modern English F = French ICE = International Corpus of English ME = Middle English NED = New English Dictionary NPC = No Phrase Constraint ODNW = Oxford Dictionary of New Words OE = Old English OF = Old French OED = Oxford English Dictionary UBH = Unitary Base Hypothesis WWW = World Wide Web

chapter 1

Introduction Polysemy, heterogeneity and ambiguity in word-formation patterns

One of the most persistent problems with postulated word-formation rules is their exclusion of formations that can nevertheless be found in actual usage. An example of a particularly heterogeneous word-formation pattern in English is the formation of nouns by suffixation with -ee. In its most typical guise, a word of this kind (e.g. interviewee) is formed with a verb that can take an object (‘interview someone’) and often co-occurs with a complex noun of the same base suffixed with -er (e.g. interviewer). The following dialogue is taken from the results of a Web-corpus investigation of neologisms of -ee formations which is part of the present study. The nature of the data collection, its goals, size and analysis will be explained in more detail in Chapter 2 and Chapter 5. In the example below, we can see the coinage of such a model new -ee word in the making: beseechee C 62 Writer A: nothing like a good beseecher Writer B: … and of course a good beseechee Writer B: hehee Writer A: whatever that means Writer B: the one who is being besought … of course Writer B: beseeched Writer B: the recipient of the beseeching Writer A: these linguistic gymnastics are exhausting me Writer B: *dismayed* sorry  (www.avocadosalad.org, 2002)

. The number 62 identifies beseechee as corpus item 62, i.e., C 62 (cf. also List of abbreviations and Appendix 2). All websites cited were accessed between February and July 2005 unless otherwise stated, the exact date of access within this time period is not given for each individual Web citation. Typos and idiosyncratic orthography choices are maintained.



Heterogeneity in Word-Formation Patterns

As the conversation above suggests, the new word beseechee not only co-occurs with beseecher, but the presence of the latter seems to trigger the new coinage – a phenomenon which has been referred to as “paradigm pressure,” i.e., “paradigmatic forces which affect the coining of new words” (Bauer 2001: 71). While the logic of the sense relation of a converse antonym is compelling in many -er and -ee suffix pairs – after all, an employer needs an employee in the same way as an advisee is dependent on an adviser, or a trainer is only complete with a trainee – there are also numerous examples of -ee words where this pattern does not exist: retiree, for example, neither has a *retirer as an opposite concept, nor does it seem to refer to the recipient of the action of retiring in the same way as beseechee is “the recipient of the beseeching.” Similarly, there is no *divorcer to the divorcee, no *interner to the internee, etc. As we shall see below, there are many more “deviant” cases of -ee formations that have served as sources of mystification in a number of linguistic discussions on the “apparently strange conditions on the productivity of these affixes” (Plag 2003: 45). This chapter aims at providing a first overview of the heterogeneity of this word-formation pattern. After this initial problematization, it will offer a number of theoretical and methodological perspectives which are to be explored in the subsequent chapters. The standard textbook view tends to describe -ee suffixation as verb-based rather than noun-based (cf. Bauer 1983; Katamba 1994), despite some instances – arguably also biographee – where no verbal base existed at the time of the new formation (see also discussion in Section 2.2). Katamba’s (1994: 65) definition of -ee suffixation as “(passive) person who undergoes action indicated by the verb” sums up a number of customary conceptions about this derivation type – (a) passivity, (b) personhood, (c) verbal base – which have to be questioned on account of the following examples: (a) In the following established and new -ee words, the referent is not “passive”: absentee: (1) a. One who is absent, or away, on any occasion. b. absent voter. Also (chiefly U.S.) absentee vote, voter, voting. (2) One who systematically stays away from his country or home; a landlord who lives abroad. (OED) escapee: One who has escaped; esp. (a) an escaped convict from a penal settlement; (b) an escaped military or political prisoner (OED).



Chapter 1.  Introduction

expressee C 291: One who expresses an opinion, as in the example below: “LEGAL DISCLAIMER: The stories, views and opinions expressed on this site are solely those of the expressee.” (www.groups.msn.com, 2005) survivee C 885: One who has survived (a catastrophe, an accident), as in the following example: “Question is: Do you offer a Titanic survivee ice for his whiskey?” (www.deathlist.net/forums/index, not dated)

(b) The idea that -ee words necessarily have human referents can also be found in statements like the following: “So what do ee words have in common? Only one characteristic is agreed on by everyone: ee refers to a person. An abusee could not be a maltreated parrot or a broken window. This is true of old coinages, or any of the 200 or so new 20th century ee words” (Aitchison 1999: n.p.). The examples below, from various sources (Bauer 1987; OED; my corpus items) show that pets, farm animals (i) or linguistic and technical components (ii) can indeed be possible referents of -ee nominalizations: Examples (i): animate but non-human: trainee: 1841. First applied to an animal undergoing training, later extended to human reference. 1850. “The trainers first double up one of his fore legs, which they bind fast with a cord; this they pull, and thus compel the trainee to come down upon his bent knee”. (Fraser’s Mag. XLI 658, cited in OED) vivisectee: 1886. Animal undergoing vivisection. “Whether any attempt at the absolute prohibition of vivisection would not react to the disadvantage of the unhappy ‘vivisectees’.” (Pall Mall G. 3 June 5/2, cited in OED) brushee C 75: One who is brushed (pet) “Some super Maltese brushers lay the brushee on his sides and/or back, but standing and sitting are the usual brushing positions for me and mine.

. As already explained in Note 1, some of the examples given here are not listed in established dictionaries but have been found in my web-search (Chapter 5) and can be taken as examples of mostly recent word-formations with -ee.





Heterogeneity in Word-Formation Patterns

Generally, I use a grooming arm to steady the dog, but sometimes brushing is easier without it.” (www.malteseonly.com/brushouts.html, 2004) cleanee C 92: One who is cleaned, including non-humans: “This benefits both fish – the cleaner gets a meal and the cleanee leaves as a healthier fish.” (richard-seaman.com/Underwater/Belize/OtherFish/index.html, not dated) milkee C 470: One who is milked (farm animal) “If you get dairy goats, you’ll probably want to build this nifty stanchion to make milking easy for both the milker and the milkee.”  (www.motherearthnews.com/Do-It-Yourself/2002-06-01/Milking-Stanchion) Examples (ii): inanimate and non-human possessee: 1982 David Gil “Case Marking, Phonological Size, and Linear Order” Syntax and Semantics 15: 122 “Possessive constructions in which it marks the subject possessee phrase … are used by all but the die-hard purists”. (Bauer 1987: 318) dependee C 159: Matching component in a technical operation: “If a match is found, a persistent dependency is created between the matching component, the “dependee,” and the calling component, the “dependent”.” (Docs.sun.com, not dated) surfee C 881: Notice the balance between the length of the timer and the time it takes pages to load. As you surf you may like sites that only require 10 seconds before you can surf again, but this does not benefit pages that load slowly. So observe not only from the surfer perspective but also from the “surfee” point of view. (www.ebizyourhome.com/articles/traffic_exchanges.html not dated)

(c) The third condition cited above, “action indicated by the verb,” is also questionable. In the following examples, the verbal base of the -ee formations is nonexistent, at least not in English: . It could be argued here, of course, that the webpage visited (i.e., the “surfee”) stands metonymically for the (human) creator of the webpage. As personification of machines is rather common, a final decision on the “person” or “non-person” status of surfee is difficult to make here. Please note that surfee has various possible meanings (cf. discussion in Section 2.3).



Chapter 1.  Introduction

lessee: 1481. One to whom a lease is granted; a tenant under a lease (OED). “Lessees … [shall] fynde goode and suffycient suertie” 1495. Act 11 Hen. VII, c. 9 §2, cited in OED. refugee: 1685. A fugitive; someone who seeks refuge in a foreign country (cf. OED). benefactee: 1982. A description of a grammatical role in linguistic terminology “like NP1 it is not agentive, and it is typically affected by the verb, generally as a benefactee or malefactee” (David Gil, “Case marking, phonological size, and linear order,” Syntax and Semantics 8: 315, in Bauer 1987). therapee C 903: 1999. Someone undergoing therapy. “The seating was slightly different, two simple chairs nudged close together like therapist and therapee.” (www.guardian.co.uk/world/1999/mar/05/clinton.usa2)

Such initial “deviant” examples may serve as an indication that the formation and interpretation of -ee words are possibly more diverse than can be captured by a single word-formation rule. Recent work in morphology and lexical semantics has taken a closer look at the heterogeneous and polysemous nature of derivations with -ee (cf. Barker 1998), sometimes in combination with the more widely explored -er suffix (e.g. Baeskow 2002; Booij & Lieber 2004; Lieber 2004). While approaches and interests in this derivational affix vary, it is undisputed that the diversity in formation patterns, meanings and interpretations of -ee words poses a challenge both for the idea of word-formation rules in particular and for the relationship between lexical semantics and word-formation in general. The derivational affix -ee makes for a prime example in exploring issues of heterogeneity, polysemy and ambiguity in word-formation. Rochelle Lieber, in her introduction to Morphology and Lexical Semantics (2004: 2), points out that lexical semantics of word-formation (as opposed to words) has so far been neglected, and she poses a number of questions that should

. Note that this is also an instance of “non-passivity.” . Note that this is also an instance of “non-person reference.”





Heterogeneity in Word-Formation Patterns

be further explored, among them the polysemy question, i.e., why does a particular affix in English have different meanings? As we could see from our initial examples cited above, the polysemy question is also an interesting one to explore in -ee suffixation. Altogether, our preliminary investigation of the diversity of -ee words raises questions in three areas: – Firstly, polysemy is certainly one of the most striking issues to consider: why does the affix -ee sometimes create patient nouns (employee, interviewee) and othertimes agent nouns (retiree)? Does the affix have any unitary core of meaning at all, and if so, what is it? – Secondly, the actual production process also raises the question of heterogeneity: which types of bases (verb, noun) does the affix attach to? What influences the choice of base? What role does analogical coinage play in the production process (e.g. in our initial example of beseechee)? – Thirdly, it would be interesting to explore how the meaning of -ee words is interpreted from the user’s perspective, and where multiple meaning possibilities might lead to ambiguity in the interpretation process. Ambiguity in individual words, as it is understood here, is necessarily related to the polysemy of the word-formation, i.e., the fact that the -ee suffix has various possible meanings can result in interpretations of, for instance, the established word abandonee perhaps meaning ‘one to whom anything is formally or legally abandoned’ (OED) as well as ‘someone who has been abandoned’. Similarly, a more recent word from my corpus, faintee (C 298), might be interpreted as ‘someone who faints’ or ‘someone who is fainted at’ (cf. Section 2.2.2 for a more detailed exploration of syntactic ambiguity in -ee word-formation). The fact that -ee can sometimes compete with -er might also lead to an uncertainty in the choice of suffix for a new word, as can be seen from the following example on the aforementioned faintee:

. The other questions in Lieber’s (2004) list are: secondly, the multiple affix question: why are there several affixes (e.g. -er and -ant for agent nouns) which perform the same functions?; thirdly the zero-derivation question: how do we account for semantic change without formal change in word-formation?; and lastly the semantic mismatch question: why is the correspondence between form and meaning in word-formation not always one-to-one? While most of these could also be applied to -ee suffixation (obviously except for the zeroderivation question), I will concentrate on polysemy as the starting point of my investigation of -ee formation.



Chapter 1.  Introduction

faintee C 298: Eventually the aforementioned fainter (faintee?) regained consciousness, was helped off the train by a couple of staff members and we proceeded on our way. (www.roberthampton.me.uk/news/archives/2004 07)

There are, in fact, good reasons why the fainter may be realized as faintee, as we shall see in Section 2.3 on semantic constraints on -ee word-formation. There are also good reasons why the abandonee is more likely to be interpreted as the direct object of the verb (‘someone who has been abandoned’) today even though the original meaning (1848) is given as an indirect object formation of the type ‘one to whom something is Ved’ (cf. also Chapter 3 on the historical development of ee words). Before further investigating the versatility of -ee formation on the basis of a discussion of phonological, syntactic and semantic constraints (Chapter 2), I would now like to outline the theoretical and methodological approaches which will be used in this study.

1.1

Theoretical and methodological approaches in this study

There are, certainly, more complex scholarly descriptions of -ee word-formation than our first discussion of “passivity,” “personhood” and “verbal base” of -ee words may have suggested. In the 1980s and 1990s, notably Bauer (1983, 1994) and Barker (1998) have given rather detailed syntactic and semantic characterizations of -ee formation patterns and described constraints on their productivity. Whereas Bauer’s analysis aims at specifying the various syntactic patterns of -ee as a deverbal nominalization, Barker argues that syntactic descriptions do not cover the word-formation constraints of -ee, and, instead, proposes a semantic analysis of this word-formation. Barker contends that -ee formation can be characterized by three semantic features: i.e., sentience, lack of volition, and episodic linking of the word to the denotation of its stem; he even proposes that these features in -ee form a separate thematic role. More recently, Baeskow (2002), Portero Muñoz (2003), Lieber (2004) and Booij & Lieber (2004) have looked at -ee formations as a particular challenge to semantic description, often in combination with the almost equally heterogeneous -er derivations. Baeskow (2002) discusses the -ee suffix as part of a contrastive analysis of derived person references in English and German in the framework of her minimalist word-formation theory. In their effort to explain deviant examples of -ee and -er words, Booij & Lieber (2004) use a framework of lexical semantic representations (Lieber & Baayen 1997) to ascribe semantic content to





Heterogeneity in Word-Formation Patterns

these suffixes in the form of the features [+material, +dynamic]. Their contrastive (English-Dutch) approach seeks to explain why Dutch seems to lack a specific process of word-formation analogous to -ee in English. Lieber uses the suffixes -ee and -er which “present substantial descriptive problems in other frameworks of semantic analysis” (Lieber 2004: 12) to develop her own framework of lexical semantic description. An aim of Lieber’s outline is to illustrate that polysemy in complex words must be treated like polysemy in simplex lexical items. While it is undisputed that the above treatments of -ee have provided much valuable insight on some of the constraints in the production of -ee words, there are quite a few phenomena which do not fit in either a compact syntactic or a semantic framework. On the basis of a critical review of Bauer (1983, 1994), Barker (1998), Baeskow (2002), Portero Muñoz (2003) and Booij & Lieber (2004) in Chapter 2, I will argue that the many heterogeneous patterns of this wordformation process cannot be accounted for by a single homogeneous framework alone. Rather than contributing to yet another model for the precise compositional semantic description of the suffix, I would like to look at the polysemy of this word-formation pattern from a different perspective. The theoretical part of this perspective includes: firstly a diachronic view, secondly, it comprises a discussion on analogical coining, semantic networks, and the co-occurrence between word-formation patterns (-ee and -er), and thirdly, it takes into account influences through language contact. The methodological part of my overall study seeks to challenge previous measurements of productivity and proposes a new way to establish the relationship between actual and possible words. Rather than relying on random word collections, for instance, from newspapers (as in Barker 1998; Bauer 1983, 1994), or on dictionaries and reverse dictionaries (Baeskow 2002), this study makes use of the largest and most up-todate electronic corpus, the World Wide Web as a new and adequate tool to search for neologisms (cf. Chapter 5).

(a) Theoretical considerations My theoretical discussion will support is the notion of a continuum of “typicalness” in -ee formations, i.e., there are more prototypical and more marginal examples of -ee words when looked at from a synchronic point of view. I will argue that the “fuzzy edges” to this word-formation, i.e., the many exceptions, especially to the previously proposed compact models, are not random accidents or inexplicable deviations. Rather, as it will be emphasized, they are part of the complex history of this word-formation, where several syntactic and semantic processes have operated in the course of the 600 or so years of its existence. In contrast to



Chapter 1.  Introduction

almost all previous descriptions of -ee formation patterns, which make their case from a synchronic point of view alone, this study advocates a diachronic perspective on the formation of -ee words in order to account for the heterogeneity and simultaneous presence of several word-formation processes which have evolved in the course of the history of -ee suffixation. My analysis also takes seriously the perspective of language change and its social origins and motivations in word-formation. By offering a link between a synchronic and a diachronic point of view, it aligns itself with recent claims (e.g. Bauer 2001; Fischer 1998; Lass 1997; Romaine 1998) that the historical development of word coinage has often been neglected in the strict separation between synchronic and diachronic positions on language. In her introduction to The Cambridge History of the English Language, Vol. IV, Suzanne Romaine (1998: 4), quoting Strang’s (1970: 19), writes that “at every stage the history of a language must be studied in the light of its use in the world,” thus emphasizing the fact that every language has an internal and external history. Romaine expresses regret that scholars generally treat these two aspects of languages as separate enterprises often placing more importance on pursuing processes of internal evolution over external factors: Traditional histories present the language as changing largely in response to internal linguistic pressures. Language history is viewed as a series of changes with little attempt to answer the question of who originated them and what motivation others might have had for adopting and spreading them. These questions about the social origins and motivations for change naturally become harder to answer the further back in time we go, but have become increasingly difficult to ignore in the context of the greater understanding modern sociolinguistic research has (Romaine 1998: 4) yielded […]. 

This study, therefore, will not only look at the different stages of -ee formation with a detailed analysis of the neologisms produced for each century, but will also explore the reasons for and the motivations of change. Furthermore, it will look at productivity patterns both from a diachronic and a synchronic point of view.

(b) Methodological considerations Particular emphasis will be placed on the details of data collection and documentation. One of the shortcomings of previous investigations is that some of the analyses (Bauer 1983; Portero Muñoz 2003) rest on rather thin and arbitrary . Baeskow (2002) is an exception to this lack of inclusion of a diachronic perspective. . In Bauer’s case this is also determined by the genre of the text, i.e., Bauer (1983) is a more general textbook on English word-formation and not a separate study on -ee formation.



10

Heterogeneity in Word-Formation Patterns

data; in Bauer (1994: 40–47) some more recent examples of his own random collection of -ee words (cf. also Bauer 1987, 1993) are added and analyzed according to the patterns they seem to follow. Furthermore, as Bauer (1983: 291) himself notes in his summary, “it is far easier to justify the use of a particular word-formation process post hoc than it is to predict which process will be used before the event” – a limitation which weighs even more heavily if the analyses are applied to just a few selected cases. While Barker’s (1998) investigation is more detailed and resourceful – he presents quite a number of new examples of -ee words assembled from primary and secondary sources – little attempt is made to compare new or very recent productivity patterns with older ones, or to single out any characteristics of the more recent cases of -ee formation. Since -ee suffixation is used merely as an example to explain larger theoretical frameworks, the data on -ee words is not highlighted in Baeskow (2002), Lieber (2004) and Booij & Lieber (2004) and the examples they use are from established sources like dictionaries and reverse dictionaries. Baeskow (2002: 57) also acknowledges the unfeasibility of making predictions on the productivity of suffixes with such traditional tools as they do not include words with predictable meaning and consist of exclusively lexicalized derivates. None of the studies cited above has engaged in any structured, corpusbased investigation of the productivity of either historical data or new and recent -ee words. This study seeks to close this research gap by testing the productivity of 1,000 potential -ee words which have not previously been documented in any other data source. I have chosen this method of testing a random list of potential -ee words because, unlike in coded corpora, a search for *ee# on the WWW is not possible. My list of 1,000 potential words was created by analyzing the OED and selecting those bases for which, according to my sources (see Appendix 1), no -ee derivation had been listed. Until the time of data collection for this study, in 2005, approximately 500 -ee words had been documented altogether in various sources in the course of the 600 or so years of the existence of the study (cf., my detailed word lists in Chapter 3 and Chapter 4, and documentation in Appendix 1).10 My analysis on the World Wide Web will provide evidence that the productivity of this word-formation pattern may be significantly higher than assumed, both in ad-hoc creations and nonce words, as well as in formations which seem to be already well established in usage but not documented in dictionaries or other public sources. . A similar figure is also given in Barker (1998: 701). However, the figures, which he cites for the occurrence in various centuries (without documentation), add up to merely 386. 10. Appendix 1 also documents the sources of these findings, for instance, dictionaries like the OED, reverse dictionaries and scholarly work like Bauer (1983, 1987, 1994) and Barker (1998).



Chapter 1.  Introduction

The use of the Web as a corpus has received some attention and gained in popularity in recent research for a number of reasons which will be dealt with more thoroughly in Chapter 5 (cf. also Hundt et al. 2007; Kilgarriff & Grefenstette 2003; Resnik & Smith 2003). This investigation has purposely made use of the World Wide Web (websites in English) as an unfiltered data source for a number of reasons, among them the variety of diverse text types and the global connectedness of the data and its users. By employing this uncoded source, the study also takes seriously the input of various varieties of English to the contemporary production of -ee words, including the possibility of non-native speaker input. While the (likely) inclusion of non-native speaker creations in the study of neologisms in English may be seen by some as distorting the value of the results, I would like to argue that non-native speaker contributions in a globalized English have been a reality for a long time and cannot be ignored. I would also like to point out that the linguistic biographies of speakers/writers in established corpora are also often not known. The analysis of the very new and recent -ee words will reveal the predominant syntactic and semantic patterns, and will also acknowledge the simultaneous presence of different patterns and a resulting heterogeneity or even ambiguity in different text types and different varieties. Both in the historical analysis and the investigation of contemporary patterns of -ee usage, I will highlight that this word-formation pattern has been influenced by language contact between different languages and varieties. The analysis will again focus on the usage of -ee words in different varieties of English, for example, different patterns of usage in British and American English, as well as in “smaller” varieties such as Australian and New Zealand English, English in India, etc. The fact that, in Web usage, these varieties are often difficult to pin down and are subject to international interaction in communication (especially in chats or blogs but also online newspapers, etc. which transcend the national/local level of communication) is not seen as an obstacle but rather as programmatic for the study of contemporary new -ee words as a phenomenon of global English with a common core.

1.2

Outline of chapters

Chapter 2 takes a critical look at earlier syntactic and semantic descriptions of -ee nouns (e.g. Baeskow 2002; Barker 1998; Bauer 1983, 1987, 1993; Booij & Lieber 2004) and the statements made with regard to the phonological, syntactic and semantic constraints on the productivity of this word-formation. Throughout this chapter, examples from both earlier documented data and from my corpus search are drawn upon in order to illustrate my points.

11

12

Heterogeneity in Word-Formation Patterns

At first (Section 2.1), phonological limitations to -ee verb derivations are discussed. Previously established phonological restrictions (e.g. the -ate deletion rule, e.g. in nominee, Bauer 1983: 247) are questioned on the basis of old and new examples of -ee formations. Another phonological (and sometimes orthographic) restriction, the final vowel limitation, is suggested. In Section 2.2, the syntactic basis of the word-formation is put under scrutiny. Here, long established claims of -ee as a de-verbal nominalization are carefully examined, initially by looking at established and new noun-noun formations of -ee and classifying the pattern for formations from a nominal base. Furthermore, the verb as a base is considered with regard to formal limitations of its derivation to an -ee word. Several phonological and morphological restrictions are singled out and investigated in the light of older and more recent (corpus) productions of -ee words. Bauer’s (1983) description of four syntactic patterns of verb-derivations with -ee is outlined and given critical attention in Section 2.2, with special consideration given to those cases where a clear classification of the pattern is not possible because of multiple (synchronic or diachronic) occurrences. To complete the syntactic overview in Chapter 2, a critical review of several other proposals of syntactic theories towards a categorization of -ee suffixation is provided (e.g. passivity, ergativity, unaccusativity). The third part of Chapter 2 is devoted to more recent semantic descriptions of -ee formation (Barker 1998; Portero Muñoz 2003; Booij & Lieber 2004; Lieber 2004). Barker’s analysis of -ee formations, to date still the most detailed and insightful treatment of this specific suffix, is reviewed with regard to the three semantic properties (sentience, lack of volitional control, and episodic linking to the denotation of the stem) he proposes as vital for producing -ee words. By drawing once again on established and new examples of -ee formation, it is illustrated that, while these semantic properties apply to many -ee word examples, they cannot fully account for others. Yet again, it becomes clear that a purely synchronic perspective, which does not consider the external history of the word-formation, is ultimately limiting. In the latter part of Section 2.3, a prototype theory perspective is proposed and applied to -ee word-formation, as such a perception allows a graduation between more typical and more marginal formations. While this may be fruitfully applied to the synchronic level, it does not explain the reasons for the heterogeneity of the pattern and the more atypical formations. Rather, it must be seen as a first step towards the development of a theory of morphological networks (cf. also Schröder 2008) where creative neologisms that deviate from the prototypical formation type can trigger analogical coining and eventually create new rules. In conclusion to this chapter, the syntactic and semantic sources for ambiguity in this formation are revisited and summarized.



Chapter 1.  Introduction

Chapter 3 puts into practice the claim that a diachronic perspective has to be added to address the many changes this word-formation has undergone. It therefore investigates the history of -ee suffixation in English from its beginnings in the fourteenth and fifteenth century, through to the nineteenth century, by taking a close look at the concrete lexical items that were formed during different periods. By examining the new entries for each century, it can be shown that certain “trends” can be followed with regard to semantic and syntactic features of -ee word-formation. Developmental steps in the word-formation process can be scrutinized, for example, from analogy formation (based on deviant cases) to rule formation. The diachronic part of this study, furthermore, proposes a viewpoint on -ee formation as a language contact phenomenon. Having its inception as a borrowing from Legal French in the fourteenth century, the word-formation has its origin in a language contact situation, in which English gradually took over domains like the legal sphere from the erstwhile hegemonic French. It can be shown that, in its development as a productive word-formation on native bases, -ee suffixation underwent several drastic syntactic changes in the fifteenth, sixteenth and seventeenth century; not only did it change its chief pattern of noun-noun derivation but it also developed a predominantly indirect object relation to a verbal base. It is revealed here that, among other influences such as parallel developments with other suffixes, continued borrowings from French affected the course of the word-formation to make direct object formation an increasingly productive pattern. Semantically, it experienced a widening in several directions, moving away from, but not abandoning completely, to which the legal sphere the word-formation was tied in its early usage. For the nineteenth century, the wealth of new words is analyzed with regard to their syntactic and semantic patterns. This can also be placed within a context of the history of the growing diversification of the English language and an increasing influence of the now (politically and) linguistically independent United States of America. Throughout the discussion of the word-formation patterns during the respective centuries, the external history of the formation is considered. In conclusion to this chapter, the changes across the centuries are briefly summarized in Section 3.7. Chapter 4: A particular focus of the discussion so far has been devoted to syntactic versus semantic constraints on the formation of new words. The productivity of -ee words, especially with regard to twentieth-century formations, are now discussed in Chapter 4. Various definitions of productivity in word-formation (e.g. in Aronoff & Anshen 1998; Bauer 1983, 2001; Kastovsky 1986a; Plag 1999) are examined both generally in through various methods of data collection and analysis. Furthermore, distinctions between gradual productivity, qualitative, quantitative, diachronic and synchronic productivity are discussed. Drawing

13

14

Heterogeneity in Word-Formation Patterns

on the insights on diachronic versus synchronic patterns in -ee formation from Chapter 3, particular emphasis is placed on a diachronic analysis of the productivity of this word-formation for the present study. In this chapter, previous distinctions between morphological productivity – an unintentional coinage of a number of formations – and morphological creativity – a conscious coinage of a new word on an unproductive pattern (e.g. Lyons 1977; Schultink 1961) – are centrally positioned in a discussion of the relationship between morphology and the lexicon, i.e., abstract rules versus actual words (cf. Aronoff & Anshen 1998). Neologisms, often regarded as “marginal” for morphological theory (cf., for instance, Lieber 1992), is then given particular attention with regard to their value in providing clues to repeated and misleading statements about the productivity of particular patterns, as well as in helping us understand the development of semantic change and polysemy (cf. also Lehrer 1996). New words, i.e., neologisms (e.g. Elsen 2004; Fischer 1998; Lehrer 1996), ad-hoc creations or nonce words (e.g. Hohenhaus 1996), and their stages of development have received increased scholarly attention in the last decade. In this chapter, I look at such stages, especially with regard to twentieth-century -ee formations from recognized sources, and analyze the factors influencing “ephemeral” usage or stable growth. The data of twentieth-century -ee formations are also examined in relation to similarities and differences in semantic and syntactic patterns, to continuities in new formations versus new trends, etc. A discussion on the problems and merits of measuring productivity is led by using the twentiethcentury -ee words as an example. Finally, the theoretical debate of “actual” versus “possible” words in word-formation theory is continued so as to provide the basis for the empirical test in Chapter 5. Chapter 5: The productivity of new -ee words is put to the test for new wordformations of the late 1990s and early 2000s: the theoretical discussion on syntactic and semantic constraints is followed by an empirical analysis of a sample test of 1,000 potential new -ee words (or -ee words not yet listed in the OED and a number of other sources, cf. Appendix 1) in the largest corpus of written language, the World Wide Web. The question of whether or not the Web can be seen as a corpus is discussed on the basis of recent relevant literature on the subject (cf. Kilgarriff & Grefenstette 2003; Resnik & Smith 2003). The specific conditions of a Web-search, the advantages and the latent problems for this particular study are then considered, in addition to the procedure of analysis. Drawing on a vast range of texts (from official documents and serious news reports, to semi-private weblogs and chats in internet forums) and users, the Web investigation focuses on relatively recent developments and analyzes the environments in which they occur. The question of alternative word-formations with -ee and their influence on the success of a possible -ee word is also pursued.



Chapter 1.  Introduction

The data obtained by testing 1,000 potential -ee words is then classified according to the specific syntactic and semantic criteria for their occurrence. A particular focus is placed on whether or not a co-occurrence of an -er form influences the creation of a new -ee form. Each of the “successful” new -ee words is therefore examined for possible collocations with corresponding -er words in the same environment, for instance, whether apologizee is formed in collocation with apologizer, disrobee with disrober, etc. While most of the 1,000 potential new -ee words have been formed as a verb-derivation, nevertheless, the collocation test offers clues as to whether or not the existence (or concurrent creation) of an -er word influences the success of the -ee word. Semantic criteria such as human and animal (“sentient”) versus instrument denotations are looked at in the body of successful new -ee words, as is the question of agency versus lack of volitional control. A further question is that of semantic field of the vocabulary and the role it plays in the formation of these recent words. Particular emphasis is also placed on ambiguous new words that occur as several semantic and syntactic types. What does such heterogeneity tell us about the role of the speaker in morphological theory? The next part of Chapter 5 deals with those potential -ee words that did not “make it” in the test. Are there any particular ineffective patterns? What roles do formal restrictions play? Are there any morphemes in a complex verb that do not produce a likely -ee word-formation? What can be said about semantic criteria in non-successful -ee formations? In the study of neologisms, hapax legomena are always an interesting phenomenon, which deserves a closer look. Do they show any particular syntactic or semantic patterns? In what text types are they most likely to occur? And, related to this, what kinds of “users” (speakers, writers) are most likely to produce new formations? In which environments? How do they relate to new -ee words which show a higher frequency of occurrence? It is clear that in this type of search in a non-finite corpus, only the temporal restrictions of the search date determine the status of a hapax legomenon. Therefore, the nonce word of, for instance, March 2005 might not be a single occurrence any more. A diachronic follow-up might, therefore, be worth considering. The present investigation of inventive new -ee words, however, necessarily adheres to the particular time frame for which it has been designed, in order not to obscure and distort the data. Given the large (and unexpected) success of the testing of the potential words, the question remains of whether the popularity, and with it the productivity of -ee words, has increased in recent years. As is shown in Chapter 3 and 4, the productivity patterns of -ee words have undergone changes and have also followed certain trends. However, in the past, as in this research, the method of selecting and documenting has to be taken into account. The availability of such large and varied data that can be found on the Web will necessarily produce a greater likelihood of findings. The question

15

16

Heterogeneity in Word-Formation Patterns

therefore remains whether we are dealing with new productivity or whether this output has to be seen in the light of the new research possibilities. Finally, in Chapter 6, the analysis of the Web-corpus centres on the distribution of a sample of -ee words in different geographical environments. It has traditionally been claimed (Mathews 1945; Marchand 1969) that -ee word-formation is particularly popular in the US (as opposed to the UK). The analysis shows whether this is the case for the contemporary situation, or only for some particular lexical items that may be seen as “Americanisms”. After a first discussion of historical and current examples of particular American usages, a second regional focus concentrates on Australia. In the history of the Australian language, some particular -ee words with very specific euphemistic meanings have emerged (cf. Baker 1945; Ramson 1966). In more recent years, hypocoristic word-formations have been popular in Australian English usage, especially nouns with -o and -ie suffixes, as well as -ee. In the latter case, an ambiguous meaning between hypocoristic and non-volitional can often be detected. In the analysis, some of the ambiguous cases will be discussed, especially in the light of various possible influences on Australian English. Subsequently, a structured comparison of the frequency of occurrence of some particular -ee words in different national websites (i.e., websites whose domain is e.g. .uk, .au, .nz, etc.) which points to a particular geographical origin will be conducted and analyzed. It is shown that some regional and national trends can be observed in the use of established and new -ee words. However, the use of web data as data that easily transcends local and national boundaries raises questions of the localization of Englishes, of growing globalization trends and of an internationalization that is also affected by non-native speaker influences. The final discussion of the findings is devoted to this issue.

1.3

Aims and scope of this study: Wider implications

Evidently, this study is concerned with a very specific word-formation pattern in English. It deals not only with its phonological, syntactic and semantic features, its history and the resulting heterogeneity of its formation, but also with its diachronic, recent and new productivity and the new and creative formations in an international context, and in more specific uses in varieties of English. It is hoped that this study will contribute to a fuller understanding of the formation and usage of English nouns with this particular suffix. Furthermore, and perhaps more importantly, this analysis also seeks to raise and discuss questions on theoretical issues which go far beyond the particular interest in the -ee suffix in English word-formation. By highlighting a special



Chapter 1.  Introduction

theoretical and/or methodological concern in each chapter, this study also hopes to fruitfully engage in debates on (i) solid semantic and syntactic categories versus gradual approaches of typical versus marginal formations, (ii) the usefulness of upholding a dichotomy between synchrony versus diachrony in word-formation, (iii) productivity measures and the question of data collection, (iv) the relationship between morphology and lexicon in the study of neologisms and ad-hoc creations. Moreover, this study seeks to encourage (v) innovative and experimental uses of data collection on the Web, as well as (vi) a view on the English language which considers both local and global uses. In sum, the present investigation sees its innovative potential in the study of this particular word-formation with regards to the following points: – On the synchronic level, it advocates a decidedly inclusive model of syntactic and semantic description which does not exclude the more marginal formation patterns but rather sees the formation as ordered in a continuum between more prototypical and more marginal patterns. – It emphatically proposes a distinction between a diachronic and a synchronic level when looking at productivity constraints and dominant syntactic and semantic models. – It considers the polysemy and sometimes ambiguity of the formation to be the result of historical changes and the simultaneous presence of several layers of -ee formation patterns. Therefore, the heterogeneity in -ee suffixation is viewed as a language change phenomenon and the social conditions and motivations of this change are an essential part of the historical description of this word-formation. – It documents in a detailed manner the chronological (first) occurrence of previously described -ee words and lists their sources of documentation. – It adds to these approximately 500 already acknowledged -ee words a structured corpus search on English websites for 1,000 potential new -ee words (not listed in any earlier sources). – By analyzing this new and innovative data, it throws new light on frequency, productivity patterns and collocations of -ee words. It also contextualizes the various text types in which single occurrences of -ee words (hapax legomena) are most likely to occur and looks at the route of establishment for new -ee words. – It describes -ee suffixation as a language contact phenomenon, both in its origin, its development over the centuries and its contemporary influences. Throughout the study, processes such as borrowing and parallel changes in correlative and homophonous word-formation patterns are taken into account.

17

18

Heterogeneity in Word-Formation Patterns

– By looking at the language use of specific -ee items in various varieties of English (UK, USA, Australia, etc.), it proposes a perspective on this word-formation in the light of the globalization and internationalization of Englishes. The use of internationally accessible websites in English (distinguished by their country of origin) is therefore programmatic for this outlook.

chapter 2

Phonological, syntactic and semantic constraints on the formation of -ee words

Morphological descriptions are necessary abstractions of existing complex features of words and deal with the internal structure of a language. One of their chief purposes is to predict potential complex words of a language, even if they do not yet exist. While morphological rules are by definition regular, the actual lexical items produced may be highly irregular and unpredictable. However, such a separation between the morphology of a language – which deals with potential words – and the lexicon of a language – which deals with existing words – may obscure the fact that the two are interdependent. As Aronoff and Anshen (1998: 237) point out, “the morphology, which forms words from words, finds the words that it operates on (its bases) in the lexicon.” In other words, speakers produce new words on the basis of abstract mental rules, which are, nevertheless, based on the corpus of existing words in a language. Morphology, therefore, cannot ignore uncertainties with regard to complex words in the lexicon for the sake of the more “model word-formation” but must try and incorporate more deviant productions. There are quite a few rather heterogeneous word-formation patterns in English, among them the highly productive suffix -er, where the exact syntactic and semantic properties are difficult to describe and predict. This chapter will place the heterogeneity and possible ambiguity in suffixation with -ee at its centre. By giving a critical overview of the most detailed descriptions of word-formation with -ee to date (Baeskow 2002; Barker 1998; Bauer 1983, 1994; Booij & Lieber 2004; Lieber 2004; Portero Muñoz 2003), a particular focus will be given to borderline cases, i.e., on the many cases of -ee formations, which do not fit a general pattern or are ambiguous in their meaning. With the exception of Baeskow (2002), all of the above descriptions adhere to a synchronic viewpoint of -ee formation. While motivations and causes for the uncertainty of -ee will be dealt with in this chapter, I will also argue that the heterogeneous nature of the suffix can only be explained from a diachronic point of view. A diachronic analysis of -ee and the resulting simultaneous patterns of -ee formation will be given in some detail in Chapter 3.

20 Heterogeneity in Word-Formation Patterns

The high variability of -ee formations has been remarked upon in several early descriptions of the suffix, however, not in any exhaustive way. Marchand (1969: 267–268) describes some of the possible types of derivation as well as nonpassive uses of -ee words. Quirk et al. (1972: 998) provide a comment on their original description (verb-derived personal noun, passive suffix) which states that “often the noun, while retaining its passive meaning, is not directly derived from a verb base: nominee, refugee. Some examples do not have passive meaning at all: absentee.” In the last two decades, a small number of scholars, notably Bauer (1983, 1987, 1993, 1994) and Barker (1998) have attempted a synchronic syntactic and semantic classification of -ee suffixation with the aim of describing constraints on the productivity of new word-formations. While Bauer’s analysis (1983, 1994) rests largely on syntactic properties, Barker (1998) argues that this word-formation can only be explained from a semantic point of view and describes the -ee formation type in a thematic role framework. In the early 2000s, notably Baeskow (2002), Portero Muñoz (2003), Lieber (2004) and Booij & Lieber (2004), the -ee suffix was used as an example of a particularly heterogeneous word-formation pattern, which challenges traditional semantic description. In this chapter, these studies will be reviewed and their findings on phonological, syntactic and semantic constraints on -ee formation will be scrutinized closely. This will be done not only on the basis of the examples mentioned in the previous studies, but also in the light of recent -ee words. The source of these contemporary examples of -ee formations are some items collected in scholarly texts and dictionaries, and more importantly, neologisms found in my structured Web corpus search. In this corpus, 1,000 previously unlisted potential -ee words were tested for their occurrence in a meaningful context on the Web (i.e., English language webpages).

A note on the data cited from the corpus material This chapter does not focus on the technicalities of the data collection. Rather, Chapter 5 will give a full description of the methodology employed, its justification, as well as the results with detailed examples. For the purpose of using some of the examples of new -ee formations to enrich the theoretical discussion on phonological, syntactic and semantic constraints, it is sufficient at this point to note that the criteria for selecting the test words was that a) they were – at the time of the data collection (February – June 2005) – not listed in any of the established sources (OED; other selected dictionaries, relevant scholarly texts; cf. Appendix 1) and b) they had to be a recognisable derivation of either a contemporary English verb (the majority) or -er noun (less frequently). The choice of verbs



Chapter 2.  Phonological, syntactic and semantic constraints

was a random “blind” choice from the dictionary; a pre-selection in the form of a consideration of syntactic features (e.g. transitive/intransitive patterns), phonological and orthographic shape, seeming meaningfulness/meaninglessness of the derivation or likelihood was avoided. Out of the 1,000 test words of the corpus, 748 proved to be “actual words,” i.e., they were produced in a meaningful context in an English-language text on the Web. Firstly, this challenges notions of the -ee suffix as “semi-productive” and adds a wealth of data (748 new words in addition to the previously established ≈ 600) on which phonological, syntactic and semantic constraints might be tested. Secondly, this test provides a model of a new method for investigating neologisms. For more details on the method of data collection, please refer to Chapter 5.

2.1

Phonological limitations to -ee verb derivation

The productivity of any word-formation pattern will be restricted to some extent by pragmatic and structural constraints. Thus, phonological limitations can also be found in the formation of -ee words. Plag (2003: 88) describes the suffix as “autostressed,” i.e., “it belongs to the small class of suffixes that attract the main stress of the derivative.” If we take a closer look at arguably the most dominant syntactic -ee formation pattern, i.e., verb derivation, we will see that the phonological shape of the verb plays a role in whether or not an -ee word can be formed from the verb stem. Some formal limitations on the verb base of -ee words have been noted as, for instance, the following in Bauer (1983: 247). 1. If the base of the verb ends in -ate, this is usually deleted, e.g. in alienee (*alienatee), evacuee (*evacuatee), liberee (*liberatee), or nominee (*nominatee), except in cases where “the final consonant of the base would then be /k/ in a word of Romance origin (i.e., spelled with a c)” (Bauer 1983: 247), e.g. in allocatee, dedicatee, dislocatee, educatee or locatee. Anderson (1992: 280) shows, however, that -ate deletion only applies to verbal stems with secondary stress on the final syllable, thus mandáte and narráte form mandatee and narratee rather than *mandee or *narree – where the stem would also become unrecognizable. Elaborating . While this is certainly true, it would be an interesting question whether or not this applies equally to all words formed with this suffix. It has been suggested to me in a discussion at a conference (Anglistentag, Halle 2006) that the affixal stress is not as pronounced in newer formations as it is in very established ones – which could be taken as a sign of the “nativization” of the once borrowed suffix. As my study relies on written material alone, this question cannot be pursued here.

21

22

Heterogeneity in Word-Formation Patterns

on -ate truncation, Plag (2003: 88) points out that a haplology restriction (cf. Raffelsiefen 1999) also applies: -ate is systematically deleted “in those cases where -ee attachment would create identical onsets in the final syllables, as in, for instance, *ampu.ta.tee (cf. truncated amputee), *rehabili.ta.te (cf. rehabilitee).” Baeskow (2002: 583–584) discusses -ate deletion as part of her subcategorization frame and holds that [(base of X-ate)] is an optional subcategorization feature which activates a morphophonemic feature [Ø → {eıt} / /k/ ____]. Therefore, Baeskow contends, -ate is not really deleted – since the input is not really the complete infinitive of the verb – but rather, it is “automatically” inserted in those cases where the base of the -ate verbs is phonologically marked and ends on /k/. As we shall see below, such a mechanistic view may cause more problems than it answers. In many of the established -ee formations one can find competing forms, sometimes even in the same source (for the various sources, cf. Appendix 1), e.g.

communicatee and communicee, donatee and donee, interrogatee and interrogee manipulatee and manipulee operatee and operee

as well as cases where this rule seems to have been ignored, e.g.

congratulatee consecratee constipatee delegatee remuneratee separatee

With the exception of separatee, where a truncation might have been prevented by blocking – separee as part of the compound chambre separee is an already existing loanword from French, the disregard for the proclaimed -ate truncation rule seems more random; neither Anderson’s (1992) above mentioned limitations nor Raffelsiefen’s (1999) haplology restriction applies here. In my test corpus of 1,000 potential new -ee formations (cf. Appendix 2), 88 of the randomly selected test words happened to be based on derivations of . Note that in Lieber’s (2004) morpheme-based semantic theory, such adjustment rules become superfluous since the (meaningful) affix -ee can link itself directly to the (also meaningful) bound root (e.g. nomin-) without having to make a detour via the infinitive (nominate) and its subsequent truncation.



Chapter 2.  Phonological, syntactic and semantic constraints

verb stems ending with -ate. Since the aspect of -ate truncation was not given any special attention (cf. above on the procedure of data collection), the -ate derived forms were usually used in their untruncated version. Only in some cases, where both the truncated and the truncated form appeared in the results, were both cases listed. The result of this test word search shows the following 60 formations where the -ate was maintained (see Table 1). For some of these new forms found on the Web, the aforementioned conditions apply. Like mandate and narrate, sedate and translate would be excluded from the truncation rule, just as rate as a monosyllabic stem could not possibly delete the -ate. Confiscatee, excommunicatee, indicatee, lubricatee, medicatee, Table 1.  New -ee words formed with verb base ending in -ate ITEM

C No.

ITEM

C No.

abrogatee agitatee animatee anticipatee commemoratee compensatee confiscatee contaminatee (and contaminee) denigratee devastatee discriminatee dominatee elevatee emaciatee emancipatee evaluatee excommunicatee expropriatee frustratee humiliatee imitatee impersonatee incarnatee incriminatee indicatee infatuatee infiltratee initiatee instigatee intimidatee

   5   23   31   35 103 106 110 118 119 155 173 193 228 238 241 243 273 277 292 326 374 383 388 392 395 397 399 401 405 412 419

investigatee isolatee lubricatee mediatee medicatee migratee mutilatee navigatee negotiatee perpetratee placatee pollinatee postulatee procrastinatee propagatee ratee regeneratee reincarnatee reinstatee relegatee relocatee remonstratee repatriatee retaliatee sedatee self-deprecatee subordinatee (and subordinee) suffocatee toleratee translatee

420 423 453 466 467 469 487 493 497 548 564 584 592 617 630 656 681 684 685 691 692 699 705 728 782 784 867 868 877 910 925

23

24

Heterogeneity in Word-Formation Patterns

placatee, relocatee, self-deprecatee and suffocatee would also keep the -ate according to the rules laid out above, in order to avoid velar softening. For others, the reason why the formations were successful without complying with this reputed -ate deletion rule is less clear. In a relatively small number of words, only the truncated form was tried (congratulee C 112) or appeared as the only form in my Web search: denominee C 157 and inauguree C 391. The above result should not be interpreted as implying a preference for untruncated forms as, in most cases, only one form was tested. This record suggests, however, that in these more recent formations, the -ate deletion rule does not play a major role, at least not to the extent that it can be maintained as an obligatory rule; there are far more realizations of -ee derivations of -ate verbs (60) than “failures,” i.e., unsuccessful test words (28) (see Table 2 below). However, the number of -ate verbs among the non-realized words is proportionally slightly higher (32.2% of all -ate verbs tested) than verb bases with other endings (25.2% of all formations). As was pointed out above, my list of successful formations with -ate also includes cases which Bauer exempted from the deletion rule, i.e., where the final consonant of the verb would be a /k/, for instance confiscatee C 110, excommunicatee C 277, indicatee C 397, lubricatee C 453, medicatee C 467, placatee C 564, relocatee C 692, self-deprecatee C 784 or suffocatee C 877. While an -ate deletion would indeed cause the problematic softening of the final /k/ sound to /s/ in these nouns (e.g. *excommunicee, *indicee, *relocee etc.), the occurrences where such a deletion would lead to a distortion of the verb stem are too manifold to only highlight this particular issue. Equally problematic for the recognition of the stem Table 2.  Unsuccessful test words formed with verb base ending in -ate ITEM

C No.

ITEM

C No.

*accommodatee *acerbatee *denominatee *deprecatee *disintegratee *dissociatee *domesticatee *exacerbatee *exculpatee *ingratiatee *insinuatee *laudatee *overcompensatee *personatee

   9   12 156 161 200 219 227 274 278 404 409 436 521 552

*predominatee *recriminatee *recuperatee *renunciatee *reprobatee *saturatee *segregatee *underestimatee *understatee *uneducatee *valedatee *vituperatee *vociferatee

604 672 674 704 714 763 783 944 947 950 959 968 969



Chapter 2.  Phonological, syntactic and semantic constraints

and the pronunciation of the -ee formation, I would argue, are instances where the final sound before the -ate is either /i:/ – which would have to merge with the stressed /i:/ pronunciation of the suffix, e.g. in emaciatee C 241 (*emaciee), humiliatee C 374 (*humiliee) – or /g/ – which in most cases would become /dȝ/, e.g. in relegatee C 691 (*relegee), instigatee C 412 (*instigee) etc. 2. A second and possibly more significant formal limitation is instances where the verb stem in question for the -ee derivation ends in a single vowel or diphthong. There are exceptions, both in the list of earlier, previously documented -ee words and in the new corpus data, but they are far less frequent than wordformations maintaining the -ate ending of the verb stem. Thus, some of the most common verbs have not been derived from a passive noun with -ee suffix: there is a gazee but no *seeee or *eyeee; there is an escapee but no *freeee; there is a cheatee but no *lieee, etc. An exception to this is sayee – “a person to whom something is said,” which is listed in the OED as “rare” with a first entry appearing in 1902. Among the already established words (Appendix 1, various sources) which fall into this category, most have verb stems ending on a diphthong, and many connect to the -ee suffix by means of a glide in the oral production or – and this seems to play an important role – through a graphic “pseudo-consonantal” marker “w” or “y”. Examples:

advowee avowee borrowee bowee drawee employee ennuyee payee prayee rescuee sayee vowee

In my own corpus analysis (cf. Chapter 5), there are similar exceptions where verb stems ending on a vowel/diphthong were successfully realized as -ee words, most of them also ending in orthographic “w” or “y”. Yet, there are also a good number of test words in my corpus in this category that clearly did not prove to be actual

. In the written form, orthographic restrictions probably apply in this and in other examples (*eyeee, *freeee, * lieee).

25

26 Heterogeneity in Word-Formation Patterns

Table 3.  New -ee words with vowel end limitation Successful words

Unsuccessful words

ITEM

C No.

ITEM

C No.

bellowee blowee envyee fancyee followee-up identifyee notifyee overthrowee pacifyee playee pursuee qualifyee repayee reviewee rowee satisfyee screwee showee slayee swallowee weighee

  60   69 265 300 314 379 498 529 531 569 642 646 706 740 754 762 776 799 804 888 982

*defyee *disavowee *eschewee *eyee *glorifyee *lieee *mystifyee *overseee *remarryee *testifyee

147 181 267 295 340 443 488 528 695 900

10 out of 31 = 32.2% unsuccessful in this category (as opposed to 25.2% altogether)

words, as can be seen in the following overview in Table 3. The proportion of unsuccessful words to all -ee words in this category (with vowel-end limitation) is 32.2% (10 out of 31 words). David Gold (1977: 160) is uncertain regarding this limitation when he discusses his own creation queree – ‘person who is queried.’ It appears that he had received criticism for violating “the usual pattern for -ee formations by eliminating the final (vowel) sound of the stem” (ibid). Gold’s reaction reveals the speaker’s helplessness concerning some of the conflicting “rules” and limitations of this word-formation: Creswell […] writes that ‘failure to find any -ee word formed on a base ending in a vowel is probably not accidental; apparently there is a basic phonotactic barrier to the creation of such forms.’ I do not know whether the barrier does not exist in my idiolect of English or whether it does and I broke it when writing queree, but at least I am sure of having written it spontaneously and unwaveringly. Queryee, (Gold 1977: 160) on the other hand, sounds strange. 



Chapter 2.  Phonological, syntactic and semantic constraints

This example provides some insight into individual speakers’ reactions to the heterogeneity of the formation, here on a formal/phonological level. As we shall see in more detail below (in Sections 2.2 and 2.3), there are quite a number of (syntactic and semantic) ambiguities that the speaker has to cope with and which cannot be explained by a single prescriptivist model. Gold thus ends his note on queree with the suggestion that “a study of nouns ending in -ee, which have generated some prescriptivist wrath, would be useful” (1977: 160).

2.2

Syntactic limitations of -ee as a verb derivation

Despite the beginnings of -ee formation as a noun-noun derivation (see also Sections 3.1 and 3.2), the verb is seen as the contemporary base of -ee word–formation; at least, this is what most sources agree on: thus. Bauer (1983: 244) states that “characteristic for this word-formation is that it is verb-derived”; Portero Muñoz (2003: 129) even calls this formation a “deverbal nominalization.” Yet, in contemporary formations the -ee suffix does not only attach to the verb. This morphological process is in fact, a counter-example to one of the early generative principles of word-formation, the “unitary base hypothesis,” which states that “the syntacticosemantic specification of the base […] is always unique. A WFR [Word-Formation Rule, S.M.] will never operate on this or that” (Aronoff 1976: 48). While verb-derivation can be seen as the dominant contemporary pattern, noun-noun derivation is still a possibility in present-day formations, as can be illustrated by the following examples, which are, with one exception, all relatively recent (twentieth or twenty-first century) instances: aggressee: Despite all the new freedoms everybody claims they have, I still feel strange when I am the aggressee (John D. McDonald, Free Fall in Crimson), (1981, in Bauer 1987) astrologee C 46: (no v. *‘to astrologe’). There is a very strong alliance between the astrologer and astrologee. Both collaborate for the success of the astrologer (www.washingtonpost.com/ac2/wp-dyn/A26526, July 2004) †benefactee: like NP1 it is not agentive, and it is typically affected by the verb, generally as a benefactee or malefactee (David Gil, “Case marking, phonological size, and linear order,” Syntax and Semantics 8: 315), (1982, in Bauer 1987) biographee: ‘one who is the subject of a biography’ (OED 1841)

27

28

Heterogeneity in Word-Formation Patterns

blind datee: The golden moment when the blind datees come back and tell what a great time they had (BBC2, Clive on Saturday), (1989, in Bauer 1993) †cliticee: Proclitics tend to lose syllabicity in order to attach themselves to the cliticee (John M Anderson and Colin Ewen, Principles of Dependency Phonology), (1987, in Bauer 1993) elevatee C 238: The elevator is full of the smell of a fellow elevatee’s tuna sandwich and … and … really GOOD elevator music. (www.thewholenote.com/wholenote, April 2000) By the time the thing stops at the third floor, he is out the door before he can see the grimaces of the elevatee he left behind. (www.halfbakery.com/idea/Secret_20Floor, not dated) festschriftee: ‘someone to whom a festschrift is dedicated’ (Barker 1998) jihadee C 427: The jihadee kills invading Americans selflessly, with a knowledge or his or her eternal reward. Under such rules of engagement, only the jihadee, with an apparently better grasp of principles supposedly shared by the world’s other great religions, can win. (www.ospolitics.org/worldview/archives/2003/09/24/iraq, September 2003) It’s hard for reasonable people to understand the true madness of extremism – as if some public condemnations of the current oppressor (“jihadee of the day”) will be met with logical understanding and exempt any particular group from being victims of the fanatics. (danielburgin.typepad.com/petrie_dish/2004/08/, August 2004) laryngectomee: ‘a person who has undergone a laryngectomy’ (OED 1956) †malefactee: cf. benefactee (1982, in Bauer 1987) moneylendee: ‘someone to whom money is lended’ (Barker 1998) missionee: ‘one who is susceptible to the arguments of an emissiary or a missionary’ (OED 1951) philantropee: [No example given] (Barker 1998)



Chapter 2.  Phonological, syntactic and semantic constraints

professee C 621: “This model redefines the common, society mandated concept of education as a one way dynamic between lecturer and student, professor and professee.” (www.uvm.edu/~envprog/bittersweetvine/bit-stu.htm, 1996) return addressee: [No example given] (Barker 1998) therapee C 903: The seating was slightly different, two simple chairs nudged close together like therapist and therapee. (www.guardian.co.uk/world/1999/mar/05/clinton.usa2, March 1999) I have mixed feelings about this because I don’t think I gained as much from this therapist/therapee relationship as I would have liked. (poetical.diaryland.com/011211, December 2001) †traitee C 919: “Traits are generally inflicted upon the “traitee” with the “is” keyword, though other modalities are possible” (www.perl.com/pub/a/2004/04/16/a12.html, April 2004)

For some of them, the verb underlying the subject nominalization (e.g. ‘to lend’ → moneylender → moneylendee) or the compound noun (‘to address’ in return address → return addressee) can still be seen as one possible base of the formation. However, for most cases it is much more plausible to explain the -ee as suffixed to the compound noun where, in English, the head is usually the rightmost element; as Barker notes in his discussion of the semantics of -ee, “a blind datee is someone who participates in a blind date, and not someone on a date who is blind” (Barker 1998: 716). Moreover, there are no verbs (as of yet) *‘to astrolog(ue),’ *‘to aggress,’ *‘to benefact,’ *‘to mission’ or *‘to laryngectome.’ While ‘to profess’ (= ‘to declare’, ‘to announce’ or ‘to take vows’) certainly exists as a verb, the formation professee here seems to be plainly based on professor as a complementary opposite to the new

. “[…] well-behaved traits are really just roles applied to declared items like containers or classes. It’s the declaration of the item itself that makes traits seem more permanent than ordinary properties. The only reason we call them “traits” rather than “properties” is to continually remind people that they are, in fact, applied at compile time. (Well, and so that we can make bad puns on “traitor.”) Even ill-behaved traits should add an appropriately named role to the container, however, in case someone wants to look at the metadata properties of the container” (ibid.).

29

30

Heterogeneity in Word-Formation Patterns

word. A similar case is elevatee: ‘to elevate,’ meaning ‘to raise, lift up’ exists as a verb both in a concrete and figurative meaning but the context in the examples here – ‘someone with whom an elevator is shared, someone in an elevator’ – makes it rather unlikely that it is verb-derived. This is also an interesting formation in that the correlative noun to elevatee is an instrument rather than a person (cf. discussion in Section 3.7). Jihadee, on the other hand, has a somewhat special status due to the stem being evidently borrowed and the word class of the stem not, perhaps, being transparent or even important to English speakers. Baeskow (2002: 587) formulates two subcategorization frames, a verbal and a nominal one, for the formation of -ee words. Her study is notable in that she treats -ee derivation from nouns not as merely a deviation. However, her technical description does not shed any new light on how or why -ee noun derivations are formed or why the suffix can be attached to multiple bases. Bauer (1987: 315), despite counter-examples, insists on verb-derivation even for patterns of this type when he writes that “the example biographee […] seems to presuppose the nonexistent verb *[sic] biograph. […] benefactee and malefactee fall into the same category.” The fact that Bauer does not take into account the possibility of a noun-noun derivation becomes evident when he states that “lessee is phonologically lexicalized, since the expected form of the base would be /lis/ rather than /les/” (1983: 248). But lessee is a fifteenth-century word formed as a correlative noun to lessor (OED first entry: “[a1481 LITTLETON Inst. §57 Il y ad le Feoffor, & le Feoffee, le Donor & le Donee, le Lessor & le Lessee.]”) and is certainly not derived from ‘to lease.’ It

. Arguably, a similar case could be made for detectee (C 170) which appears in my corpus (2 entries) also only as a complementary opposite in relation with detective. . There are also examples in my corpus where elevatee is formed as a deverbal noun, meaning ‘someone initiated, uplifted,’ usually in a religious sense. This is an example of a formation with an ambiguous meaning. . It has to be said, however, that ‘to jihad’ by now also seems to have been successfully established as a verb, cf. the following examples (accessed September 2005) (a) “You’d think the assatollahs of Same-Sex activism just might consider the years of intercine social warfare in this country post-Roe v Wade before jihading into the judiciary.” Darleen’s Page, March 9, 2005, (www.darleenclick.com/weblog/archives/2005/03/jihading); (b) “I’m allergic to some or all cats so my body is Jihading them? even if they like me … hmmm.” October 6, 2004, (forums.catholic.com/showthread). An extended google-search (September 2005 – separate from the main corpus search) for ‘jihading’ on English language websites provides more than 300 links, most of which are making sense (in their own logic). . This is not quite accurate: In the OED, ‘to biograph’ is listed as a verb (1883) even though it appears after the first listed occurrence of biographee (1841), thus ‘to biograph’ seems to be a denominal verb formation.



Chapter 2.  Phonological, syntactic and semantic constraints

seems that here the strictly synchronic viewpoint obscures the heterogeneity of the formation, which is partly a result of the historical development. However, some of the examples above suggest that correlative noun-noun derivation (astrologer-astrologee, therapist-therapee, aggressor-aggressee, benefactor-benefactee) is also possible in more recent formations without the existence of a corresponding verb. In most cases where an agent noun of the -er/-or/-ist type exists, it is itself a subject nominalization of a verb or, like in the case of ‘to biograph,’ a backformation to supplement the verb. Therefore, the role of an existing correlative noun for the formation of an -ee word is difficult to determine but many of the co-occurrences of -er/-ee forms in the examples of the corpus analysis (cf. Section 5.3) suggest that it has at least a reinforcing character. While most of my test words were based on verb derivation and can therefore not show any evidence of a possible productivity of a noun-derivation pattern, my collocation test in Section 5.3 shows that many of the successful new -ee formations appear in juxtaposition to a corresponding -er word, as in the following examples:10 applauder/applaudee C 38: The twelve steps that follow are based on my own experiences as a charter member of the Shy Club for Men. They can help you make the transition from audience to presenter – from applauder to applaudee, a transition that can make a major and lasting contribution to your income and self-esteem. (www.newentrepreneur.com/Resources/Articles/Look_more_confident/ look, not dated) beggar/beggee C 59: So she observed both the beggar and the beggee; in her paper, she introduces characters such as Jerk, who sells Garbage Pail Kids cartoons while passing out […] (cornell-magazine.cornell.edu/Archive/July%201999/JulyLegal.html, July 1999) haunter/hauntee C 356: To make a good ghost movie you must have a plot, haunter, and a hauntee. You must also have a scary place like an old house or a haunted hotel or something […] (library.thinkquest.org/J0112622/movies.html, not dated)

. Cf. also a misunderstanding of the verb-base of trustee (cf. discussion of syntactic ambiguity in Section 2.2.2). 10. Of all the 748 successful -ee words in the test, 552 items (73.8%) occurred in at least one instance in “lexical solidarity” with an -er word, whereas 196 items (26.2%) produced no such collocations, cf. detailed results and more examples in Section 5.3.

31

32

Heterogeneity in Word-Formation Patterns

sneerer/sneeree C 817: I am in awe of these techno-nerds, but it is much more fun being the sneerer with Dilbert than the sneeree with Netscape. (www.enneagramcentral.com/edt_0207.htm, not dated)

Booij and Lieber (2004), in their contrastive study of -ee and -er affixation in English and their analogies in Dutch, suggest that noun derivations like biographee are formed on the base of biography rather than on a verb. This is achieved through the semantic framework of co-indexation (cf. also Section 2.3.1). The various types of non-deverbal -ee noun formations might be summarized as follows: 1. -ee noun formation as correlative noun to an existing -er/-or/-ist noun without a verbal base (e.g. aggressee, benefactee, malefactee, philantropee, therapee, traitee); 2. -ee noun formation as correlative noun to an existing -er/-or/-ist noun which do have a possible but unlikely verbal base (e.g. professee, elevatee, chargee11); 3. -ee formation with a compound noun as the base (e.g. moneylendee, return addressee); 4. -ee formation with a noun as the base which is not a correlative agent noun (e.g. jihadee, laryngectomee, missionee). The opposite case, i.e., verb derivation of -ee without an existing correlative agent, is easier to illustrate, simply because there are more verbs without a corresponding -er subject nominalization than the other way round, e.g. *deporter, *indoctriner, *paroler, or *rehabilit(at)er. Adjective derivation seems to be a rare exception. As we will see in the historical part in Chapter 3, the most notorious case is absentee (1537), which could, however, also be explained by simple borrowing or verb derivation with a French loan (s’absenter). In my (mainly verb-derivation based) corpus analysis of -ee words, I have come across some isolated examples of possible past participle derivation (e.g. stuckee,12 drunkee13), which suggest that 11. In its 1884 meaning: OED ‘The holder of a charge upon property, or of a security over a contract.’ 12. Ex. in my corpus word decodee (C 143) – “This leaves the DF-fragmenter as the stuckee […]. 13. Ex: “Well let me tell you a story about my friend. I don’t think the teachers would quite appreciate it if I told you that’s he’s an underage drinker but oh well. He is. Well anyway this friend I have will get drunk, so drunk that I swear this guy starts to loose his mind. One time we where playing some kind of Nintendo game and he’s like: “I crossed my eyes and titled [sic] my chin down doing an impression of this drunkee (www.geocities.com/ sargon_of_azon/fred.htm).



Chapter 2.  Phonological, syntactic and semantic constraints

this type could, in principle, be possible. It is surely not a major pattern, however, and I have not made any attempt at structured research in this area nor have I included the examples in my corpus list. Finally, there are some cases where -ee noun derivation preceded the -er form (cf. also Barker 1998: 709), for instance in guarantee (1679), which is documented before the synonymous guarantor (1853), or committee (1495) – here still referring to an individual person – to which committor was later formed as a correlative noun (cf. OED). As the discussion has shown, the type of formation -ee undergoes with regard to the word class of its stem is far from clear, not only from a diachronic perspective (cf. Chapter 3) but also in a synchronic view. While de-verbal nominalization is the most prominent pattern, derivation from other word classes (noun-noun, potentially adjective-noun) is also possible. The likely influence of a co-occurring -er noun on the production of a new -ee noun (even if it is chiefly verb-derived) also has to be taken into consideration.

2.2.1

Verb derivation with -ee: Syntactic properties

The syntactic properties of verb derivation with -ee have been dealt with in a number of studies. In Bauer (1983: 245), they are portrayed rather straightforwardly as consisting of four different types of verb derivation. Since this classification forms the basis of a number of syntactic descriptions and elaborations (Baeskow 2002: 553ff., 585–587), and has also inspired a lot of criticism (e.g. Barker 1998), it will be reviewed here at some length.

(i) Direct object (DO)

+ Verb + Transitive + Noun + Human DO

Bauer (1983: 245) describes the transitive verb-derived form ‘one who is Ved’ as “the most common meaning (in both the established and nonce forms).” Barker (1998: 705) also sees this type as the “prototypical or basic in some sense” and notes that 53% of his total -ee word body belongs to this category. As we will see in the historical development of -ee forms, such a claim is true at least for new Theoretically, drunkee could also be derived from ‘drunkard.’ However, the presence of the rival ‘drinker’ in this passage and the co-occurrence of ‘drunk’ makes it more likely that the latter serves as a base for the derivation.

33

34

Heterogeneity in Word-Formation Patterns

formations that have been created since the nineteenth century (62.5% DO forms in the nineteenth century, cf. Section 3.6). It has to be noted, however, that quite a few -ee words which today seem to belong to this type are historically indirect object types (ii) and some have even changed from indirect object to direct object meaning, e.g. recoveree, (cf. Section 3.3). Many established nineteenth- and twentieth-century examples from all contexts (gen., legal, milit., hum.) fit this context (all OED):

advisee – ‘one who is advised’ evictee – ‘one who is evicted’ trainee – ‘one who is trained’ hecklee – ‘one who is heckled’

(ii) Indirect object (IO) “The second most common pattern” (Bauer 1983: 245) are formations with ditransitive verbs (i.e., with an indirect object), mostly of the type ‘one to whom something is Ved.’ In Barker’s (1998: 705) count, this type comprises 16% of all of his overall body of -ee words. + Verb + Di-transitive + Noun + Human IO

Examples of this type are also abundant. However, it is worth noting that this is the preferred type in legal use, especially but not exclusively in historically earlier formations (cf. also Chapter 3). Examples (all listed in OED):

vendee – ‘one to whom a thing is sold, the purchaser’ promisee – ‘one to whom a promise is made’ abandonee – ‘one to whom anything is formally or legally abandoned’ mortgagee – ‘one to whom property is mortgaged’

For some verb derivations, both direct and indirect object interpretations are possible, e.g. addressee – ‘one who is addressed,’ or – ‘person to whom a letter, packet, etc. is addressed’ (OED).

Equally, the meaning of presentee (listed in Barker 1998: 705 as indirect object type) originally referred to the direct object (‘a person presented,’ OED), and still does in institutional (legal, clerical) contexts, as can be seen in this quote from



Chapter 2.  Phonological, syntactic and semantic constraints

the website of the British clergy association: “As already stated, unless a parish has passed resolutions A & B (see above) neither representative may reject solely on the grounds that the Patron’s presentee is a woman.” (www.clergyassoc.co.uk/ content/interregnum, italics added). But presentee might also be used to mean ‘person to whom something is given’ (The American Heritage Dictionary). This meaning is also given in an example like “It would be an insult to the presenter, so the presentee keeps the pen safe in a desk drawer.” (www.kamakurapens.com/ makie/JapaneseAndFountainPens.html, italics added). For some words, the fact that the meaning has changed over the centuries makes a synchronic percentage count of the various types highly questionable. In this, as in the subsequent chapters, the meaning of -ee words will therefore be examined in the specific context of usage and the period of time in which they have been employed.

(iii) Object with preposition (O Prep) In Bauer’s (1983) list, the third group of verb derivations is based on verbs that permit a prepositional phrase including a human noun, ‘one who is Ved preposition.’ + Verb + Noun + Human O Prep

Examples (OED):

staree – ‘one who is stared at’ flirtee – ‘one who is flirted with’ noddee – ‘one who is nodded to’ gazee – ‘one who is gazed at’

Barker (1998: 705) states that 7% of his -ee words belong to this type, the same figure as for the next and last category, that which Bauer (1983: 247) calls the “agent type.”

(iv) Subject (Su) Here, -ee is synonymous with -er and the human noun is the subject rather than the object of the base verb: + Verb + Noun + Human Su

35

36

Heterogeneity in Word-Formation Patterns

This type, which seems to contradict the notion of -ee formations as “passive” or “patient nouns” (cf. Bengtsson 1927; Quirk et al. 1972; Katamba 1994), consists of two subcategories. The first one refers to the subject of an intransitive verb, e.g. waitee – ‘someone who is compelled to wait’ (i.e., in a waiting room) (Bauer 1987) standee – ‘one who is compelled to stand’ (for instance, in a bus) (OED) resignee – ‘resigner’ (OED) dinee – (Barker 1998) surfee – ‘surfer, someone devoted to surfing’14 (C 881) failee – ‘someone who fails’ (C 297)

The other possibility is the subject of a transitive verb, e.g.

attendee – ‘one who (merely) attends a meeting, conference, etc.’ (OED) parkee – (Lehnert 1971) signee – ‘one who has signed a contract or register’ (OED) expressee – ‘one who expresses an opinion’ (C 291)

Bauer (1983: 248) concludes that “in all, then, there are four separate patterns which have been productive in the formation of established -ee derivatives.” He explains failures to fit in any of these patterns with lexicalization, or, as Lyons (1977: 547) called it, “fossilization.” Lexicalization occurs when, due to some changes in the language system, the lexeme takes on “a form which it could not have if it had arisen by the application of productive rules” (Bauer 1983: 48). While lexicalization often explains unanalyzable synchronic forms from their history, the examples that Bauer cites here (amputee as semantically lexicalized, lessee as phonologically lexicalized, and trustee as morphologically lexicalized “since the verb from which the meaning is derived is entrust, not trust”), do not hold. The formation of lessee as a correlative opposite to lessor rather than a verbal nominalization can still be a productive rule, trustee as derived from an older meaning

14. There are several possible meanings to surfee: 1. A surfing addict: “One was a barefoot surfee called Dave Smurthwaite who only packed seeds when the surf was flat. The other was Matthew Alexander, who had a preference for organic vegetables.” History of the Diggers Club (www.diggers.com.au/HistoryOfDiggers); 2. A website, or someone whose website has been frequented, i.e., “surfed”: Example: “As you surf you may like sites that only require 10 seconds before you can surf again, but this does not benefit pages that load slowly. So observe not only from the surfer perspective but also from the “surfee” point of view” (www.ebizyourhome.com/articles/ traffic_exchanges).



Chapter 2.  Phonological, syntactic and semantic constraints

of the verb ‘to trust’ is equally analyzable, while the formation of amputee15 might be explained by the -ate deletion rule Bauer has himself suggested.

2.2.2 Syntactic ambiguity in -ee verb derivation As we have seen in the discussion above, the meaning of the examples of -ee words may vary – a fact which makes their belonging to one specific pattern questionable. The meaning of quite a few -ee words is therefore potentially ambiguous.16 The types of ambiguity include: (a) Cases where it is not entirely clear which category they belong to. For instance, signallee – ‘one to whom a signal is made’ could be read as either an indirect object or an object + preposition type (cf. also Portero Muñoz 2003: 132). Similarly, the meaning of trustee at the time of its first occurrence (1647) is given as both ‘one who is trusted’ and ‘one to whom something is entrusted,’ i.e., this could be interpreted as either a direct object or an indirect object pattern.17 (b) Cases where the meaning of the word has changed along with the type of formation. As we will see in the historical development in Chapter 3, this is particularly relevant for cases where the meaning has changed from indirect object to direct object or vice versa. Thus, one of the examples listed under indirect object here according to the definition it is given in the OED, abandonee – ‘one to whom anything is formally or legally abandoned,’ might also be found in a direct object use, according to the following examples from my web search (all accessed May 2005), where the abandonee more often18 denotes an abandoned person, especially abandoned in a partner relationship or an abandoned child:

15. The OED cites, in fact, also the verb ‘to ampute’ (1623) but this has been obsolete for some centuries. Since the first occurrence of amputee is given in 1910, we might safely assume that the then and now current ‘to amputate’ is the basis of the word-formation. 16. I gratefully acknowledge the point made by an anonymous reviewer that the meaning of the words is usually not ambiguous in contextual interpretation. However, their multiple potential meanings in different contexts, I would argue, seem to justify the attribution of an ambiguous/ polysemous character to these formations. 17. Bauer (1983: 248) evidently missed the possibility of the first meaning when he writes that trustee must be “morphologically lexicalized since the verb from which the meaning is derived is entrust, not trust.” In the meaning of trustee (1647) as ‘one who is trusted,’ the verb base is most likely the transitive verb ‘to trust’ which has been attested at least since 1374. 18. Even though the original “legal meaning” is also still in use, consider, for instance, from the website National Bankruptcy Conference: […] The abandonee of property under section

37

38

Heterogeneity in Word-Formation Patterns

[…] This would leave a repentant “abandoner” in a really tough position if the “abandonee” moves on and marries someone else. Hope that helps! Blessings! (www.singleservings.org/board/board_topic, not dated) […] The fact that it’s more painful to be the abandonee than the abandoner is rarely acknowledged by the latter, because both sides want to be considered “the injured party.” (www.emotionalfeelings.tripod.com/emotional_feelings, not dated) […] Hoping to save the child from an orphanage, the boys try to find a new home for the abandonee on their own. (www.entertainment.msn.com/movies/movie, not dated)

Similar meaning and position changes would be transferee, from (1736) “one to whom a transfer is made,” to (1892) “one who is transferred or removed; e.g. from one position or grade to another” (OED) or releasee, cited in the OED as (1744) “one to whom an estate is released,” but in contemporary usage more likely realized with a direct object meaning as “someone released (from prison or another institution)”: […] The designated program shall conduct an assessment in accordance with the provisions specified in 77 Ill. Adm. Code 2060.417 and as further specified by contract to determine if the releasee is likely to be rehabilitated through substance abuse treatment. 1. The designated program shall obtain the releasee’s informed consent prior to the provision of services. […] (www.ilga.gov/commission/jcar/admincode, not dated)

An indirect object formation can also become a subject formation, as the following case suggests: Surrenderee in its (1662) meaning is given as “The person to whom an estate, etc. is surrendered: correlative to surrenderor” (OED) but seems to have changed to “someone who has surrendered,” i.e., a surrenderer (for instance, in a war activity), as can be seen in these examples (all from the web, accessed May 2005): A surrenderee who claims to be a former top official of the underground New People’s Army and a former consultant of the Nationalist People’s Coalition has linked the Canlaon People’s Rights Movement and five other persons to the ambush of the late Canlaon City Vice Mayor Jose Cardenas.

554 should have a prepetition claim for damages, that is, for the net cost of remediation […] (www.law.uchicago.edu/NBC/Overview, not dated).



Chapter 2.  Phonological, syntactic and semantic constraints

“Surrenderee links eight to Cardenas ambush,” by Edmund Sestoso, Daily Star, 25.09.2002. (www.visayandailystar.com/2002/September/25/negor) Last year, the Union home ministry approved the package for Jammu and Kashmir that provides every surrendered militant a monthly stipend of Rs 2,000 for three years and a fixed deposit of Rs 1.5 lakh, which can be withdrawn after three years if the surrenderee exhibits “good behaviour.” Also on offer are vocational training and assistance in obtaining bank loans for starting a business. The scheme was cleared at the request of the Jammu and Kashmir government, which initially sought a grant of Rs 3 lakh for each surrenderee but settled for Rs 1.5 lakh. (www.telegraphindia.com/1050307/ asp/frontpage/story, March 2005)

There are also established examples of this type: is a meetee the object or the subject of the meeting, i.e., can the meetee also be the meeter? Bauer notes on this, “in the case of meetee, this might be an instance of pattern 3 in American English, where one meets with a person. The most likely interpretation, it seems to me, is one with a conjoined plural subject, as in The management and union representatives met last night, where everyone concerned is a meetee” (1983: 247). (c) Cases where more than one pattern is used simultaneously, for instance, subject and direct object uses. This differs from (b) in that it is not a diachronic development but a synchronic multi-formation. Polysemy and ambiguity in (b) are the result of the evolutionary process of -ee derivative formation. Multiple meanings and interpretations that can be found in the use of individual synchronic formations (c) are based on the availability of these historically formed multiple patterns which allow analogical coinages from several models. Among the new words (i.e., from my corpus), such ambiguity is not uncommon, as the following examples (expressee, failee) will clarify: expressee C 291: subject (pattern iv): “LEGAL DISCLAIMER: The stories, views and opinions expressed on this site are solely those of the expressee.” (www.groups.msn. com, 2005) object (pattern ii): The intention behind any expression, must be a desire that the expressor (you) will, after expressing, get along in the world a little better with the expressee (your boss, spouse, child or colleague). (www.barrowford-stress-clinic.co.uk/assertiveness.htm, August 2004)

39

40 Heterogeneity in Word-Formation Patterns

failee C 297: subject (pattern iv): Timothy Prescott, a multi-failee himself, strode over. “That’s right generous of you, Richards.” (www.geocities.com/diadem8/fanfic/ab/before6, not dated) This will ensure the other trainees don’t get bored and the “failee” doesn’t get disheartened; you can return to this exercise later, […] (222.uel.ac.uk/subaqua/site/articles/pages/how-to-pool.htm, February 1996) object (pattern i): I think you can be the victim of someone else’s failure. So my parents are moderately middle-class people – my dad’s a lawyer – who haven’t quite had the life they expected. My Dad’s judgement led to him mistiming and mis-judging a couple of crucial deals that would have meant he could retire at thirty. So I guess that makes my mother a failee: she’s living a certain degree of failure, as a result of the actions of someone else. It’s a good reason not to get married, too, methinks. (www.troubled-diva.com/2003_10_19_troubled-diva_archive.html, October 2003)

The last example, idiosyncratic as it may be, illustrates an important semantic property of -ee words, one that Barker (1998) terms “lack of volitional control” and which will be discussed in more detail in Section 2.3. Bauer (1983: 250) sees an increase in the productivity of pattern iv (Subject pattern) and notes that “why this should be when there is such a plethora of suffixes for indicating persons without a ‘patient’ meaning (-er, -ist, -ite, -man, etc.) is far from clear.” For some of the examples cited above, -er suffixation seems to have been prevented by a simple case of blocking: waiter (profession), diner (type of restaurant) or, to some extent, attender (or attendant, as profession) have, after all, a different meaning from ‘someone who waits,’ ‘someone who dines’ or ‘someone who attends a meeting’ and *meeter, due to its homophony with meter, would, at least in oral communication, be ambiguous. More puzzling is the co-existence of -er/-ee synonyms as, for instance, in stander-standee, parker-parkee, signer-signee or failer-failee. Why use a contrasting word-formation pattern if a more common one (for ‘agent’ meaning) is already established? Bauer (1983: 290) sees semantic reasons – similar to Barker’s (1998) characteristic “lack of volitional control” – as the only satisfactory explanation for his examples (dilutee, embarkee, retiree and standee): The only reason I can think of for why these should use the -ee suffix rather than -er is that, although the people concerned are potential subjects for the verbs in the bases, they are felt to be affected by circumstances rather than acting of their



Chapter 2.  Phonological, syntactic and semantic constraints

own free will. This “explanation” is far less convincing for embarkee than for the other three items.

The same also applies to parkee and signee, one might add. Barker (1998: 705) thus assigns no value to describing -ee words from a syntactic standpoint and states that “the main difficulty for syntactic theories is that from a syntactic point of view, the set of possible referents for -ee nouns just does not seem to be a natural class.” As we shall see in the discussion below, Barker also dismisses previous analyses of -ee as “passive,” “unaccusative” or “absolutive” (Barker 1998: 705–708). The latter point was tentatively suggested in Bauer (1983: 250): One suggestion (made to me by Bernard Comrie) is that it shows a trace of ergative-absolutive patterning in English: that -ee is coming to be used to mark the object of a transitive verb and the subject of an intransitive verb, like the absolute case in ergative languages. Whether or not this is the case (and there are a couple of examples which cannot be explained by this), it may be true that such formations in -ee show a coming trend in word-formation patterns.

In the following sections of this chapter, this and other syntactic hypotheses will be critically discussed before I address (in Section 2.3) semantic hypotheses of -ee word-formation and their possible meanings.

2.2.3

Syntactic hypotheses of -ee word-formation: A critical review

We have seen that scholars like Bauer (1983) have resorted to both syntactic and semantic explanations for the diversity of -ee words, even though his analysis rests on mainly syntactic criteria. Other early syntactic hypotheses have tried to find a more uniform explanation for the production of -ee words and characterized the formation as passive (Bengtsson 1927), as absolutive (cf. Bauer 1983 above) or as unaccusative (Horn 1980).

(i) -ee as passive The passivizability of -ee words has been supposed from early publications in the field (cf., for instance, Bengtsson 1927) and is still seen as a core feature of -ee words in many treatments of the suffix. Barker (1998: 705) states the passivizability hypothesis as follows: “-ee nouns can refer to entities that can serve as the referent of the subject of a passive (or pseudo-passive) version of the verbal stem.” We have already seen in the course of the introduction (Chapter 1) and the debate of -ee as a verb derivation (Section 2.2) why this hypothesis of -ee as passive is problematic. While the rule stated above can be used to accommodate all

41

42

Heterogeneity in Word-Formation Patterns

those formations involving a direct object, many of the indirect objects do not fall into this class: The advisee was advised (by the adviser) The interviewee was interviewed (by the interviewer) *The borrowee was borrowed

Notably the legal lexical items in the historical development of -ee (recognizee, assignee, also: abandonee, nineteenth century), where the formation with the indirect object of the verb was the rule rather than the exception, would be a claim against the notion of -ee as truly “passive.” Such an argument, however, places great emphasis on the grammatical features of the individual verb, i.e., whether or not the verb can actually take a direct object, whether or not indirect and direct objects are both applicable, etc. To avoid such a narrow view, Barker modifies the passivizability rule to deal in grammatical relations rather than in individual lexical items: […] since some indirect objects can occur as the subject of a passive sentence, therefore every indirect object participant can be expected to potentially end up as the referent of a corresponding -ee noun. Thus the failure of individual verbs to allow indirect objects to passivize would not be relevant.  (Barker 1998: 706)

But even such a wider view of what may be seen as passive cannot possibly include the “agent type” -ee formations, where the nominalization is the subject rather than the object of the verb (e.g. waitee) or, indeed where the verb may not even be the base of the formation (e.g. astrologee, professee), cf. the discussion in Sections 2.2.1 and 2.2.2. In order to explain the correlation between -ee formation and passivizability, Barker (1998: 706) proposes a semantic rationalization rather than continuing to pursue a syntactic classification. He sees the semantic component of “lack of volition” as essential in -ee noun formation which is able to explain both the passive and the non-passive constructions: “Lack of volition is typical of the referents of direct objects and other passivizable grammatical relations (explaining the correlation), but is not limited to those relations (explaining the exceptions to the passivizability hypothesis).” As has been established in Bauer’s (1983: 250) statement quoted above, such a semantic argument could also account for agent formations such as waitee or standee – assuming that the actions the human referent undergoes here are unwilling and involuntary ones. Whether or not this semantic feature – lack of volition – can provide an overall explanation for the formation of -ee nouns, or whether this might lead to other inexplicable gaps, will be discussed in the Section 2.3 below. In



Chapter 2.  Phonological, syntactic and semantic constraints

that section, Barker’s hypothesis of -ee noun formation, which includes a number of such central semantic features, will be described in more detail.

(ii) -ee as absolutive Ergative-absolutive patterning in -ee formation in English is discussed in Barker (1998). Ergative-absolutive languages have a basic morphological case for the marking of the subject of an intransitive verb and the object of a transitive verb, independent of their semantic roles (agent, patient) in the sentence (cf. Bußmann 1990). What seems at first like a neat way to place apparently disparate constructions like advisee, evictee, trainee, hecklee (DO of transitive verb) and waitee, standee, resignee, dinee, surfee or failee (Su of intransitive verb) into one syntactic category, becomes intensely problematic when we look at the one case which is clearly not possible to classify as an absolutive in ergative languages: the subject of a transitive verb. This is, as we know from our discussion of the syntactic possibilities of the -ee formation, nevertheless realized in English -ee nouns, e.g. in attendee, parkee, signee, representee. Besides, the ergative-absolutive hypothesis cannot account either for those -ee formations which are noun-derived (e.g. professee) or for those which there is no corresponding verbal argument (e.g. amputee). (iii) -ee as unaccusative Within the framework of Relational Grammar, and based on the idea that syntactic properties of verbs are determined by their meaning, -ee has also been described as an unaccusative. The Unaccusative Hypothesis (first linked to Perlmutter 1978) suggests that the class of intransitive verbs is not homogeneous but consists of two subclasses, the “unaccusative” verbs and the “unergative” verbs, which are each associated with a different underlying syntactic configuration. As Levin and Rappaport Hovav (1995: 2) point out: Because of the convergence of semantic and syntactic properties that characterize it, unaccusativity provides fertile ground for exploring the relationship between lexical semantics and syntax. The importance of the Unaccusative Hypothesis is that, if correct, it allows us to use unaccusativity as a means of identifying aspects of verb meaning that are relevant to the syntax and of appropriately formulating at least some of the linking rules.

The idea of -ee as an unaccusative was developed by Horn (1980) who suggests that -ee will attach only to verbs that have a direct object in deep structure – which then also allows for -ee nouns referring to indirect objects; if the stem verb takes an indirect object, it will also allow a direct object. For intransitive verbs, the Unaccusative Hypothesis would then predict that -ee attaches to “unaccusative” verbs

43

44 Heterogeneity in Word-Formation Patterns

(surface subject is an “initial 2” – direct object) but not to “unergatives” (surface subject is an “initial 1” – subject and there is no “initial 2” – direct object). There are a number of existing cases of -ee words that cannot be accounted for by this hypothesis: firstly, if narrowly interpreted, it would not allow for -ee nouns which refer to the prepositional object of the verb. Secondly, the focus of this hypothesis is entirely on the verb (and, therefore, verb-derivation) and does not include noun-noun derivation in -ee words, or -ee words which do not have an analyzable English verb as its stem (refugee, absentee, etc.). In sum, it should be clear that all of the above syntactic classifications that try to explain -ee formations by means of a single syntactic argument have severe shortcomings: a. None of the above can account for “the significant number of -ee nouns which refer to participants for which there is no syntactic argument position at all” (Barker 1998: 708) (e.g. amputee); b. Subject formations, especially from direct objects, cannot be accounted for; c. Noun derivations from correlative -er nouns or other nominal bases cannot be accounted for. Baeskow (2002) tries to overcome these deficiencies in her formulations of two subcategorization frames for the formation of -ee nouns, a nominal and a verbal one. From a purely descriptive point of view, this model includes the cases cited above. However, it does not really explain the question of why this diversity of input is possible.

2.3

Semantic explanations: -ee formation in a thematic role framework

We can state that, while detailed synchronic syntactic descriptions can illustrate the heterogeneity of this word-formation pattern, they do not really clarify many details about the production process of -ee formation. Most recent studies (Barker 1998; Booij & Lieber 2004; Lieber 2004; Portero Muñoz 2003) have therefore focused on the semantic properties of -ee suffixation. Barker (1998) has given the most thorough description to date of the semantic characteristics of, and constraints on, -ee formation. His model has been taken up as a basis for extension and critical discussion by most other researchers. It will also be dealt with here at some length. Barker (1998) does not deny that there are syntactic constraints to the formation of -ee words but sees a semantic approach as ultimately more fruitful in explaining the various types and meanings of -ee nouns:



Chapter 2.  Phonological, syntactic and semantic constraints

[…] the fact that the semantic approach automatically accommodates without adjustment the more exotic -ee noun types (governed preposition uses, uses for which there is no corresponding verbal argument, and uses with non-verbal (Barker 1998: 709) stems) certainly counts in its favor.

He basically attributes three semantic properties to -ee words: (i) the referent of an -ee word must be sentient, (ii) the use of an -ee noun entails lack of volitional control on the part of its referent, and (iii) the denotation of an -ee noun must be episodically linked to the denotation of its stem.

(i) Sentience As we have already seen at the outset of our explanations of the formation criteria of -ee words (Chapter 1), the idea that an -ee referent must be a “person” (Katamba 1994: 65; Aitchison 1999) does not always hold. Some of the new formations cited from my corpus, such as milkee (C 470), brushee (C 75) or cleanee (C 92)19 imply, however, that the referents are able to somehow experience the action they are undergoing. The non-human animate referents here (cow, dog, fish) are animals or pets to which the speaker attributes this capacity. In contrast to this, animals to which we do not attribute a sense of feeling or experience are not the referent in an -ee formation. In the following examples of new -ee words from my corpus, the eatee (C 237) is neither the slaughtered animal nor the meat on the table but indeed a human being who – this will be important in the discussion (iii) below – is a also a participant in the action, i.e., most of the entries20 refer to the so-called “cannibal-trial.”21 eatee C 237: Afterwards, the Eatee was in a bathtub full of water, so he could bleed to death. The Eater would check up on him every so often. On the video, the Eatee said, “If I am still alive in the morning, let’s cook and eat my testicles.” (www.zug.com/gab/index.cgi?func=view_thread&thread_id=53079, April 2005)

19. For the full description of the examples, cf. Chapter 1. 20. There is one instance of an older use of a sentient eatee which has not been recorded in the relevant linguistic data sources so far): “But toward him the girl had no heart of gratitude because she had no brain of understanding. She, who had been sold for a fat pig, considered her pitiful role in the world to be unchanged. Eatee she had been. Eatee she remained.” (Jack London, Jerry of the Islands, Chapter 4). 21. A rather spectacular case of cannibalism in Germany – apparently with the consent of the victim – that gained worldwide attention. The trial opened in December 2003.

45

46 Heterogeneity in Word-Formation Patterns

If the eatee had backed out of it, it certainly would have been […] No matter the eatee wanted to be eaten, the eater is still in a heap o’trouble. […] (www.invisibleadjunct.com/archives/000418.html, not dated) I did some looking into the case, and it seems the eatee was in on the eating. […] Before the guy passed out (the eatee) he said he had to pee. […] (websleuths.com/forums/archive/index.php/t-4596, November 2004)

Barker (1998: 710) sees “sentience” of the referent as the clearest semantic requirement which must be satisfied by an -ee noun. He also views the sentience requirement of -ee words as a contrast to -er nouns, “since most instrumental uses of -er can easily refer to an inanimate object; for instance, a pager can be either a person who pages someone or the electronic device that delivers the page” (Barker 1998: 710). The problem is that there are also quite a few -ee words that refer to inanimate objects, especially in the technical and instrumental sphere. The following are just a few examples among my new words which have either an exclusively or a predominantly technical meaning: adaptee C 15: Subject-oriented programming and the adapter pattern: Since the Adapter Pattern does not retain state about the adaptee of the adapter in the source, this symptom of Object Schi­zophrenia does not arise here, […]. (www.research.ibm.com/sop/sopcadap.htm, not dated) contaminee C 119: However, contaminees such as soils and fly ashes from waste incinerators often contain a considerable amount of other chlorides, which may act as a main source of chlorine in the formation of PCDD/Fs via thermal processes. (cat.inist.fr/?aModele=afficheN&cpsidt=13898929, October 2002) decodee C 143: This leaves the DF-fragmenter as the stuckee22 until either the bundle is opportunistically DF-decoded (at which time the decodee or any downstream node may elect to take cus­ody) or, more probably (at least the way I usually think about things) until the DF-fragments hit the ultimate destination. (mailman.dtnrg.org/pipermail/dtn-interest/2004-April/001430.html, April 2004)

22. Note also this interesting -ee word-formation that seems to be derived from a past participle (cf. Note in Section 2.2)



Chapter 2.  Phonological, syntactic and semantic constraints

dependee C 159: Instead, let’s mix this “dependent of ” list with the “dependee” list from example 3. If a node sets another node as a dependent of it, the target node stores […] (www.gamasutra.com, August 2003) dominatee C 228: […] the dominatee declares itself as the leader if both its neighbors are dominators, or one of its neighbor is a dominatee but has larger ID. Distributed Construction of Connected Dominating Set in Wireless. (ieeexplore.ieee.org/iel5/7798/21442/00994519.pdf?arnumber=994519, 2002)

Inanimate referent denotation is also a common phenomenon among established words. Exceptions to the rule are also acknowledged in Barker’s (Barker 1998: 710) discussion of “sentience”, yet rather sweepingly dismissed as limited to “sharply circumscribed linguistic and metalinguistic uses” (1998: 711). Indeed, a special category of inanimate (and therefore non-sentient) -ee words is formed by linguistic and metalinguistic terms, e.g. governee, possessee or controllee but, as could be seen above, inanimate and non-sentient items are not restricted to this field. How can such exceptions to this “crucial rule” be explained? One possibility, I would suggest, is a semantic extension of role behaviour in human interaction to technical components, i.e., the sentience is attributed in a figurative sense. Another explanation, and one which has not been considered in any of the -ee descriptions, is the possibility of parallel development between -ee and an implicitly associated -er suffix. There are several similar phenomena in -ee and -er development, one of them being the extension of previously human nouns to instrumental or abstract nouns, as will be shown below (for a fuller discussion of other types of -er suffixes, cf. Section 2.4). It is interesting that Barker views the sentience requirement of -ee words as contrasting with -er nouns (cf. above). As we have seen in Section 2.2, it is evident that -ee nouns are sometimes implicitly or explicitly formed as correlational opposites to an -er noun; whether or not this is more the exception rather than the rule for contemporary formations will be seen in the analysis of the corpus of new -ee words in Section 5.2. Due to the manifest (historical and possibly contemporary) relationship between the -ee and the -er suffix, I will, at this point, insert a brief discussion of the semantic variability of -er, often called an “agential suffix” (e.g. Quirk et al. 1972), which is regarded as almost completely transparent. Portero Muñoz (2003: 132), for instance, claims that – unlike for -ee formations – “a syntactic explanation is plausible for nouns in -er since all the different

47

48 Heterogeneity in Word-Formation Patterns

semantic groups are neutralized by the syntactic function of subject, which simplifies the explanation,” even though she acknowledges some exceptions. In contrast to this, I would like to emphasize that -er formations are far from the straightforward type they are very often claimed to be. Indeed, they have a high potential for ambiguous formations. There are many -er nouns which denote non-sentient entities. One of the most cited is cooker – which is not the person but the instrument; similarly for other cooking terms and other instruments where the noun is not a nomen agentis, e.g. blotter, cutter, roller, poker, two-seater, tooth-picker, etc. (cf. Koziol 1937: 150). Very often, instrumental -er nouns have a personal noun in their ancestry (e.g. boiler, hanger, cutter, cf. also Chapter 3) the meaning of which has obviously been extended to non-human denotation. These non-sentient cases notwithstanding, the idea of animation or sentience does seem to be a relevant criterion at least for the more “typical” cases and will thus be considered in my discussion of prototypical and marginal -ee formation in Section 2.3.2. The notion of sentience is also important for Barker’s framework because it is implicitly tied to another one of his criteria, i.e., “lack of volitional control”: Establishing the sentience requirement is especially important […] since I claim […] that a use of this suffix entails a lack of volitional control on the part of the referent, and it only makes sense to consider volitional control if the referent is (Barker 1998: 711) capable of volition in the first place. 

This point will be considered in more detail in the following section.

(ii) Lack of volitional control Barker (1998: 717) states that “the use of an -ee noun entails a lack of volitional control on the part of its referent either over the occurrence or the duration of the qualifying event itself or (given a punctual qualifying event) over its immediate direct consequences.” In a similar way in which “sentience” somewhat replaces and widens previous concepts of “human” or “person” as part of earlier descriptions of the semantic make-up of -ee words, “lack of volitional control” is related to the notion of “passivity” (in a semantic sense) or “patient” (as part of a thematic role framework). The modification made by Barker has a few advantages in that both elements here, lack of volition or lack of control, need not necessarily mean that the referent is passive in the activity or event referred to. Thus, refugee – one of the -ee nouns which have an agent rather than a patient function – can be better explained in Barker’s terms than in previous frameworks, because the group of persons (or sentient entities) concerned here are an active but involuntary part of the refugeeing action.



Chapter 2.  Phonological, syntactic and semantic constraints

For numerous -ee nouns, the “lack of volitional control” constraint is compelling. The most obvious cases are those where “something bad” happens to the -ee noun referent in the course of the action, as in e.g. muggee, blackmailee, murderee, amputee. Yet, the more jocular formations like kissee, huggee, squeezee also play with the notion of “lack of volition” or “lack of consent.” For -ee words in the legal sphere, such as debtee, obligee, pledgee or trustee the semantic feature concerned is arguably one of commitment and obligation rather than lack of volition. While this criterion rests on a semantic argument, the notion is problematic or at least ambiguous where (syntactically) the -ee formation is the subject rather than the object of the base verb, i.e., where agency is at least syntactically implied. If we take another look at the lexemes discussed in this category in Section 2.3 (iv), we see that “lack of volitional control” explains some of the formations (a) but is contradicted in others (b): (a)

failee resignee signee standee waitee surfee

To varying degrees, a sense of helplessness or powerlessness is implied in these formations, i.e., someone who is compelled to wait or stand for a bus, someone who is forced to resign, etc. Surfee, one of the new -ee words from my corpus, is somewhat special but, nevertheless, implies the “lack of volition” property; in the agent meaning it appears to belong to a whole group of -ee words which have become somewhat fashionable in recent formations, with the meaning “addict of X,” i.e., a surfee is someone addicted or devoted to surfing. Here, the (self-inflicted) “lack of volitional control” is playfully integrated into the agency of the noun as “someone who cannot help doing X.” Possibly, this semantic feature merges with the hypocoristic meaning of the homophonous23 -ie suffix (as in cabbie, kiddie, etc.), a phenomenon that is especially popular in Australian English usage (cf. also discussion in Chapter 6). On the other hand, the element of lack of volition or lack of control is not given in other subject formations:

23. Or, rather, almost homophonous; the non-hypocoristic -ee word surfee – i.e., one whose website has been surfed, or the website itself – receives word-final stress, while, for instance, in the hypocoristic bookie the stress is not placed on the suffix.

49

50

Heterogeneity in Word-Formation Patterns

(b)

attendee dinee expressee parkee

The dinee has control over his or her dining, as do the expressees over their expressing. In attendee – ‘one who (merely) attends a meeting, conference, etc.’ (OED) – a weakening or lessening component seems to come into play, even though the -ee suffix here is clearly not an -ee diminuitive (such as bootee for baby shoes) or a hypocoristic (such as bookie for book maker). Barker (1998: 718) offers two explanations for subject formations with -ee, stating that either “although the qualifying event itself is indeed under the control of the -ee participant (as befitting a verbal subject), there is a clear lack of control over the direct consequences of the event, and the use of the -ee form emphasizes this lack of control,” or “there may be strong entailments showing that even though the -ee participant is an active participant, their actions are nevertheless compelled or constrained by external forces independently of any entailments that are part of the meaning of the stem.” These explanations work with the examples Barker cites (e.g. escapee, retiree). However, as can be seen from our list of words in (b) above, they do not work with all of them. Again, we might state that the criterion “lack of volitional control” may not apply to all formations to the same degree and that there are more prototypical and more marginal formations in this respect. Barker (1998: 719) also acknowledges that this feature might not be an absolute property and one that might be subject to semantic drift: Even if lack of volition is a legitimate constraint on productive formation of nouns in -ee we would naturally expect some discrepancy between the constraint and the observed meanings of well-established uses such as escapee due to semantic drift. That is, because the degree to which uses of a particular -ee noun type conforms to [the “lack of volitional control” hypothesis] can differ little by little, we should expect some established -ee nouns to lose some part of the entailment of lack of volition through a gradual shift in meaning. It would be comparatively less likely for an established -ee noun to come to apply to inanimate objects, or to lose their episodic linking, since that would require a discontinuous change in (Barker 1998: 719) meaning. 

In the discussion of twentieth-century formations (Chapter 4) and recent formations from the web-corpus (Chapter 5), the varying degrees of the property of “lack of volitional control” will be readdressed.



Chapter 2.  Phonological, syntactic and semantic constraints

(iii) Episodic linking Barker’s third semantic constraint for the formation of new -ee words, “episodic linking,” basically rests on aspectual semantic properties of the verb from which the -ee noun is derived. Barker (1998: 711) gives a definition of episodic linking as: “the referent of a noun phrase headed by an -ee noun must have participated in an event of the type corresponding to the stem verb,” e.g. a complainee is only a complainee if he or she has participated (or is even a regular participant) in a complaining event. The -ee nouns formed are episodically linked, i.e., they do not describe an individual-level property of the referent (e.g. intelligence, height, etc.) but a “stage-level” property of an individual, i.e., something that may be true for an individual at one point in time but false at another point in time, for instance, the stage of being a kissee is linked to the duration of the kissing event. Other stages are altered more permanently through the event, for instance, a retiree remains a retiree after the event of retiring has taken place. Barker (1998: 697) notes striking parallels between the episodic linking as a factor in the (usually) deverbal noun formation -ee and the semantic constraints of innovative denominal verbs (e.g. in the sentence, I was bottled by the spectators, in Clark and Clark’s (1979) description), which is not altogether surprising since both types can ultimately be reduced to the pattern ‘the Ved N.’ As such, the sentence above could easily result in a consequent deverbal noun formation like the bottlee. The one other suffix in English word-formation, which can also form episodically linked nouns, is -er. Given the historical and also the frequently semantic connection between -er and -ee, one might suspect a parallel construction here. However, there are several differences between the episodic linking of -ee and of -er suffixes (cf. Levin & Rappaport 1988; Rappaport Hovav & Levin 1992): a. An -er nominal will only be episodically linked if it is interpreted as an agent, e.g. teacher (as opposed to an instrumental, e.g. blotter) b. The nominal inherits detailed argument structure from its stem In contrast, Barker suggests that “episodic linking in -ee nouns is uniform, and does not depend on the syntactic argument structure of the stem” (Barker 1998: 713). Rather, the meaning of the verb and its aspectual properties (punctual versus non-punctual) determine whether or not the event in question causes a permanent or temporary change of state for the participant. Among the -ee words in which the permanence of the change of state can be seen are adoptee, amputee, divorcee, inductee, patentee, retardee, and retiree, and as examples from my corpus might be added, also circumcisee C 88 and convertee C 123. Others are anti-­punctual (e.g. employee, dependee C 159), naturally punctual (e.g. appellee,

51

52

Heterogeneity in Word-Formation Patterns

delegatee, photographee, assassinee C 43, burglaree C 76) or naturally non-punctual (e.g. borrowee, trainee). Potential problems with this criterion of “episodic linking” include the fact that the aspectual properties are connected to the verb and thus exclude -ee formations which are not verb-derived. Nevertheless, Barker (1998: 717) contends that “all that is required in order to satisfy the definition of episodic linking given […] is that the stem be associated with a set of eventualities that can serve as qualifying events, and the attested uses of nominal stem -ee nouns satisfy this requirement.” What about inanimate, “non-sentient” entities? Can they also alter their state or participate in an event? This might apply to a more figurative sense; a decodee C143 might be interpreted as having participated in a decoding event and therefore altered its state. Similarly, an inanimate use of commandee (C 102) in a linguistic description, as in “call these the “(strong/weak) command domain” and the “(strong/weak) commandee domain,” respectively,24 the commandee could metaphorically be seen as participating in a commanding event. Barker’s description of semantic constraints is certainly the most elaborate explanation of the semantic properties of -ee word-formation to date and has gained wide acceptance, despite some obvious exceptions to his model pointed out above.25

(iv) -ee as a thematic role? However, Barker goes even further and argues for a classification of -ee as a thematic role, not quite “on a par in importance with notions like Agent and Patient, which pervade language (or at least, pervade linguistic description)” (Barker 1998: 723) – but nevertheless as a separate thematic role based on the aforementioned semantic criteria. One of his arguments is that the components of -ee noun meanings do not apply to any particular verbal thematic role: “The closest potential match is to thematic roles such as Theme or Patient or Undergoer. However, none of these thematic roles are restricted to sentient entities” (ibid.). Barker furthermore compares the semantic argument of -ee nouns to similar phenomena in other languages, which do not, however, closely match. The idea of linking this word-formation to a separate thematic role thus seems rather far-fetched and has not really been developed. Portero Muñoz (2003), on the other hand, tries to integrate -ee nouns into the semantic continuum of established thematic roles: 24. (www.zas.gwz-berlin.de/mitarb/homepage/gaertner/files/1173797674.pdf) Generalized Trans­formations and Beyond, by Hans-Martin Gärtner, Draft version, December 2001, p. 168. 25. Cf., for instance, Plag (2003: 88).



Chapter 2.  Phonological, syntactic and semantic constraints

Arg of DO 1st arg of do (x, …) 1st arg of pred (x, y) 2nd arg of pred (x, y) Arg of state AGENT

EFFECTOR MOVER ST-MOVER L-EMITTER PERFORMER CONSUMER CREATOR SPEAKER OBSERVER USER

LOCATIVE PERCEIVER COGNIZER WANTER JUDGER POSSESSOR EXPERIENCER EMOTER ATTRIBUTANT

THEME STIMULUS CONTENT DESIRE JUDGEMENT POSSESSED SENSATION TARGET ATTRIBUTE PERFORMANCE CONSUMED CREATION LOCUS IMPLEMENT

PATIENT ENTITY

Figure 1.  Continuum of thematic relations (Portero Muñoz 2003: 139) […] thematic relations are regarded as forming a semantic continuum, whose anchor points are Agent (the wilful, volitional, instigating participant), at one end, and Patient (the non-wilful, noninstigating, maximally affected participant) at the other. In between the endpoints, there is a continuum of thematic relations of more agent-like or more patient-like roles: themes are not affected in the same ways and to the same degree as Patients are, but they are also entities which undergo something, since they get moved around: therefore, they are at the right(Portero Muñoz 2003: 139) hand end of the continuum. 

In her model based on lexical representation of Role and Reference Grammar (cf. Van Valin & Foley 1980; Van Valin & LaPolla 1997), Portero Muñoz (2003) looks at the way semantic macroroles can be related to a number of thematic roles. Macroroles, i.e., generalizations across the argument types found with particular verbs, such as Actor or Undergoer, might prototypically be applied to Agent or Patient, but they might also be Effectors (e.g. Actors with verbs of cognition) or Locatives and Themes (e.g. Undergoers with verbs of position and motion). These marcroroles are then related to verbs of particular lexical domains Portero Muñoz specifies as likely domains for -ee nominalization:26

26. Cf. also Barker’s (1998) list of domains which have “compulsion and obligation” as the overriding features and include legal terminology; personal violence; crime and police work; prisons and punishment; military and war; humor; business; politics, government and bureaucracy; social norms; education; publishing; and communication.

53

54

Heterogeneity in Word-Formation Patterns

– – – – –

Verbs of transfer of possession (e.g. borrow, donate, loan) Speech act verbs (e.g. address, complain, write) Verbs of motion (e.g. arrive, deport, invade) Verbs of position (e.g. stand, sit) Verbs of affecting (a rather vague category which includes any kind of action which has an effect on a second entity, be it physical (e.g. beat, kill) or abstract (e.g. arrest, abandon, divorce)

Portero Muñoz’s model thus also tries to find a systematic way of delimiting the different sets of verbs which are more eligible for -ee nominalization than others, a point which she finds lacking in Barker’s model (cf. Portero Muñoz 2003: 133). It is not transparent what the advantage of the above model might be. It seems as if it merely aims at predicting those types of verbs that might be used to form new -ee words. But how successful is this forecast? One of the most obvious flaws in Portero Muñoz (2003), similar to the problem in Barker’s model, is that she looks at the meaning of the formations only from a synchronic point of view and does not make any distinction between more recent formations and, for instance, sixteenth-century formations. Polysemy and ambiguities in meaning, so typical of -ee words, are not even discussed. Another issue with the model is that it relies completely on the verb derivation of the new forms; noun-noun formation is not discussed as a possibility. In addition, Portero Muñoz’s empirical basis is rather slim as she only relies on scattered examples of already documented words, and does not look for any new formations.

2.3.1

Lexical conceptual structure and co-indexation of affixes

Recent semantic studies (e.g. Lieber 2004; Plag 1998, 1999, 2004) have looked at polysemous affixes using the theory of lexical conceptual structure (cf. Jackendoff 1983, 1990) as the starting point. In these studies, the meaning of a given derived form becomes apparent as a result of the interaction of the semantics of the base with the semantics of the suffix. Here, “the syntactic category of the base can be disregarded because the only restriction necessary is that the base can successfully be interpreted as an appropriate argument in the LCS [Lexical-Conceptual Structure]” (Plag 1998: 227). Lieber (2004) as well as Booij & Lieber (2004) attribute independent semantic content to affixes like -er and -ee. Lieber (2004: 18) notes that this independent semantic contribution goes “beyond an effect on the argument structure of their base.” She develops a “theory of co-indexation” which permits the referential



Chapter 2.  Phonological, syntactic and semantic constraints

properties of an affix to integrate with those of its base. As she states “co-indexation is a device we need in order to tie together the arguments that come with different parts of a complex word to yield only those arguments that are syntactically active” (ibid.). With regard to one of the “problem cases” in -ee formation, Booij & Lieber (2004: 345) offer the noun biographee as an example of co-indexation, which, as the authors claim, is formed on the base of biography: […] This complex noun is formed on the base biography, which is an abstract processual noun having two arguments of its own. The first of these is the “R” argument, and the second argument, which is syntactically realized as the object of an of prepositional phrase in English. The “R” argument in this case is the referent of biography, which is clearly non-sentient, so in coindexing, the affixal argument skips over this one and finds a better match in the second argument, which indeed can be sentient and nonvolitional. The result is a concrete dynamic noun whose referent is sentient, but nonvolitional, as required. There doesn’t need to be a verbal base for biographee to be a “patient” noun of sorts; this reading follows from the semantic content of the affix combined with the conditions on the (Booij & Lieber 2004: 345) coindexation of the affixal argument. 

Their procedure certainly permits us to account for polysemous affixation without considering the syntactic category of the base. But why are some affixes (like -ee and -er) more polysemous than others? What makes -ee formations so particularly heterogeneous, both in a syntactic and a semantic sense? Also, to disregard the syntactic category of the base altogether neglects cases of ambiguity caused by the choice of syntactic pattern (cf. Section 2.2.2), i.e., the question of how the same formation can result in different meanings. How does the speaker select a pattern on which a new word is formed? After all, even though a range of possibilities exists, the syntactic and semantic properties are not represented in an equal proportion.

2.3.2

Ontological classification and prototypes: The question of categories

In my view, neither the syntactic nor the semantic theoretical approaches that try to explain -ee formations by a single compact and synchronic model alone, can account for the diversity and ambiguities that this word-formation offers. The question is whether they should. Rather than adding to the semantic models described above, this study suggests accepting the heterogeneity and polysemy of “natural kinds,” i.e., “terms whose denotation is determined by language-users’

55

56

Heterogeneity in Word-Formation Patterns

casual interactions with exemplars”27 (cf. Carlson 1991; Gross 2006). Instead of trying to confine the formation of -ee words into one model, I will propose placing -ee formation within the framework of such a cognitive view of language users and describe -ee words in terms of semantic prototypes and semantic networks. The idea of semantic prototypes originated in psycholinguistics and can historically be related to psychological research on how cognitive categories are learned by children in the course of their linguistic and cultural socialization (Lyons 1995: 97). Prototype theory rests on the notion that speakers of a language (including experts, such as lexicographers) would have difficulties in specifying the exact properties of a “natural kind” word as, for instance, ‘bird,’ ‘car’ or ‘teacher,’ or classifying individual entities non-arbitrarily into categories of such natural-kind expressions. Words denoting natural and cultural entities are inherently fuzzy and indeterminate. Nevertheless speakers use these words without difficulty since speakers normally operate with prototypes or stereotypes of a particular entity. Natural-kind words have subclasses of more typical realizations so-called “nuclear” or “focal” extensions and, in contrast, there are also “non-nuclear” or “non-focal” classes. The idea of semantic prototypes argues against what has been called the “checklist theory of definition,” derived from the classical, Aristotelian notion of essential and accidental properties, and rejects the notion that “every member of a natural kind […] must possess (in equal measure) all those properties which, being individually necessary and jointly sufficient, constitute the intension of the class and subclass […] to which it belongs” (Lyons 1995: 99). One of the advantages of prototype theory over a more essentialist view of the conceptualization of words is not only the idea of gradualness, i.e., an entity can belong to a particular subclass in various degrees, but also that it allows for differences in conceptualization over time and according to cultural context. By extension, prototype theory can also be applied to word-formation patterns like -ee suffixation. Clearly, there are more central and more marginal examples of -ee words, i.e., some in which the most common and typical features are unified, and others which belong to the class of -ee words on the basis of just one or two traits.

27. There are, of course, various definitions of, and many philosophical debates on, “natural kinds” which, in the scope of this study, cannot be reiterated in detail – the central idea is that they are object-groupings that reflect real distinctions in nature as opposed to arbitrary groupings of objects whose behaviour cannot be described by shared basic properties (cf. Gross 2006).



Chapter 2.  Phonological, syntactic and semantic constraints

For example, the formation – – – – – – –

interviewee with the properties: verb-derived with existing correlative -er noun Direct Object relation to the verb sentient and probably human role participant non-volitional and non-active part in the event can be used in a legal as well as more general contexts

makes a more central -ee word in contemporary usage than, for example – – – – – – –

festschriftee with the properties: noun-derived without existing correlative -er noun Indirect or Prepositional Object relation to the noun sentient and probably human role participant non-volitional and non-active part in the event highly specific context

without devaluing the belonging of festschriftee to the class of -ee words. – verb-derived – with existing correlative -er benefactee noun – Direct Object relation to the verb – sentient and probably human – role participant – non-volitional and non-active part in the event – can be used in a legal as well interviewee, – noun-derived as more general contexts accusee, etc. – without existing correlative -er noun – Indirect or Prepositional Object relation to the noun retiree – sentient and probably human – role participant – non-volitional and non­active part in the event festschriftee – highly specific context

Figure 2.  Prototypical and marginal patterns in -ee formation

57

58

Heterogeneity in Word-Formation Patterns

Similarly, benefactee (noun-derived, with co-existing correlative -er noun, non-sentient and non-human, highly specific context) would also make a rather marginal -ee word, whereas retiree (verb-derived, without existing correlative -er noun, possible DO relation to the noun, sentient and probably human, role participant, non-volitional or non-active part in the event, general context) would be more central than festschriftee but less typical than interviewee, accusee, and similar examples. As pointed out above, the specific entities which make focal and non-focal extensions of -ee-ness do not necessarily have to be the same at all times and in all cultural contexts. As we shall see in Chapter 3, a prototypical -ee word in the sixteenth century had properties rather different from our core example above, i.e., interviewee. For contemporary usage, the prototypical -ee word in Australian English might also be different from one in British English – a point to be discussed in Chapter 6.

2.4

Conclusion: Heterogeneity, polysemy and ambiguity revisited

In Chapter 2 of this study, several possible reasons for the heterogeneity of -ee word-formations have been offered. One might summarize some of the sources for the frequent heterogeneity and ambiguity in -ee nouns, as we have explored it so far, as follows: – Heterogeneity and ambiguity through changing and parallel rules in the development of -ee verb-derivation (Direct/Indirect Object, Subject etc. cf. Section 2.2.1). – Heterogeneity and ambiguity through underlying noun-derivations from -er words: if -ee words can also be formed as correlative nouns from -er words (as we have seen in Section 2.2.2), the correlation may sometimes be extended to include non-human, immaterial, etc. words. This strikes me as particularly relevant with regard to technical descriptions/tools (cf. discussion on sentience/lack of sentience in -ee and -er nouns in Section 2.3). With regard to the latter point, the possibility of a parallel development might be used to explain other “non-typical” or “marginal” -ee formations, for instance, with regard to agent/patient behaviour. In addition, there are human nouns formed with an -er suffix which are not agent nouns; examples include Londoner, New Yorker, islander (location markers) or widower, where the suffix is a masculine gender marker derived from a female person reference. Quirk et al. (1972: 998) note that -er also has ‘passive’ meaning in a few nouns, e.g. cooker can mean in



Chapter 2.  Phonological, syntactic and semantic constraints

British English ‘an apple for cooking’ (Cf. OED: 1887 “They are a large, juicy apple, agreeable to eat and splendid cookers”). Sometimes an -er word can equally denote agent and patient; for instance, milker can be both the person who milks a cow or the cow itself (like in OED 1802 “They [Yorkshire cows] are excellent milkers”). Apart from the classic verb-derived human noun (agent or non-agent) formation with -er, there are various possibilities for -er formation (cf. also Marchand 1969: 273–281 for various types of -er suffixation): – Animate but non-human (e.g. pointer, retriever, trotter, etc.); – Agent is a device or tool (e.g. blotter, eraser, fertiliser, knocker, poker, stopper, toaster, voucher), or – Immaterial agents (e.g. reminder, eye-opener, thriller; cf. Marchand 1969: 274– 275) – Patients (e.g. cooker, milker, cf. above) – Adverbial complement and object types (e.g. slipper, wrapper, jumper, drawer– these are all articles denoting clothing) – Words denoting place of activity (e.g. boiler, locker, counter, dresser → the latter two from French comptoir, dressoir) – Agent nouns for which there is no English verb-base, especially with the variant -or (e.g. doctor, author). (cf. also Baeskow 2002: 272; Lieber 2004: 17) These are two of several possible roots of the versatility and possible haziness of word-formation as we have explored it thus far on a (largely) synchronic level. What seems clear to me, however, is that the heterogeneity and ambiguity of -ee formations cannot be investigated in isolation, nor can one afford to omit a diachronic perspective. While advocating a description of the syntactic and semantic properties of -ee words in the framework of prototype theory (Section 2.3.2), I do not suggest that it offers per se an explanation for the high variability. In other words, the description of the properties of -ee words and the causes for the variety of patterns are treated as two separate issues. What it does provide is a first step towards the development of a theory of morphological networks (cf. also Schröder 2008: 232) where creative neologisms that deviate from the prototypical formation type can activate analogical coining and eventually create new rules. The description of the history of -ee words in Chapter 3 will show how new formations can be based on a number of syntactic and semantic possibilities which have arisen in the 600 years of the development of -ee words through an interplay between analogical coining and rule-governed production. The diachronic overview will also include factors such as language contact as a cause for the creation of marginal coining in the various periods.

59

60 Heterogeneity in Word-Formation Patterns

On the one hand, my analysis will illustrate how the semantic and syntactic possibilities for the formation of -ee words have increased. On the other hand, it will also provide data for movements of particular syntactic types (e.g. Indirect Object formations) from margin to centre in the sixteenth century and back to the margin in the nineteenth century. Ultimately, heterogeneity and polysemy in -ee words bear witness to a crucial fact in word-formation theory (an issue which will be discussed in more detail in Chapter 4), i.e., that word-formation is not merely a generative morphological process which follows a narrow set of predictable rules but that it is also influenced by the lexicon, that is, by the set of already existing, often irregular, actual words of that word-formation type as well as other competing influences outlined above. Words, as Mark Aronoff (1976) points out in his early work on Word Formation in Generative Grammar, persist and change and do not always behave according to the logic of morphology: But words are peculiar, not only in that not all of those that should exist actually do, but also in that those that do exist do not always mean what they are supposed to mean, or even look like what they are expected to look like. Words, once formed, persist and change; they take on idiosyncrasies, with the result that they are soon no longer generable by a simple algorithm of any generality. The word gravitates toward the sign. The actual words of a language, the members of the set of dictionary entries, are as a result not a subset of the items which are generated (Aronoff 1976: 18) by a regular morphology […]. 

The particular relationship between morphological rules and actual words will be borne in mind in the diachronic exploration of the development of this word-formation in Chapter 3 with the theoretical topic of language change. In Chapter 4, I will readdress the notion of morphology and the lexicon so as to discuss productivity and creativity in -ee word-formation in the twentieth century; I will then embark on my empirical test of possible (or potential) versus actual words and their analysis in Chapter 5 and (for varieties of English) in Chapter 6.

chapter 3

The career of -ee words A diachronic analysis from medieval legal use to nineteenth-century ironic nonce words

For most of the twentieth century, as Jean Aitchison (1981: 48) remarks, “synchronic linguistics was considered to be prior to diachronic linguistics,” with diachronic linguists relying largely on the previous work of synchronic linguists. However, by ignoring the “messy bits,” synchronic linguists often leave out those parts of their description that are essential for understanding language change. In Chapter 2, we have seen that a synchronic perspective on the syntactic and semantic patterns of -ee word-formation is limited and thus unsatisfactory. Without a historical understanding of the word-formation we might therefore ask, “why do what look like functionally compromised architectures (irregularities, apparent ‘irrational’ complexities) persist?” (Lass 1997: 15). No existing synchronic model can fully explain the ambiguities and their sources in the formation of -ee nouns and several models are even misleading when, for instance, words which have undergone meaning changes are interpreted from a static point of view alone. In this chapter, I will argue that a diachronic analysis is necessary to further explore the various layers of -ee formation and to explain the influences on the “fuzzy edges” of meaning and formation patterns. In this sense, I follow Lass’s (1997: 14) view that Synchronic structure and historical origin (of course) interact; but the state of the language as a whole is not fully describable in purely synchronic or functional terms. Portions of apparent ‘synchronic’ states are relics of the historical processes that brought them into being, evolutionary scars on the present-day body. The two dimensions are complementary, but in the end history probably has more to say about synchrony than the other way round. An extreme (but largely true) statement is Lehmann’s dictum (1952: 23n) that ‘the only explanation for a linguistic form is an older form.’

The diachronic perspective presented in this chapter also emphasizes that productivity in a word-formation which has evolved over a period of several centuries

62

Heterogeneity in Word-Formation Patterns

must be analysed with regard to different stages in its development. This section will thus also provide a basis for the discussion of diachronic and synchronic productivity in Chapter 4.

3.1

Towards a contextualized history of -ee suffixation

Lexical growth in the development of the English language has been achieved to a considerable degree by extensive borrowing from other languages, especially in the Middle English and Early Modern Periods. During these periods the vocabulary increased tremendously, not only through borrowed lexis itself but also through new productive elements that owe their existence to borrowed lexis. Affixation with foreign elements, for instance, added to the already existing stock of native resources for word-formation, often resulting in competing patterns (e.g. native prefix un- versus rival a-, dis- in-, or non-; cf. Nevalainen 1999: 332). Suffixation with -ee is one such productive element which came into the language as a consequence of borrowing and became an indigenous word-formation pattern. In the following, I will briefly outline the origin and development of the -ee suffix. Because of its implicit association with the -er suffix, I will also point out some contrasts or parallels with -ee, without intending to give a full account of the complexities of -er affixation. Compared to the older human noun derivation in English with -er, -ee suffixation entered the language more recently and can be traced back to the Middle English Period. The -er formation as an agent noun has Germanic roots; the -ere ending can be found in large numbers in Old English (e.g. bæcere – ‘baker,’ bōcere – ‘scribe,’ cwellere – ‘killer,’ fiscere – ‘fisher,’ fugelere – ‘fowler,’ leornere – ‘learner,’ etc.). The ending in Old English denoted male person reference, a meaning which changed slightly in Middle English, when -er(e) started to refer to both male and female persons, e.g. thredere – ‘(male or female) threader’ (cf. Koziol 1937: 150). In contrast to this, -ee suffixation in English is French-derived and first appeared in nativized French loanwords towards the end of the Middle English Period, in the fourteenth and fifteenth century. By then, the Middle English diglossic . It is estimated that in the Early Modern Period, 45–50% of the new vocabulary recorded were loan words (Wermser 1976, as quoted in Nevalainen 1999: 332), as opposed to approximately 3% borrowed lexis in Old English. . This is, of course, only one possibility of several word-formations with -er in different periods and with various sources. Cf. Marchand (1969: 273ff.) for an overview of different types of -er suffixation.



Chapter 3.  The career of -ee words

language situation, with French as the language of government, law and education, and English as the language of the wider population, had begun to crumble, and English started to enter domains formerly reserved for the high prestige language French. After 1362, English was used at court, Parliament opened in English, and English became the language of instruction in schools. These institutional changes, initiated by a change in language politics, must be regarded as a milestone towards re-establishing English as the principal language in England. Because of the long period of diglossia, many words belonging to the high prestige domains, or which were used in a more formal linguistic register, were borrowed from French and gradually incorporated into English. Borrowings from French and Latin also included nouns ending in -er, -our and -or (e.g. appelour, …) and this phonological coincidence of an already existing Germanic ending with a French and Latin-derived one certainly gave a boost to the productivity of the -er suffix from the Middle English Period on (cf. Koziol 1937: 150; Marchand 1969: 275), making it one of the most prolific suffixes in English and one which is based on a range of formation patterns and meanings. As we shall see below, French-derived -ee nominalizations were restricted at first to the rather specific domain of the language of the law until the semantic range of usage of this suffix began to widen. Kastovsky (1986b), in his discussion of the interaction between word-formation patterns and the semiotic-communicative functions of word-formation, points out that the first loans and new -ee formations in this legal sphere had a purely labeling function. Unlike other new productive elements (cf. negative prefixes above), -ee did not have a native rival or competitor from the existing native morphological stock as there is no real semantic predecessor. As Nevalainen (1999: 396) notes, “the adoption of the passive benefactive suffix -ee in Early Modern English marks the only significant addition [to the main semantic types of English deverbal nouns].” However, -ee suffix is not the only suffix that first appeared in Middle and Early Modern English to accommodate French and Latin legal terms. The participial suffix -ant/-ent like attendant (1555), dependant (1588), etc. were formed in Early Modern English as both personal and instrumental deverbal nouns (cf. Nevalainen 1999: 396). The figure below, taken from Görlach (1999: 462), shows how the hegemonic French and Latin languages dominated the legal domain more than others, but were being rapidly replaced by the emerging Standard English after the changes . Known as the Statue of Pleading, initiated in 1356 and enacted in 1362 (cf. Baugh & Cable 2002: 149). . As opposed to later formations, which also engaged in the process of syntactic recategorization; Kastovsky sees these two functions – labelling and syntactic recategorization – as the two end points of a functional scale (Kastovsky 1986b: 412).

63

64 Heterogeneity in Word-Formation Patterns

English dialects

law documents

poetry, literature

French

Latin

English dialects French

Standard English

Standard English

Latin French

scholarly texts

Latin

Standard English

Standard English

spoken

English dialects Latin

French

700 800 900 1000 1100 1200 1300 1400 1500 1600 1700

Figure 3.  Taken from Görlach (1999: 462): English dialects, Latin, French and Standard English in varieties of written and spoken English

in language politics in the mid-fourteenth century – a Standard English which is, however, marked by extensive lexical influence from its predecessors. One can conclude from this dissolving diglossia that English suffixation with -ee is basically the result of a language-contact situation, one of the primary sources of language change. The borrowing of French legal terms and consequent adoption of the -ee affix for productive English word-formation was motivated by external social forces, i.e., by a change of language politics and the subsequent need to fill lexical gaps in this particular domain previously occupied by French. During the now relatively long history of -ee words, starting from the late Middle English Period until the early twenty-first century, the formation has undergone quite a few changes, which will now be examined in some detail. While such changes can not be exactly pinpointed to a specific place in time, I will, for reasons of convenience and to provide a clearer overview, take a look at the new entries of -ee words in each century for the first five to six hundred years of usage, from the fourteenth century until the nineteenth century. The more recent . Fourteenth and fifteenth century will be taken together as there are very few (4) entries from the fourteenth century.



Chapter 3.  The career of -ee words

formations, i.e., twentieth-century new -ee words, will be analyzed in Chapter 4 together with a more general theoretical discussion on productivity. The reason for this separation is threefold: a. In order to place emphasis on the difference between diachronic and synchronic productivity and to make visible syntactic and semantic changes over the course of the development of -ee which have been neglected in earlier (purely synchronic) descriptions of -ee formations; b. The discussion of productivity in the twentieth century offers a forum in which to develop the very recent (mainly late twentieth and early twenty-first centuries) formations of my web-corpus analysis; and c. The documentation of twentieth-century sources is more diverse than for the previous centuries discussed in this chapter; because of its relative temporal vicinity, my twentieth-century documentation is not only dictionary-based (e.g. Lehnert 1971; Muthmann 1999; OED) – dictionaries notoriously lag behind in including neologisms – but will also include new examples found in articles and other scholarly works (e.g. American Speech; Barker 1998; Bauer 1983, 1993, 1994). The source for my chronological lists of -ee words and their first attested use between 1330 and 1899 is the entries in the Oxford English Dictionary. Despite some limitations (cf. Nevalainen 1999: 337) the OED can still be seen as the best possible source for a diachronic reconstruction of vocabulary, especially for the earlier formations (cf. also Mair 2004). Legal dictates are mainly written texts; when -ee words were employed in this sphere, it is expected that they are documented in the OED. Nevertheless, the caveat of chance documentation in the analysis of historical data remains – as is often the case in studies which deal with historical material. While completeness of the total of -ee word findings cannot be guaranteed in the subsequent lists, the figures for new entries in the various centuries closely correspond to or even exceed those given, for instance, in Bengtsson (1927), and reiterated in Barker (1998). . Cf. William Labov (1994: 11) on this issue: “Historical documents survive by chance, not by design, and the selection that is available is the product of an unpredictable series of historical accidents. The linguistic forms in such documents are often distinct from the vernacular of the writers, and instead reflect efforts to capture a normative dialect that never was any speaker’s native language.” For the present study, however, these limitations do not weigh as heavily as in many others, especially in the early history of -ee words when they (presumably) appeared mainly in legal texts, i.e., written documents which are often preserved over centuries. . Barker (1998: 701): “By the count of Bengtsson (1927, rear matter) supplemented by my own collection of naturally occurring examples culled from primary and secondary sources,

65

66 Heterogeneity in Word-Formation Patterns

3.2

Anglicized Law French beginnings

For the socio-historical reasons described above, suffixation with -ee first appeared in English in legal contexts, where English steadily replaced French as the dominant language of that domain. Even though this political language change was enacted very rapidly, the linguistic change that followed as a consequence was a slow process. Marchand (1969: 267) states that “although by the decree of 1362 English was officially established as the language of jurisprudence, French continued to exercise a dominant influence, and legal English was for centuries little more than anglicized Law French.” Originally a substantivized French past participle (-ée for feminine and -é for masculine words), legal -ee words progressively gained entry into the English lexicon as loan words. The OED lists at least 14 entries for -ee words for the fourteenth and fifteenth centuries, 16 if one counts new attested meanings of a word as a neologism. As can be seen below, the spelling in the first OED entry of an -ee word, ordiné, is still unassimilated and resembles the Old French ending -é as past participle noun: 1330: # ordinee (OF, ME: ordiné) In ME. a. OF. ordiné, pa. pple. of ordiner to ORDAIN; in mod. use formed anew: see -EE. †A. adj. Admitted to holy orders, or into a religious order or fraternity; ordained. Obs. c1330 R. BRUNNE Chron. (1810) 225 an went is Ottobone orghout e cuntre, & quaynted him with ilkone, lewed & ordine. c1400 Rule St. Benet (E.E.T.S.) 22/7 Princlike sal she sende an ordane nunne till her at is in sentence. B. n. An ordained clergyman or minister; now, usually, a newly-ordained deacon. c1330 R. BRUNNE Chron. (1810) 210 [To] at holy kirke, & alle e ordinez, & bisshop wo ei wirke & clerkes of dignitez. 1863 A. BLOMFIELD Mem. Bp. Blomfield I. iv. 106 In addressing his ordinees on the subject of amusements, he has deprecated fox-hunting. 1884 Ch. Times 428/4 There was a falling-off in the number of ordinees at the beginning of the decennium.  (OED) the progressively increasing number of new uses counted per century are given […]”: Century endpoint 1500 – number of types (11); 1600 (21); 1700 (26); 1800 (30); 1900 (100); 1997 (196). The figures Barker cites here add up to 384 items, yet he inexplicably speaks of “470+ (total).” . # will be subsequently used to mark those formations which can be interpreted as direct object formations of the verb. In these very first (fourteenth and fifteenth century) examples, noun-noun derivation must be assumed. However, the cited -ee formations also behave as if they could be the direct objects of the verb (in the French past participle tradition).



Chapter 3.  The career of -ee words

Table 4.  Word list – fourteenth and fifteenth century (OED entries) 1330: 1351: 1380: 1387: 1411: 1419: 1424: 1442: 1467: 1480: 1480: 1481: 1485: 1491: 1495: 1498:

# # # # #

#

ordinee (OF, ME: ordiné) presentee (OF example) vowee appellee feoffee (1) assignee (1) enfeoffee patentee assignee (2) espousee confeoffee lessee vouchee grantee committee presentee (E example)

As Nevalainen (1999: 369) points out, French loan words from this period do not differ greatly from their sources. Subsequent entries rest on the presence of correlative word pairs in OF (Old French) legal contexts such as apelour versus apelé, which were consequently adapted in English as ‘appellor’ versus ‘appellee’ (1387). Productive word-formation in English rather than a nativization of French loan words starts with the formation of new word pairs where -or (for the agent noun) and -ee (for the patient noun) were freely added to English verb stems to form nouns.10 Kastovsky (1986b) notes that, despite the first -ee words in English having, for the most part, a naming function, a syntactic-semantic function was also fulfilled: i.e., it denoted that participant in an act of transaction that is normally represented as indirect object (Dative/Goal in a Case Grammar). This paved the way for an extension of this pattern into the area of syntactic recategorization, especially

. A possible exception here is vowee – ‘advocate, patron,’ i.e., one to whom a vow has been made (OED example: “c1380 Sir Ferumbras 405 Ne were it for repreue, By Mahoun, at ys my vowee [v.r. vowe], of yn heued y wolde ee reue)” which seems to have been without a correlative agent noun at the time. The first occurrence of vower (which largely shares the meaning of vowee – ‘guardian, patron’) is listed in 1470, almost 100 years after vowee. 10. Unlike -ant/-ent, also a suffix first used in the legal domain, which does not operate on native bases in Early Modern English (Nevalainen 1999: 396).

67

68 Heterogeneity in Word-Formation Patterns

since English did not possess any regular means for expressing a recipient in the same way as it could express the agent, viz. by a derivative in -er.  (Kastovsky 1986b: 415–416)

The productive use of -ee on English stems makes it principally a new English word-formation. However, the exact point in time when this happened is difficult to determine. After all, many of the potential English verbs, and particularly those used in the legal sphere, are also French in origin. While we can see signs of an independence from the French origin in the fifteenth century – e.g. in patentee (1442), the etymology is given as “[f. patent n. + ee […] perh. first in an Anglo-Fr. form patenté]” (OED), no real existing French past participle is cited as a source. In confeoffee, a prefixation (also a Latinate in origin), is added to the Anglo-French feoffé (past participle of feoffer), thus including another productive element. Such processes of independent word-formation do not exclude the possibility of occasional borrowing from similar French patterns continuing. Suffixation with -ee on an English verb stem of Germanic origin does not take place until 1611 (writee). However, this may be due to the profusion of Latinate vocabulary in the specific register in which -ee words appeared rather than a stringent dependence on the French origin. This can also be seen in the grammatical changes that the new (English) -ee words underwent; the derivatives in -ee, unlike the Anglo-French participial nouns on which they were modelled, typically do not have a grammatically passive sense, but frequently denote the indirect object of the verbs from which they are derived. Thus, unlike their original model, English -ee nouns are not necessarily the direct object of the verb but often (especially at first) denote an indirect or prepositional object relation (benefactive) to that verb. The above list (Table 4) is divided on the one hand between formations in the French past participle tradition, such as ordinee – cf. above, presentee – ‘a person presented’ or assignee (1) – ‘one who is appointed to act for another; a deputy, agent, or representative’; all of these items could be seen as direct objects of the related verb (marked in the list with #) if these first -ee words had, in actual fact, not been loaned or formed as a correlative noun to the -er word. On the other hand, we have the more progressive formations which are further removed from the French past participle tradition in that they are indirect and prepositional object formations based on verb derivation. For example, the meaning of ‘patentee’ is not the object that has been patented, but “one to whom letters patent have been granted; […] now esp. one who has taken out a patent for some new invention, or the like” (OED). The shift from direct object to indirect or prepositional object meaning becomes most evident in cases where both have been applied successively. While



Chapter 3.  The career of -ee words

the first denotation of assignee (1419) has a direct object meaning,11 a few years later (1467) it is used in the indirect object meaning as ‘one to whom a right or property is legally transferred or made over’ (OED).12 Almost all of the newer formations are of this latter type; with the exception of presentee (2), all fifteenthcentury formations after 1419 have an indirect object meaning. It can be argued that the distancing from the French pattern and the consequent disassociation from (continued) French borrowings made the nativization process of this new English word-formation pattern possible. In the relatively short time span when this particular process of language change can be observed, we seem to have a rather typical interplay between two possible paths in the formation of new words, i.e., between the lexicon and morphology: a) new formation by analogy (i.e., formed on the basis of a specific item, from the lexicon or “actual words”) and b) by rule (i.e., formed on the basis of a general rule, that is, morphology). As Lehrer (1996: 67) points out, “what happens very often is that a neologism may originate by analogy, but it very quickly (after 2 or 3 new words) generates a rule so that additional neologisms need not be based on any specific word.”

3.3

English indirect object formations in the sixteenth century

The following words (Table 5), which were coined in the sixteenth century (OED, my search), all reflect this move to indirect passive noun formation, with the exception of electee. Donee, for example, does not refer to the object that has been given but “one to whom anything is given; esp. in Law, (a) one to whom anything is given gratuitously; (b) one to whom land is conveyed in fee tail; (c) one to whom a ‘power’ is given for execution” (OED). This is also the case for formations where the underlying verb would allow a person as direct object, as in recognizee. The sixteenth-century meaning in the law context is, however, “the person to whom one is bound in a recognizance” (OED) – usually of land rights.13

11. As in the previous examples, assignee was formed as a correlative noun. 12. Note that assignee becomes a direct object formation again in the nineteenth century, meaning ‘a convict assigned as unpaid servant to a colonial settler. Hist.’ (OED). Example: “1843 Penny Cycl. XXV. 139/2 It is comparatively difficult to obtain another assignee, easy to obtain a hired servant.” This usage of assignee has been particularly prominent in Australia, cf. also Chapter 6. 13. Example (OED): 1592 WEST 1st Pt. Symbol. §41 G, When the moietie of the Recognisours landes bee delyuered to the Recognisee.

69

70 Heterogeneity in Word-Formation Patterns

Table 5.  Word list – sixteenth century (OED entries) 1523: 1528: 1531: 1531: 1531: 1531: 1531: 1531–1532: 1537: @14 1540: 1542: 1542–1543: 1547: 1574: 1574: 1581: # 1584: 1590: 1592: 1593: # 1598:

donee bailee alienee appellee (E. version) debtee cognizee (Old Law) prayee recoveree (Law) absentee (Gen.) disseisee/-ze feoffee (2) devisee vendee discontinuee (Law) obligee (1. Law) indictee (1) mortgagee obligee (2. Gen.) recognizee electee bargainee (= purchaser)

Summary: Number of words Number of direct object formations # Number of nouns with agent meaning @

21  2  1

The list above (Table 5) demonstrates the considerable syntactic and growing semantic variety in which the -ee formation is applied in the sixteenth century: – Most words are still formed as correlative passives to agent nouns (cognizee to cognizor). For some, it is evident that the verb is used as the basis of derivation (for instance, obligee – as the agent noun obliger is formed considerably later, in 1650 – OED), in one case (absentee), it has been suggested (Bauer 1983: 244) that the new word might be adjective based. The latter case is somewhat controversial in origin as well as in its status as an agent noun. Quite possibly, this formation is based on the Law French verb s’absenter (cf. Marchand 1969: 268). A look at the first entry in the OED suggests that its 14. @ stands for agent noun.



Chapter 3.  The career of -ee words

origin is, in fact, a French loan word: “1537 in BLOUNT Law Dict. (1691), Absentees or des Absentees, was a Parliament so called, held at Dublin, 10 May, 28 H. 8” (OED). – Some (2) words have a direct object meaning (indictee, electee). They could either be formed according to the old pattern or, more likely in the sixteenth century, they could have started out as loan words from French and were subsequently integrated. After all, borrowing from French reached its peak between 1570 and 1620 (cf. Nevalainen 1999: 368) and French continued to be an important and ever-growing influence, especially in administration and law: In Early Modern English [French loan words] mirror England’s cultural and political contacts with France, as well as the influence of French emigrants, who settled in England in the late sixteenth and seventeenth centuries. The fifteenth and sixteenth-century loans no doubt in part continue to reflect the role of French as a language of administration and law, but much of the seventeenth-century variation can only be explained in terms of Anglo-French relations, which were revived during the Restoration, after the various tensions that had existed between the two countries since the 1620s were relaxed.  (Nevalainen 1999: 368)

However, with the two exceptions above most sixteenth-century words relate to the indirect or prepositional object (recoveree 1531 “Law. The person from whom some property is recovered; spec. the defendant in an action of common recovery”).15 We might also connect this diversity to a gradual change in function of this word-formation, from a chiefly labelling function (e.g. in loan words such as cognizee) to a syntactic recategorization (cf. Kastovsky 1986b), where information contained in a sentence is converted into nominal form (e.g. in recoveree). – Absentee does not fit the general pattern, as it does not have a passive meaning. The coinage of absentee is, in fact, the first instance of an English -ee word where the referent has a subject rather than an object meaning (‘One who is absent, or away, on any occasion’), a type of formation which increases in subsequent periods. Whatever the precise syntactic pattern of the formation may be (cf. discussion above: adjective-based, based on French loan words – the noun Absentee, or the verb s’absenter, etc.), the effective use of this coinage in various contexts marks the beginning of a new pattern. It is, at least in one

15. As opposed to its twentieth-century meaning (1957) “One who is recovering from a disease or an illness” (OED).

71

72

Heterogeneity in Word-Formation Patterns

case (presentee in a jocular nineteenth-century meaning),16 an obvious and indisputable model for new formations; in other cases it might have served as an implicit exemplar for subject patterns in -ee formations. This supports the general idea, already outlined in Section 2.4, that analogical coining plays an important role in the diversification of syntactic and semantic patterns in -ee derivative formation. – The semantic context in which the words are used is still largely but not exclusively the legal domain: absentee or obligee (2) have a more general meaning and seem to have also been used outside the legal sphere.

3.4

Growing diversity in the seventeenth century

The development in the seventeenth century sees a strengthening of the verb as the base of the formation. Direct object formation increases possibly under the influence of continued borrowings from French. Legal usage of -ee words is still prevalent, but we note further differentiation as far as the context’s meaning is concerned (Table 6). – Again, we can observe a generalization of meaning in some of the nouns which were previously used in institutional contexts only, mainly in legal but also religious domains (cf. devotee (1) 1645 and devotee (2) 1657). It is interesting here that the development is not uni-directional (from legal to more general use) but is also recorded for cases where the legal use is preceeded by a more general use (trustee (1) 1647 and trustee (2) 1653). The latter case in particular shows that -ee suffixation has now become productive in a wider context. The formation of writee – ‘one for whom something is written’ (Example: c1611 CHAPMAN Iliad XIV. Comm., Where a man is vnderstood, there is euer a proportion betwixt the writers wit and the writees) fits into this pattern. Noddee – ‘person causing drowsiness’ (Example OED: “a1680 H. MARTYN in J. Aubrey Brief Lives (1898) II. 46 Mr. Speaker, a motion has beene to turne out the Nodders; I desire the Noddees may also be turn’d out”) is unusual in that we do not have the recipient but the cause of the action.17 This formation 16. OED: ‘jocular [f. PRESENT a. (adv.) in imitation of ABSENTEE.] One who is present. Hence presen'teeism’ Example: ‘1892 ‘MARK TWAIN’ Amer. Claimant xxi. 211 There was an absentee who ought to be a presentee – a word which she meant to look out in the dictionary’. 17. Note that later, the word is also used in the passive sense, i.e., ‘a person who is nodded to’ (1810, OED).



Chapter 3.  The career of -ee words

Table 6.  Word list – seventeenth century (OED entries) 1600: 1601: 1602: 1602: 1602: 1602: 1602: 1610: 1610: 1610: 1611: 1611: 1611: 1611: 1614: 1616: 1616: 1620: 1621: 1624: 1627: 1644: 1645: 1647:

(#)19 @ # # @ @ # # # # @

confirmee (Law) confessee conusee compromittee co-patentee protectee seducee investee interessee obligee (3)20 resignee (1)21 sequestree submitt(i)e writee (rare) cheatee challengee patentee (2)23 complementee referee (1, Pol.) representee (1) garnishee representee (2) devotee (Rel.) trustee (Gen. Obs.)

1653: 1654: 1654: 1657–1683: 1661: 1662: 1664: [1666: 1670: 1672: 1676: 1676–1677: 1679: 1679–1688: 1680: 1683–1685: 1685: 1687: 1688: 1690: 1691: 1691: 1697:

# @

@ # # †24

trustee (Law)18 testee (a witness) vouchee (Gen.) devotee (Gen.) debauchee surrenderee nominee (1) bargee] referee (2: umpire) letteree (Law) preferee depositee guarantee (Person) legatee noddee (1)22 pawnee refugee transplantee nominee (2: for an office) referee (3, Law) patentee (3: invention) advowee nominee (Law, inheritance)

Summary: Number of words Number of direct object formations # Number of nouns with agent meaning @ Number of non-person references

47  9  6  1

18. Cf. also US specification. 19. OED: “A. One who is confessed (by a priest). B. One to whom confession is made. (Ambiguous and to be avoided)”. 20. Here: ‘person who is under obligation on account of benefits received’. 21. Here: ‘one to whom anything is resigned.’ 22. ‘Person causing drowsiness’ (OED). 23. Here: inventor and proprietor. 24. † stands for non-human nouns.

73

74

Heterogeneity in Word-Formation Patterns



already gives a taste of whimsical or humorous word formations which become so fashionable in the eighteenth and, in particular, the nineteenth centuries. Here, correlative noun pairs, used for so long in the legal environment, are imitated and adapted to another context for humorous effect. – Object role, the core feature in earlier formations (with the exception of absentee in the sixteenth century), is not given in all seventeenth-century -ee words. Both sequestree (1611) and submittie (1611) are actually the subjects of the verb, i.e., synonymous with the sequestrator and the submitter. Within the course of 20 years, the meaning of representee moves from direct passive noun (“one who is represented,” 1624, OED) to agent (“a (parliamentary) representative,” 1644, OED). The base of the formation of refugee seems somewhat unclear as there is no English verb ‘to refuge,’ nor an English correlative noun ‘refuger.’ The historical background suggests that here again we have a borrowing from French, possibly from the French verb se réfugier. The original word meaning is given as “one who, owing to religious persecution or political troubles, seeks refuge in a foreign country” and was originally applied to the French Huguenots who came to England after the revocation of the Edict of Nantes in 1685. Thus, in the first entries in the OED, the word is used in this context: “1685 BURNET Tracts (1689) I. 27 Zurich demanded the Estates of the refugies. 1691 Lond. Gaz. No. 2679/1 The Troops in the Town behaved themselves very well, and particularly the French Refugies. 1709 STEELE Tatler No. 13 2 That all the French Refugies in those Dominions are to be naturalized” (OED). Refugee is another instance of an -ee word with an agent rather than patient meaning.25 Bargee is usually categorized as belonging to a different type of word-formation with -ee, which Marchand (1969: 268) describes as a “jocular formation with the lengthened hypocoristic -y, -ie” and Bauer (1983: 244) sees it as a “diachronic diminuitive.” Bargee – bargeman is classified here in line with, for example, coachee (1790) which can also be used in a passive way (“one who is ‘coached’ ”) in a later meaning. The two (first of all unrelated) types of suffixation with -ee thus seem to interact in some words, in that speakers might reinterpret already existing formations as belonging to the other types. The OED lists bargee among the words in which “the suffix is employed app. arbitrarily.” – As a third interesting development – after generalization of use and loss of passive character in some words – one can notice that in these seventeenth25. It is conceivable that refugee is, in fact, the prototypical model in which the semantic feature “non-volitional” was established and, first via formation by analogy, reproduced in other newly formed words.



Chapter 3.  The career of -ee words

century formations, patentee – one of the oldest -ee words in English – is applied here for the first time (1691) to a non-human sense, i.e., the patentee becomes the invention itself rather than the “person to whom letters patent have been granted a patent” (1442) or “the inventor, propriator of something” (1616). There are several possibilities to explain this semantic extension of -ee to nonperson referents. 1. It may be due to general processes of semantic change. Generalization, specification, metonymy, metaphoric extension and parallelism are all rather common phenomena on the semantic level of language change. Regarding the example of patentee as a non-human noun, a generalization of meaning seems to be the most likely process in question. 2. It may be due to continued influence of borrowings from French where the past participle of the verb, similar in form to the English word-formation with -ee, is not restricted to human reference. 3. It may be due to the influence of parallel -er formation; we can also assume a certain parallelism to the -er formation pattern if we consider the development the erstwhile correlative agent noun with -er suffixation underwent. In many -er words, the agent is actually a tool, device, implement, machine or the like, sometimes by semantic extension from a human noun: boiler, for instance, changed its meaning from ‘one who boils (anything)’ (1540) to ‘a vessel in which water or any liquid is boiled’ (1725) (OED). At times, human and non-human noun meanings are coined almost simultaneously, e.g. blotter has a human reference in 1601 as “a. one who, or that which, blots; b. a scribbler, a sorry writer; c. one who stains or defiles […]” but in 1591 it was already used as “a thing used for drying wet ink marks, a piece of blotting paper or blotting-pad” (OED). Even though correlative noun formation is no longer the principal form of derivation for -ee words in the seventeenth century, the two suffixes are, by historical development, associated with one another and suggest a parallel path for some features of semantic change for both -er and -ee formation. In observing language change, we often have to consider more than one cause or factor. As Aitchison (1981: 169) notes […] anyone who attempts to study the causes of language change must be aware of the multiplicity of factors involved. It is essential to realize that language is both a social and a mental phenomenon in which sociolinguistic and psycholinguistic factors are likely to be inextricably entwined.

75

76

Heterogeneity in Word-Formation Patterns

3.5

Recession and continuations in the eighteenth century

For the first time since the beginning of productive use of the -ee suffix in English, we observe a decrease in new -ee formations in the eighteenth century, an occurrence which will be dramatically reversed in the century to follow. How is this eighteenth-century crisis in new -ee word creations to be interpreted? Why did the otherwise steadily rising productivity of this word-formation fall into a recessive stage? 1. Firstly, the decrease of new -ee word production reflects a general decrease of “lexical activity” during this period. As statistical evidence from the CED and other sources suggests (cf. Nevalainen 1999: 336), the time from the early sixteenth to mid-seventeenth century presents “the fastest vocabulary growth in the history of English in proportion to the vocabulary of the time,” with an absolute peak between 1570 and 1630. As can be seen in the figure taken from Nevalainen (1999: 339), the eighteenth century shows a dramatic slow-down with regard to vocabulary growth.

Figure 4.  Diachronic increase in number of words (Nevalainen 1999: 339)



Chapter 3.  The career of -ee words

Table 7.  Word list – eighteenth century (OED entries) 1706: 1710: 1715: 1716: [1716: 1721: 1726: 1727–1751: 1730: 1733: 1734: 1736: 1744: 1746: 1754: 1754: 1757: 1758: 1759: 1760–1772: 1761:

# † # # #

# # # @

warrantee (Law, person) visitee commissionee donatee settee] indictee (E version) assessee appointee (Mil.) underlessee promisee discontentee transferee releasee (Law) disponee (Law) lovee endorsee rubbee payee jestee (rare) dedicatee resignee (2)

1764: 1766: 1766: 1766: 1766: 1768: 1774: 1778: 1779: 1779: 1786: 1787: 1788: 1789: [1790: 1792: 1794: 1797: 1798:

# # † # #

#

petitionee (U.S. law) remittee drawee pledgee relessee appointee (Gen.) mandatee protegé complainee conferee (orig. U.S.) guarantee (2)26 snubbee examinee consignee coachee 1– dim.] emigré (appl. to Fr. e.) illuminé sornee (nonce word) cuttee

(40 words, 12# = 30%)

Summary: Number of words Number of direct object formations # Number of nouns with agent meaning @ Number of non-person references

40 12  1 1–2

2. The vocabulary need in the traditional domain of -ee words, i.e., the law, may have been saturated and, while new opportunities to expand the semantic field had already been explored and were further experimented with (see Table 7), they were not yet fully exploited. With regard to syntactic and semantic change, all of the developments described for the seventeenth century can also be observed for the 40 new formations of -ee words found for the eighteenth century:

26. Here: the act of guaranteeing, not the person, as in guarantee (1, 1679).

77

78

Heterogeneity in Word-Formation Patterns

– Examples of generalizations of contexts of use: visitee, examinee, promisee27 – Examples of humorous adaptations of a “legal role relationship”: lovee, jestee, rubbee – Occurrence of non-passive words: resignee (2) – ‘sb. who resigns from a job; a resigner’ (as opposed to earlier meaning, 1611), conferee – ‘sb. who takes part in a conference’ – Use in a non-human sense: (settee), guarantee (2) – as opposed to guarantee (1, 1679) The inclusion of settee in this list (Table 7) is somewhat controversial, and, as such, the word appears in brackets. As Marchand (1969: 268) contends, “Settee 1760 [OED lists first entry as 1716] fits in nowhere” nor can it justifiably be included in the class of (unrelated) jocular or diminuitive -ee noun suffixation of the bootee (‘infant’s wool boot’), coatee (‘close-fitting coat’) or shirtee (AE ‘shirt front’) type. Another such notorious case is coachee – used here in a diminuitive meaning, which was applied in the course of the nineteenth century to ‘one who is coached’ (OED), and so give evidence that the two basically unrelated suffix meanings of -ee might sometimes merge or playfully interact with one another. In sum, the eighteenth century sees a continuation of the developments of -ee word-formation which had started in the seventeenth century: more generalizations; inventions of humorous -ee words; loss of passive character in some words; application to non-human reference; some continued borrowings from French, some of which function as English nouns such as protegé, emigré, illuminé (cited above) but also some borderline cases such as the attributive adjectives equipee (173128) or passé (1775, cf. also below). Two observations with regard to the eighteenth-century formations may nevertheless be worth noting: a. Firstly, there is an increasing tendency to form direct passive nouns with -ee, especially in non-legal contexts (visitee, lovee, rubbee, jestee, cuttee, etc.). Among the 40 new words listed for the eighteenth century here, 12 are such direct object nouns, i.e., 30% (as opposed to roughly 19% for the seventeenth century) of these nouns. This development resembles the very beginning of the word-formation in the fourteenth and fifteenth centuries (cf. Section 3.2)

27. Even though promisee is also used in legal contexts, it can be applied in a wider sense: “1733 SWIFT Advice to Freemen Dublin Wks. 1745 VIII. 239 The persons … possessed of the sole executive power …, and hundreds of expectants, hopers, and promissees” (OED). 28. OED: “1731 Bailey vol. II. Equippé signifies a knight equipped, i.e., armed at all points […] 1775, Ash, Equipee”.



Chapter 3.  The career of -ee words

where direct object nouns with -ee were often the result of borrowings from French. In the eighteenth-century formations like lovee, jestee or cuttee, however, French loanwords as a source can be excluded. b. Secondly, one can detect a few words which are originally, chiefly or exclusively used in the United States of America (petionee, conferee), which is a reflection of the growing political (1776) and linguistic independence of the U.S.A. in the eighteenth century. The latter is marked by Noah Webster’s (1789: 20) linguistic declaration of independence: As an independent nation our honour requires us to have a system of our own, in language as well as government. Great Britain, whose children we are, and whose language we speak, should no longer be our standard. For the taste of her writers is already corrupted, and her language is on the decline. But if it were not so, she is at too great a distance to be our model and to instruct us in the principles of our language.

The divergence became most visible in the spelling deviation but also in the need to specify particular lexical items as “Americanisms.” As Romaine (1998: 10) remarks in her overview of the linguistic changes from that period, “by the eighteenth century a single, unified standard for English had ceased to exist.” This increased separation between British English and American English will also be dealt with in relation to subsequent time periods (cf. Section 6.1). The growing influence of specific American coinages is not the only lexical enrichment in English vocabulary. Generally, due to colonization and increasing global contacts, many languages have influenced the English lexicon, with some of these lexical items ending on the long vowel /i:/ and being orthographically represented by a double ‘e.’ Thus, by the eighteenth century, there was a range of lexical items of different origin which had an -ee ending. With the exception of passé, which is used as an attribute to a noun, all of them would be used in English as a noun, even though their origins need not have been nominal: Some examples of borrowed lexis from diverse sources: lascaree (1712): n. Urdu, non-person ref. (‘hunting spear’) suttee (1786): n. Hindi, Urdu, person reference (‘Hindu widow’) toupee (1727): n. French, person & non-person reference (a. ‘artificial hair,’ b. ‘person who wears artificial hair’)

79

80 Heterogeneity in Word-Formation Patterns

erminee (1736): a. French, non-person reference (‘special kind of cross’) passé (1775): a. French, person-reference, often attributive, e.g. a passé woman (‘past her prime’) sangaree (1736): a. Spanish, non-person reference (‘diluted and spiced wine’) chimpanzee (1738): n. Bantu lg., non-person reference, animate (‘a genus of African apes’) galapee (1756): n. Arawak, non-person reference (‘a West Indian tree’) corroboree (1793): n. Austral. aboriginal lg., non-person reference (‘type of native Australian dance’)

While it can be assumed that the average eighteenth-century speaker of English was able to distinguish a borrowing like chimpanzee from a native word-formation like releasee, some of the examples cited above are borderline cases, e.g. toupee, which can be extended to include person-reference metonymically,29 or passé, used attributively.30 The latter could have followed the same development path as, for instance ennuyé, which, after its use as an attribute, developed into a “quasi-noun” (OED) in the nineteenth century; thus an ennuyee is ‘one who is troubled with ennui’ (1827). Other borrowings would be used more context-specifically and there is no indication that their meanings would be actively confused or mixed with that of -ee word-formation. However, it seems plausible that the occurrence of this form in very different contexts and meanings (e.g. non-human) had an underlying effect on the diversification of -ee suffixation patterns in English, despite there being no ultimate proof for this hypothesis.

29. OED, 2nd meaning (obsolete): “†b. One who wears a toupee; a person of fashion; a beau, a spark, a buck. Obs.” First entry: 1727 Pope, etc. Art of Sinking x. 94 Then oh! she cries, what slaves I round me see? Here, a bright Redcoat, there a smart Toupee. 30. OED A. adj. “Past, past the prime; esp. of a woman: past the period of greatest beauty; also, out of date, behind the times, superseded.”



3.6

Chapter 3.  The career of -ee words

Abundance in the nineteenth century: -ee words go popular

The nineteenth century sees a wealth of new -ee word entries; four times as many are documented in the OED than for the eighteenth century.31 Again, the newfound linguistic identity in the U.S.A. resulted in vocabulary growth from that side of the Atlantic. As the editor of the Dictionary of American English observed, the flow of new words and meanings across the Atlantic changed after 1820, before which, the transfer had been westward (i.e., from Britain to the U.S.A.), and after which, the reverse occurred (cf. Fisher 2001: 69–70). One of the main sources of this development is the dramatic increase of new words that have a playful, humorous or ironic character – ironic in the sense that they mockingly transfer the legalistic character of -ee from its earlier uses to the sphere of interpersonal relationships. Thus, the pragmatic effect originates from the violation of an erstwhile restriction of -ee formations to the legal domain. Formations like staree, shavee, kissee, laughee, cursee, gazee, pluckee, kickee, hecklee or pumpee all belong to this “humorous” category. How can this abundance of humorous -ee formations be explained? Certainly, there is no social need for creating words like shavee, kissee or laughee. Fashion or trend is cited as one possible sociolinguistic factor for language change (Aitchison 1981: 124) and might play a role in the creation of such a wealth of -ee words. Also, actual borrowings from French decreased in the nineteenth century, making “mock-imitations” of such borrowings a plausible reason. And thirdly, neologisms and slang terms are often created for emphasis and vividness, “as a response to a kind of need when older words have become over-used and lose their impact, new vivid ones are introduced in their place” (Aitchison 1981: 124). Many of these playful words are classified as nonce words, i.e., words which were used just for a particular occasion, often as rather artificial and literary concoctions. They might be seen as prime examples of what is usually classified as morphological creativity (as opposed to “morphological productivity”), i.e., “the native speaker’s ability to extend the language system in a motivated, but unpredictable (non-rule-governed) way” (Bauer 1983: 63). However, unlike a hapax legomenon in a finite corpus or set of material, it is difficult to ascertain whether a nonce word remains a nonce word in the infinite set of naturally occurring language material. Many nowadays established and listed words famously started their career as a nonce word, and the definitions of when a nonce formation loses

31. This list, just as the ones for the preceding centuries, may not be exhaustive. The number of cited entries (152 for the nineteenth century), however, exceeds those of other estimated figures (cf. Barker 1998: 701 cites 100, based on Bengtsson 1927).

81

82

Heterogeneity in Word-Formation Patterns

its status are not clear-cut (cf. Bauer 1983: 45).32 The description of a word as a nonce word in dictionary sources is therefore not a reliable statement since, notoriously, dictionaries are outdated as soon as they are printed (the same goes for corpus material once the data collection is completed). As Fischer (1998: 3) points out in her definition and discussion of neologisms, “it is true that the date of first occurrence is an indication for the emergence of a new word, but it is not necessarily indicative of its spread and institutionalization.” When does the career from nonce word over neologism to established word start? Nonce-formations are spontaneously coined and they are rarely used. They can develop into neologisms, which in turn may lead to changes in the language system. For instance, developments on the level of the language norm can give rise to word elements (e.g. affixes or combining forms) which become productive and lead to new word-formation patterns in the language system. Changes on the level of parole, through the level of norm, may result in changes at the system level. The neologism develops from a diachronic process; the common vocabu(Fischer 1998: 5–6) lary is a synchronic abstraction. 

Bauer (1983: 45–47) notes that the status of a formation as a nonce word does not necessarily have to be linked to frequency of occurrence; if different speakers on different occasions happen to coin a complex word on the spur of the moment, the formation remains a nonce word. However, this requires rather special situations and usually the fact that a coinage is used over and over again goes hand in hand with a certain awareness of the speakers of using a term they have heard already – in Bauer’s (1983: 45) definition, the decisive factor which causes a change of status in erstwhile nonce words. A quick look at language usage on the Web33 confirms that, for instance, the once nonce words educatee or cursee are now wellestablished in contemporary usage. There are more than 400 entries for educatee (including dictionary entries) and approximately 180 entries for cursee,34 mostly in the sense of “one who has been cursed,” e.g.: He suspects that undoing his curse’s handiwork will be considerably harder than casting the curse was in the first place. (This probably explains why members of the family very rarely officially cast curses on their enemies – the long-term consequences may be as annoying for the curser as for the cursee).  (Review of Guns of Avalon, May 25, 2001, in www.epinions. com/content)

32. Cf. discussion of neologisms and productivity in Chapter 4. 33. Accessed June 2005. 34. Accessed June 2005.



Chapter 3.  The career of -ee words

Other nonce words, such as cudgellee, seem to be more restricted to the particular occasion in which they were recorded. All in all, there are 152 entries in the OED for the nineteenth century (Table 8). Table 8.  Word list – nineteenth century (OED entries) 1800: 1802: 1803: 1806: 1806: 1806: 1808: 1810: 1810: 1811: 1812: 1812: 1813: 1813: 1814: 1815: 1816: 1816: 1818: 1819: 1819: 1823: 1823: 1824: 1825: 1826: 1826: 1826: 1827: 1827: 1827: 1829: 1829: 1829: 1829: 1830: 1830: 1830:

# # # # # @ # # # # # # # @ # # # # # # # # # # #

staree expiree (Australian) invitee cudgellee (nonce wd.) preachee sendee engagee noddee (2) addressee hiree confidee custodee quotee divorcee gougee (nonce wd.) educatee (nonce wd.) interrogatee locatee trottee (nonce wd.) gaggee rappee banteree rescussee advisee quizzee (Gen.) disposee shavee (Joc.) ennuyée (quasi-noun) provokee kissee dislocatee cursee (nonce wd.) operatee laughee (nonce wd.) mesmerizee describee luggee (nonce wd.) remuneratee

1856: 1856: 1856: 1856: 1858: 1858: 1859: 1860: 1860: 1861: 1861: 1861: 1862: 1862: 1864: 1866: 1866: 1868: 1869: 1869: 1870: 1871: 1872: 1875: 1875: 1875: 1875–1876: 1876: 1879: 1879: 1879: 1880: 1880: 1880: 1881: 1882: 1882: 1882:

35. Here: for references of character of a person.

# # ? @ # # # # # # # # # # # # # # @ # # # # #

pillagee suspendee libellee standee (U.S.) bribee importee photographee beatee reservee advertisee civilizee objectee flirtee referee (5)35 defendee coachee (2) draftee (U.S.) licensee nominee (Econ.) jokee contestee (U.S.) baptizee callee contractee delegatee distrainee escapee complimentee evictee distributee (Law) vestee (Law) boycottee mancipee supervisee destinee sublessee drivee mesmeree

83

84

Heterogeneity in Word-Formation Patterns

Table 8.  (continued) 1831: 1831: 1831: 1831: 1832: 1832: 1832: 1832: 1834: 1836: 1836: 1837: 1838: 1838: 1839: 1840: 1840: 1840: 1840: 1841: 1841: 1841: 1841: 1843: 1846: 1846: 1846: 1846: 1847: 1847: 1848: 1850: 1853: 1853: 1853: 1853: 1855: 1855:

# # # #† # † # # # # # # # # # # # # # #(†) # # # # # # (@) # #

hangee introducee pluckee (nonce wd.) predecree kickee guarantee (sth. given as) loanee plottee pumpee (nonce wd.) abusee floggee shootee amusee questionee catchee cheekee (nonce wd.) hoaxee referee (4: sports) toastee boree treatee biographee trainee (first: animals) assignee (3: convict) permittee allottee convenee murderee arrestee managee (nonce wd.) abandonee employee (first: -é) gazee (nonce wd.) insuree fiancé guarantee (3)37 consultee geggee (Scots)

1882: 1883: 1883: 1883: 1883: 1883: 1884: 1884: 1884: 1884: 1884: 1884: 1884: 1885: 1885: 1886: 1886: 1886: 1887: 1888: 1889: 1889: 1890: 1890: 1890: 1891: 1892: 1892: 1892: 1894: 1894: 1894: 1894: 1895: 1895: 1895: 1897: 1898:

# # # #

# # # #† # # # # @ # # # # # # # # # # #

persecutee consecratee crammee (nonce wd.) pilotee tenderee transportee chargee co-assignee co-mortgagee galee revisee interviewee referee (6: publications) borrowee confirmee (eccl./rel.) chasee chattee vivisectee (animals) solicitee expellee sweatee vaccinee experimentee revoltee/revolté twistee usee (U.S., particular) adoptee presentee (jocular)36 transferee (person) dischargee jiltee reconcilee subornee hecklee deportee (esp. Indian) pardonee tippee (money) signallee, signalee

(152 words, 94# = 61.8%)

36. This was coined in opposition to absentee. 37. Here: “A person to whom a guaranty is given: the correlative of guarantor” (OED).



Chapter 3.  The career of -ee words

Summary: Number of words Number of direct object formations # Number of nouns with non-passive meaning @ Number of non-person references †

152   94    6    4

In a follow-up study to his earlier work on the -ee suffix, Bauer (1994: 41) not only analyzes twentieth-century -ee words from his own collection (cf. Chapter 4), but also conducts an analysis of nineteenth-century -ee words from the CD-Rom version of OED1. He found 100 -ee words with a first citation during the nineteenth century (but does not give any details or a list), all of which denoted human beings. An analysis of the syntactic patterns in his findings shows 54% direct object formations (somewhat less than the 61.8% in my analysis), 28% objects of a preposition formations, 2% subject formations and 16% listed as ‘none of these,’ i.e., “words such as biographee, where there is no corresponding verb, and loanee, which, in terms of current standard British English, looks as though it must be based on a noun, not a verb, although it might have been regularly formed from a verb in the nineteenth century” (Bauer 1994: 41). Along with some of the humorous new entries, many non-humorous -ee words which are firmly established in current, everyday language entered the dictionary in the nineteenth century, among them addressee, advisee, interviewee, trainee (first used for animals!) and employee. The use of this word-formation pattern in civil sectors of society saw an increase of -ee words which denote personal role relationships, even if they maintain a (serious or jocular, depending on the context) legalistic note, e.g. in confidee, introducee, consultee, reconcilee, pardonee. In this list (Table 8), we also have some examples of a development of particular -ee words in distinct varieties of English other than British or American English. The first example is from Australian English, the second one from Scots English: (1) expiree: “one whose term of punishment has expired; an ex-convict. Also attrib.” (OED) is specific to Australian English usage, as the examples also illustrate: 1802 Bentham Wks. (1843) XI. 123 “As to returns to England, the idea of preventing them on the part of the expirees … is now disclaimed.” 1838 Tait’s Mag. V. 781 “The free emigrants of South Australia, and the emancipists and expirees of Van Dieman’s Land.”

85

86 Heterogeneity in Word-Formation Patterns

(2) geggee: based on the Scots English noun gegg: ‘hoax, trick, practical joke’ and the verb to gegg: ‘to hoax, play a trick on’ (1855 J. Strang Glasgow: “The whole party including the geggee were in the highest spirits.”).

Given the development of other specific norms (including the lexicon) in other varieties of English (e.g. Irish English) in the nineteenth century, it can be speculated that these two examples are not exhaustive.38 In addition to those lexical items which are exclusively used in a particular variety, there are also cases where an already existing -ee word differs in meaning in various varieties of English (cf. Chapter 6). For contemporary usage, a more detailed analysis of the distribution of a sample of -ee words in various varieties of English will follow in Chapter 6. One of the most remarkable developments in the nineteenth century is the continued dramatic increase in formations with direct objects (marked with # in the list).39 Note that more than half of the new words (94 out of 152) listed here are of this type, as opposed to just 10% in the sixteenth century. Overall, a comparison of the percentage of direct object nouns in new -ee words between the sixteenth and the nineteenth centuries (Figure 5) shows a steep rise. This comparison is based on my catalogue of -ee words found in the OED (Table 4–Table 8). New -ee words with agent meaning are still clearly a minority, with custodee, rappee, escapee or standee as the exceptions, as well as the ambiguous insuree (both agent and patient meaning). It is worth noting that many of these subject nouns are formed from intransitive verbs or the intransitive meanings of the verbs, such as ‘to escape,’ ‘to stand,’ ‘to rap’40 or ‘to resign.’ Bauer notes on escapee (1983: 290), that “it is probably misleading to list the word […] here at all, since it is not an English formation but a nineteenth century loan from French. Its etymological source explain why -er was not used”, strikes me as not entirely convincing. The 38. Despite the fact that James Murray, the principal editor of the N.E.D. (i.e., the first edition of the OED) was himself a Scot and aware of regional differences in vocabulary, a policy of including regional English was not really articulated for the Dictionary and the OED’s coverage of regionalisms “arose in an ‘ad hoc’ way (Price 2003: 119). 39. Note that some of the direct object patterns in the eighteenth/nineteenth century are different from contemporary usage: ‘to jest,’ for example, was used in the eighteenth century as a transitive verb with direct object, ‘to jest someone’ (compare Bauer 1983: 246, who lists jestee as a different type). Ex. (OED) 1721 Ramsay Content 248 Be not aghast, Come briskly on, you’ll jest them when they’re past … 1775 Adair Amer. Ind. 427, I jested them in commending the swiftness of their horses. 40. ‘To rap’ can, of course, also have a transitive meaning but the version given for rappee suggests a derivation from the intransitive meaning – “One who raps or knocks; a rapper.” (OED).



Chapter 3.  The career of -ee words

70% 60% 50% 40% DO

30% 20% 10% 0%

16th C

17th C

18th C

19th C

Figure 5.  Direct object formations in new -ee words between the sixteenth and the nineteenth century (in %)

verb ‘to escape’ – a loan from Old French eschaper or even Vulgar Latin excappare – has been attested in English at least from the fourteenth century onwards and could also have been used as a source for escapee. The simultaneous presence of other agent nouns formed with -ee cannot explain this type of formation by borrowing alone. The continued influencial role of French in the eighteenth and nineteenth centuries – still the major international language of law, society and fashion – is an interesting one to consider, if not for real borrowings (see also words which retained a French spelling, e.g. protegé 1778, emigré 1792, illuminé 1794, employé 1850, fiancé 1853, revolté 1890), then for imitations or mock imitations of the French word-formation. Other possible influences – for instance, words with similar forms pointed out in the discussion of eighteenth-century formations (e.g. Anglo-Indian words, diminuitives, hypocoristic) – continue to play a role in the nineteenth century.

Table 9.  Summary of the development of direct object nouns in -ee formation Century

16th C

17th C

18th C

19th C

Total number new -ee words Number of direct object nouns Percentage of direct object nouns

21  2   9.5

47  9 19.1

40 12 30

152   94   61.8

87

88

Heterogeneity in Word-Formation Patterns

Formations of non-human reference nouns are still quite rare in this period. However, the few examples show that in the nineteenth century, animals could indeed be the referent of -ee nouns and, indeed, most of the formations of this type (marked in the list with †) refer to animals: trainee (later expanded to include human reference41) and vivisectee.

3.7

Changes across the centuries: Simultaneity and ambiguity

As we have seen, many shifts of meaning can take place over an extended period of time. Statistical evidence suggests that older words usually have larger semantic ranges than newly adopted words. Finkenstaedt & Wolff (1973: 108–110), for instance, cite figures of semantic change of lexical items which suggest that only about forty percent of the lexemes which date from the fifteenth century have only one sense, while as much as ninety-eight percent42 of twentieth-century origin words are monosemous. Simultaneity of various meanings is therefore more likely for those -ee words which have a relatively long history of usage. Yet, as we shall see in Chapter 4, processes such as figurative use or extension of meaning to various domains also affect more recent -ee words, making an ambiguity of meaning also possible for twentieth-century formations. In diachronic studies, Rissanen et al. (1997: 1) point out that to look at language variation (and here, the analysis of various word-formation patterns) provides us with a good opportunity to observe the actual process of change. We can trace the birth and death of variant expressions, but perhaps more interestingly, their changing frequencies and distributions within a variant field at subsequent periods of time and in various genres, and we can analyse changes in the intricate mesh of linguistic and extralinguistic factors conditioning the occurrence of these variants.

As Aitchison (1981: 48) remarks, factors such as language variation and “language fuzziness” are important for understanding language change and have often been ignored in synchronic language descriptions. “These frayed edges,” she writes (1981: 53), must be examined, not snipped away and tidied up. […] these are the

41. The first unmistakably human reference is cited in 1885 Daily News 16 Dec. 6/1 “Let her … ask whether she could be admitted as a lady pupil, as a trainee” (OED). 42. Given the fact that the publication of Finkenstaedt & Wolff (1973) already dates back a third of a century, this figure is likely to have changed. The general statement, however, that the probability of a polysemy increases with time, is still a valid point.



Chapter 3.  The career of -ee words

‘morbid symptoms’43 which occur when ‘the old is dying and the new cannot be born.’” The excursion into the history of the development of -ee words in English illustrates that (a) there are quite a few syntactic and semantic possibilities of -ee word-formation and (b) the syntactic and semantic preferences within this type of derivation have not been maintained across the centuries of its use. Rather, there have been many changes in the formation of -ee words between its first occurrence in the fourteenth century and the last part of our examination above. The parameters of change can be described as follows: Word class of base of derivation (correlative noun, verb); Type of passive object noun (indirect, direct); Passive character (object, subject); Semantic field of occurrence (legal, general, ironic/jocular); Human reference (human, non-human, including animals and technical components); f. Specific use in particular varieties of English (U.S.A., Australia, Scotland).

a. b. c. d. e.

However, while one can detect several preferences and trends at particular points in time, the syntactic and semantic features on which the formation is based do not completely supplant one another. At any one time, there is more than just one syntactic type of formation, i.e., noun and verb-derived forms, indirect and direct passive nouns, etc. The expansion of the semantic domain in which -ee words are used has not replaced the productivity of this formation in the legal context. The semantic expansion to include non-human reference has not resulted in an abandonment of new human reference nouns. Along with genuine English word-formations, French loanwords of the same form supplement the stock of -ee word-formations, especially in the eighteenth and nineteenth centuries. This type of heterogeneity in -ee suffixation is not due to synchronic linguistic and extralinguistic factors or constraints44 but reflects the history of its development over the course of six centuries. As Fischer (1998: 6) points out, a “neologism develops from a diachronic process; the common vocabulary is a synchronic abstraction.” In her study on lexical change in present-day English, she notes that the historical development of word

43. Aitchison is referring to a famous quote by Antonio Gramsci from his Prison Notebooks: “The crisis consists precisely in the fact that the old is dying and the new cannot be born; in this interregnum a great variety of morbid symptoms appear.” 44. A variability of synchronic usage according to users (e.g. geographical group) or context (e.g. text type) will be examined in Chapters 5 and 6.

89

90 Heterogeneity in Word-Formation Patterns

coinage is often neglected, due to a narrow distinction between synchronic and diachronic viewpoints on language: If language change is reduced to the difference between states of a language at different points of time, then cumulative, gradual and procedural (i.e., the historical aspects of the phenomenon) are neglected. Moreover, the heterogeneity of a language at a certain time is ignored.

I argue here that it is the simultaneity of syntactic and semantic possibilities of -ee word-formation, developed over centuries, which has led to a certain ambiguity of new formations with multiple interpretations. This will be explored in more detail and with a broad empirical basis in Chapter 4 and Chapter 5, especially with regard to twentieth- and twenty-first century productivity in my corpus analysis. Chapter 4 will firstly provide a theoretical discussion on the role of morphology versus lexicon in word-formation. This is particularly relevant when we look at the sources of new word production. Concepts of morphological productivity and creativity will then be discussed, particularly with regard to ad-hoc creations, creative neologisms and patterns of establishment, before we take a detailed look at the productive patterns of attested twentieth-century -ee words.

chapter 4

Morphology and the lexicon On creativity and productivity of -ee words

One of the basic purposes of describing the rules of a particular word-formation, such as those for -ee nouns, is to predict the formation of new words in form, regularity, and sometimes also frequency. In a nutshell, the ultimate purpose is to forecast the future productivity of a word-formation pattern. The theoretical framework of word-formation emerged in the 1960s and 1970s as a field within linguistics, developing from its original ties to transformational syntax (e.g. Lees 1960) and an autonomous lexicalistic approach (e.g. Aronoff 1976; Chomsky 1970). Word-formation, as a rather complex area that deals with the interface of semantic, syntactic and morphological-phonological features, has since gained in importance. For the purpose of the present study, there are at least two issues in word-formation theory which need to be discussed at this point: (a) the relationship between morphology and the lexicon, i.e., abstract rules of a word-formation versus actual words, and (b) the notion of productivity in wordformation, especially the relationship between productivity and creativity. There are long-standing debates on both of these topics in word-formation theory, a brief outline of which will be given in Sections 4.1 and 4.2. These theoretical discussions will be followed by a detailed look at productive patterns of -ee formations in the twentieth century. This part will also provide an inventory of those twentieth-century formations that were found in established sources (either listed in the OED, other twentieth-century dictionaries or found in scholarly articles). In a comparison with -ee formations in previous centuries (discussed in Chapter 3), I will concentrate here on a number of phenomena in these more recent formations which are typical of the twentieth century. The notions of creativity versus productivity will be reexamined in the subsequent treatment of ad-hoc creations, neologisms and routes of establishment in the formation of a new lexeme. Finally, I will come back to the question of morphology and the lexicon when discussing the concept of “actual words” versus “possible words” (Aronoff 1976; Aronoff & Anshen 1998; Bauer 2001) – one of the very foundations on which my empirical study with “possible words” is based and which is analyzed in detail in Chapter 5.

92

Heterogeneity in Word-Formation Patterns

4.1

Word-formation between morphology and the lexicon

The relationship between word-formation and other areas of linguistic investigation is rather complex. In the wake of the development of generative grammar theory, word-formation was at first merely included in this theoretical model which treated word-formations as transformationally shortened sentences (cf. Hohenhaus 1996: 144ff; Lees 1960). But do speakers really generate words each time anew in the same way they generate sentences? A purely generative approach could therefore not be upheld for long and was soon abandoned (cf. also Motsch 1977, 1981). Next to morphological rules, the notion of the mental lexicon with its stock of already existing words (actual words) has come to play an important role in the formation of new words (possible words). Along these lines, Chomsky (1970) also pleaded for a study of word-formation separate from syntax, mainly because he saw the many irregularities in word-formation as a source of distraction for the regularity of a model of syntax. This restriction notwithstanding, his lexicalistic hypothesis paved the way for further work on the relationship between morphology and the lexicon. But how are actual words stored in the mental lexicon? Jackendoff (1975) proposed a distinction between a “full entry” model for simplex words (e.g. relate) versus an “impoverished entry” model for a derived (complex) word (e.g. relation). By referring to the simplex word, the complex word does not get a fully specified entry in the lexicon. Jackendoff (1975: 668) thus argues against a simple view of memorization and defends the concept of creativity in word-formation: We have thus abandoned the standard view that the lexicon is memorized and only the syntax is creative. In its place we have a somewhat more flexible theory of linguistic creativity. Both creativity and memorization take place in both the syntactic and the lexical component. When the rules of either component are used creatively, no new lexical entries need be learned. When memorization of new lexical entries is taking place, the rules of either component can serve as an aid to learning. However, the normal mode for syntactic rules is creative, and the (Jackendoff ibid.) normal mode for lexical rules is passive. 

Following the influential work of Aronoff (1976) on morphology and the lexicon, Aronoff & Anshen (1998: 237) see the relationship between regular morphology . Cf. also Morris Halle’s (1973) model of word-formation which includes the components “list of morphemes,” “rules of word-formation,” “filter,” “dictionary of words,” “syntax” and “phonology.” . Here meant in analogy to creativity in syntax (note the difference to the discussion of the term “creativity” as opposed to “productivity” below).



Chapter 4.  Morphology and the lexicon

(potential words) and irregular lexicon (existing words) as separate but interdependent, when they state that, basically, “[…] morphology, which forms words from words, finds the words that it operates on (its bases) in the lexicon.” This means that the speaker abstracts rules from the existing lexicon and forms new words accordingly. Not only does the speaker exploit abstract morphological rules, but also concrete patterns found in actual words in his or her new words. According to Aronoff & Anshen (1998), in the mental lexicon of the speaker/ hearer, not all words will be stored alike; if an item is irregular or unpredictable, it is more likely to be stored than a regular item. We might say that morphologically well-formed complex potential words are provided by the morphology, not the lexicon of a language. Yet, the lexicon, including the more irregular and unpredictable words – e.g. idiosyncracies as the result of language change or language contact – has an influence on the formation of new words. As Aronoff (1976: 18) points out, “words, once formed, persist and change” and, therefore, actual words are never just a “subset of the items which are generated by a regular morphology.” Morphology thus may work on the lexicon. One of the indicators of this, Aronoff & Anshen (1998: 242) note, is the inheritance of irregularity, for instance, semantic irregularity (such as in the unpredictable meanings of immeasurable or naturalize). In sum, we might describe the relationship between morphology and the lexicon as one of “rival sources,” which are, nevertheless, interdependent: “The morphology depends on the lexicon, however, inasmuch as the bases of morphologically complex words are normally lexical entries” (Aronoff & Anshen 1998: 242). To illustrate the problem of morphology versus lexicon, word-formation rules versus actual words, Plag (2003: 45) uses the -ee suffix as a primary example to illustrate that word-formation rules are often “formulated in such a way that they prohibit formations that are nevertheless attested.” Citing the different syntactic patterns of -ee formations (transitive/intransitive verb-derivations, noun-derivations, etc.) he notes that, “ideally, one would find an explanation for these apparently strange conditions on the productivity of these affixes” (ibid.).

4.2

Productivity and creativity in word-formation

The way the relationship between morphology and the lexicon is viewed also has further consequences on another debated notion in word-formation, productivity and, as a related but separate concept, creativity in word-formation. Kastovsky . Cf. also Di Sciullo & Williams (1987: 3) who famously stated that the lexicon can be considered to be “like a prison – it contains only the lawless.”

93

94 Heterogeneity in Word-Formation Patterns

(1986a), for example, makes the distinction between the scope of a word-formation rule and its actual utilization in performance and notes that, in discussions of “productivity,” these two aspects should be kept systematically apart. Bauer (2001: 34) points out the problem of documenting “actual” or “existing” words as an important condition for the concept of productivity: “If productivity is concerned with the potentiality of new formations, then it must be possible to discover whether or not something is new, and this implies that it can be compared with a list of formations which are not new but “established” […]” (Bauer 2001: 34). Productivity, originally a term from the field of economics, is used in linguistics “to refer to the creative capacity of language users to produce and understand an indefinitely large number of sentences” (Crystal 1991: 279). It was first and foremost applied to the area of syntax, as part of the generative grammar paradigm, as a “set of formal rules which generate the sentences of a language and assign to each a set of appropriate structural descriptions” (Matthews 1974: 217). As we have seen above in my discussion on the limitations of generative grammar in application to word-formation, we can talk about “actual” and “possible” words but it would be rather absurd to talk about “actual” versus “possible” sentences – sentences, after all, are usually generated from a finite set of rules and not drawn from memorization. Thus, a definition of morphological productivity is somewhat more restricted: Morphological productivity may be defined informally as the extent to which a particular affix is likely to be used in the production of new words in the language. On this view, productivity is a probabilistic continuum that predicts the (Aronoff & Anshen 1998: 242–243) use of potential words. 

Plag (1999: 6) summarizes some of the restrictions which arise when we try to describe the word-formation rules for a potential new complex word: a. “[W]ord-formation rules may predict the existence of forms which are unattested or whose status as well-formed derivatives is more than doubtful”: for instance, a common word-formation rule in English predicts the attachment of the nominalizing suffix -ity to adjectival bases ending in -ous (e.g. curious – curiosity) but not all adjectives of this type form derivatives with this suffix, e.g., there is no actual formation furious + -ity → *furiosity – even though this could be seen as a “possible word.” b. They (word-formation rules) are “often formulated in such a way that they prohibit formations that are nevertheless attested.” c. “[S]ome affixes occur with a large number of words, whereas others are only attested with a small number of derivatives” (Plag 1999: 7).



Chapter 4.  Morphology and the lexicon

The problem outlined in (b) occurs to a great extent in -ee word-formation rules (cf. critical overview of previous phonological, syntactic and semantic models in Chapter 2), and indeed Plag cites the existence of various syntactic patterns of -ee formation as an example. Whereas (a) and (b) are concerned with qualitative questions, the problem described in (c) deals with quantitative aspects, i.e., the frequency of occurrence of a particular word-formation. Frequency and productivity in word-formation are often regarded as nearsynonyms. As Pounder (2000: 133–134) notes, the very definition of productivity is sometimes even equated or confused with frequency, a view which she contests in that “[…] the former concept [productivity] refers to the dynamic aspect of word-formation, namely to operations, and the latter [frequency] to its static aspect, namely to attested forms in the lexicon, so that such a usage is misleading.” Plag (1999: 22–23) also argues against a simplistic view of productivity in purely quantitative terms because such a perspective would neglect an essential idea of productivity, that of “a design feature of language” (cf. also Bauer 1994). In other words, the fact that there is a large number of words with a particular affix does not mean that contemporary speakers use the affix actively to produce new words. As Bauer (2001: 48–49) maintains, “Type frequency [= the number of items in the language that contain the process in question, S.M.] is the result of past productivity rather than an indication of present productivity.” Finally, both past and present productivity of, for instance, an affix is dependent on the size of the bases to which it can attach. There are a number of relevant aspects to be considered with regard to the productivity of a word-formation process: synchronic versus diachronic views of productivity as well as qualitative versus quantitative aspects. Before I discuss these in more detail in Section 4.3, I will first turn to the concept of creativity in word-formation, since many scholars draw a distinction between the notions of productivity and creativity. Bauer (1983: 63) sees creativity in morphology as “the native speaker’s ability to extend the language system in a motivated but unpredictable (non-rulegoverned) way.” But there he discusses only productivity in detail, although both phenomena give rise to large numbers of neologisms. “This is because it is impossible to make any worth-while generalizations about creativity because of its unpredictability, although it would no doubt be possible to provide a taxonomy of types of creativity” (Bauer 1983: 64–64). In his study on Morphological Productivity (2001), however, Bauer does give the notion of creativity more attention and distinguishes between three different types of creativity: creativity1 as the coining . Type frequency, as opposed to token frequency, which is concerned with the number of occurrences of a word in a given text.

95

96 Heterogeneity in Word-Formation Patterns

of simplexes, creativity2 as figurative extension of existing words and creativity3 as “the extension of non-productive patterns, patterns used only by isolated individuals, etc.” (Bauer 2001: 64). For our discussion of possible creative coinages of -ee words, only creativity3 is relevant but, again, common to all three types of Bauer’s creativity seem to be that they are not rule-governed. However, as he himself also notes, the notion of rule-governedness as a factor to distinguish between productivity and creativity is not a particularly useful or water-tight one. Instead, he proposes to distinguish the two concepts along the line of “predictability.” Neologisms, and especially creative neologisms that are attributed to such “non-rule governed coinage,” have frequently been considered marginal for morphological theory in which such a distinction between morphological productivity (as an unintentional coinage of a number of formations) and morphological creativity (as a conscious coinage of a new word on an unproductive pattern) is made. In Deconstructing Morphology, Rochelle Lieber (1992) comments on morphological creativity in rather negative terms: With unproductive processes a new word may sometimes be coined, but such coinages will always draw attention to themselves. They will be perceived by the native speaker as odd, amusing, repulsive, or otherwise remarkable. Such words are often coined by advertisers to draw attention to their products. They may show up in Safire’s New York Times column, or on the back page of The Atlantic or in the “Among the New Words” feature of American Speech. But they are not necessarily to be taken seriously by a theory of word-formation. (Lieber 1992: 3)

As we have seen in the outline of the syntactic and semantic patterns of -ee words as well as in our historical overview of this word-formation pattern, there are quite a few humorous nonce formations with -ee, especially in the nineteenth century (e.g. trottee, cursee, laughee), to which such a description – that they draw attention to themselves – could be applied. And as we shall see in the description of twentieth-century formations below, this development continues and, indeed, -ee formations have been the subject of Safire’s column and various articles in American Speech. But does this mean that they do not contribute anything worthy of note to the general word-formation pattern? If we take into account the development of -ee words as described in Chapter 3, we can note that the changes in the semantic and syntactic make-up of the -ee formation pattern were always initiated by just a . Some researchers also have labelled this as a distinction between plain morphology, i.e., “the ordinary productive (and non-productive) word-formation and word structure rules of a language” (Zwicky & Pullum 1987: 332) and expressive morphology which is “associated with an expressive, playful, poetic, or simply ostentatious effect of some kind” (ibid.: 335).



Chapter 4.  Morphology and the lexicon

few unusual coinages, upon which, one might argue, other formations followed. This would then be a case of analogous formation followed by rule-governed formation. That is to say, the unusual formation, the exception at a given point in time (for instance, direct object formation – indictee, electee – or subject formation – absentee – in the sixteenth century) sets an example from which a new or modified rule will be formed. Botha (1968: 135) has labeled such a process “rule-changing productivity,” as opposed to “rule-governed productivity.” Quite a number of scholars (e.g. Derwing 1990; Derwing & Skousen 1989; Kiparsky 1974; Skousen 1989) have stressed the importance of analogical coining in word-formation. In her analysis of prefixation in English, Schröder (2008: 232) uses a morphological network approach which, to her, “has much greater explanatory power than the presumption of any (set of) morphological rule(s), with the advantage that no one rule needs to be chosen over another as an explanation for a particular complex form” (ibid.). In the history of -ee formations (Chapter 3), the interplay between “rule-changing productivity” and “rule-governed productivity” seems to have played an enormous role in the development of polysemy and heterogeneity of this word-formation pattern. The rather nonchalant attitude towards creativity in word-formation, such as expressed above, is therefore somewhat unfortunate. As Lehrer (1996: 64) also notes, “creative neologisms can tell us a great deal about important aspects of word-formation. Moreover, they are extremely productive!” (emphasis in the original, S.M.). Kastovsky (1986a: 586) summarizes an important aspect of the productivity problem with regard to neologisms: […] neologisms are usually checked against the mental lexicon and may be rejected if they are not part of it, even if they are coined on the basis of a productive pattern and are morphosemantically completely regular. This is basically a performance phenomenon and has to do with the fact that the output of wordformation rules is lexical items and lexical items are linguistic units that typically form part of a stored inventory. Consequently, what is not part of it does not exist.

In the last decade there have been quite a few studies on creative neologisms, nonce formations and ad-hoc creations in connection with the debate on the value of morphological creativity (Elsen 2004; Fischer 1998; Hohenhaus 1996; Oesterreicher 1999) and its fruitful engagement with the issue of morphological . For an overview of the “rules and analogies” discussion, cf. Plag (1999: 17–22), also Bauer (2001: 75–97). For a reassessment of analogical coining, cf. also Schröder (2008). . Cf. also Eugenio Coseriu’s (1967) notions of “norm” versus “system” (cf. also Fischer 1998: 5; Hohenhaus 1996).

97

98 Heterogeneity in Word-Formation Patterns

productivity. With regard to -ee formations across the centuries, it is clear that both notions – creativity and productivity – have to be considered. As pointed out above, the two are ultimately as interdependent as the lexicon and morphology. Although Lehrer (1996) is sometimes not quite clear in what she includes in her usage of the term neologism, the reasons she provides for including creative formations based on metaphors, irony, word-play and jokes into morphological theory are valid and insightful: a. In other areas of linguistics study (e.g. semantics, pragmatics) creative metaphors etc. have become recognized as legitimate objects of study – so that morphological theory may also take non-literal language into account. b. The study of creative neologisms may shed new light on old misleading statements about productivity or non-productivity of particular processes (Lehrer cites misleading statements on verb-headed compounds, e.g. baby-sit, breastfeed, which were deemed not productive but empirical evidence showed that the opposite was the case). c. Studying creative neologisms may also contribute to the debate on “analogy versus rule,” i.e., “whether the creation of a new word (or new expression) is formed on the basis of a specific item or a general rule. If we encounter pizzaburger for the first time, is it formed by anaology with hamburger, or is there a general rule of the form” (Lehrer 1996: 67). She points out that neologisms very often originate by analogy, but quickly (after 3 or 4 words) generate a rule. Additional new words then do not need any specific words as a base but are formed by rule (cf. also rules and analogy, Plag 1999: 17–22), for instance, when -(a)holic – originally part of alcoholic – becomes a suffix denoting ‘addict of X’ and the speaker forms new lexemes without resorting to the original basis. d. The analysis of creative neologisms, according to Lehrer (1996: 68), can also help us understand the development of semantic change and polysemy: “If one looks at the range of meanings of words with multiple senses, or at the variety of words with a particular affix, one is almost compelled to become a lexicalist and argue that any possible rule has too many exceptions to be plausible. However, if we look at creative word-formation in context, we see that there are ‘fashions’ ”. A study of the processes of word-formation in a cultural-historical setting is therefore instructive and can help to shed light on processes of language change.



Chapter 4.  Morphology and the lexicon

e. The study of creative neologisms may provide more information on processes such as blocking. f. In contrast to the belief that neologisms are ephemeral and do not stay in the language, Lehrer (1996: 70) also shows that this is not true so that quite a few of the words listed by Algeo (1991) as neologisms (one third) were still valid. g. One of the objections against the study of creative neologisms is that they are often seen as “not serious.” This notion may also be attributed to a selective perception, in which the more humorous cases receive more attention. However, even if humor plays a larger role in creative neologisms than in merely productive ones, this can hardly be held against their serious study: after all, humor is a serious topic in linguistics (cf. also Hohenhaus 1996: 297ff.). h. Another objection is made on the grounds that not all speakers have the same competence in constructing neologisms and some speakers may have no productive control of them at all (e.g. Zwicky & Pullum 1987) – an assertion which may be true, yet, may also apply to other aspects of linguistic creativity. As noted above, quite a few -ee lexemes and especially -ee formations in the nineteenth and twentieth century can be judged as being the result of morphological creativity. The points outlined here, especially the claims made in (b), (c) and (d) may thus be seen as pertinent to a study of creative -ee neologisms. Yet, is -ee affixation also seen as a productive word-formation pattern? Whether this is the case has been much debated: Bauer (1983: 288), for instance, notes that the relatively high number of -ee forms listed in his study may be due to his personal awareness of the suffix rather than its inherent productivity. Ultimately, his assessment of the productivity of this suffix is also dependent on a comparison with other subject nominalization suffixes: When subject nominalizations are considered in the literature, the word-formation process that is usually discussed is -er suffixation. This is not, however, the only word-formation process involved in subject nominalizations. Others are ant suffixation (attendant), -ee suffixation (escapee, […]) and conversion (nurse, n, v). At least two of these are still productive, though neither of them is as pro(Bauer 1983: 286ff.) ductive, in this meaning as -er suffixation. 

He remarks that semantic field plays a role in the productivity of -ee and that for contemporary usage, “it is frequently playful or whimsical in tone” (Bauer 1983: 244).

. One might add that the same also applies to other constraints on productivity, including phonological, morphological, semantic and aesthetic factors.

99

100 Heterogeneity in Word-Formation Patterns

That the productivity patterns of -ee word-formation differ according to the various syntactic patterns is already pointed out in Bauer (1983: 245–247, 1994: 41, 46) (cf. also below, Section 4.4), a view which Barker (1998: 23–24) addresses in a critical comment: […] -ee noun formation in general is sufficiently productive to deserve a generative explanation. However, the individual arguments […] against a syntactic treatment go through only to the extent that the various subclasses of -ee noun types are sufficiently productive. That is, it might be possible for someone committed to a uniformly syntactic account of derivational morphology to argue that even if -ee noun formation in general is productive, some of the more exotic and problematic -ee noun types are not.

Barker estimates that at least three of the four syntactic patterns (direct object, indirect object and subject uses) are productive, i.e., he does not include the subject type in the “more exotic and problematic -ee noun types” as long as the appropriate semantic conditions (sentient, non-volitional, episodically linked to the verb) are met. Therefore, Barker claims, productivity of -ee derivations is dependent more on semantic rather than syntactic criteria, i.e., “the fact that the semantic approach automatically accommodates without adjustment the more exotic -ee noun types (governed preposition uses, uses for which there is no corresponding verbal argument, and uses with non-verbal stems) certainly counts in its favor” (Barker 1998: 24). In his corpus study on suffixation, Schmid (2005: 175) classifies -ee suffixation as productive. There is, however, no further classification into syntactic subclasses or discussion of semantic properties involved in producing new -ee words. Portero Muñoz (2003: 130) describes -ee as “semi-productive,” between productive suffixes (like -er and unproductive ones (like -id, -ment, -th), but without giving any further details about the nature of the term “semi-productive” or its application to the -ee suffix. Most of the assessments of the productivity of -ee given above remain rather vague or rest on the idea of frequency. In the following section, we will now look at various types of productivity (cf. Aronoff & Anshen 1998; Plag 1999; Pounder 2000) and their possible applications to -ee suffixation.

4.3

Types of productivity: Synchronic and diachronic processes

As could be seen above, the notion of productivity may mean rather different things for different researchers. Therefore, various aspects of the concept need to



Chapter 4.  Morphology and the lexicon 101

be discussed when we deal with the question of whether or not -ee suffixation is a productive word-formation pattern.

(i) Gradual productivity Arnonoff & Anshen (1998: 243) note that “some linguists treat morphological productivity as an absolute notion – a pattern is either productive or unproductive – but there is a good deal of evidence for the existence and utility of intermediate cases.” Absolute productivity should be rare or non-existent, especially in derivational morphology, and what might be labeled as “unproductive” at least must have been productive at one point in time. Most word-formation patterns should be somewhere in between. Portero Muñoz (2003: 130), having defined -ee formation as “semi-productive”, aptly points out that such suffixes “in the gray area in between” are, in fact, the most interesting ones for the researcher because for both alternatives, very productive and rather unproductive word-formations, there is very little theoretical interest in a discussion of productivity. (ii) Qualitative productivity Qualitative10 morphological factors, according to Aronoff & Anshen (1998: 243), can best be illustrated by examining affixes which are similar in their semantic and syntactic conditions. Therefore, one of the factors for dealing with the productivity of -ee is the question of rival suffixes. It is notable that there is no direct competitor to the meaning of -ee derivations. With regard to single or rival affixes, Bauer (1983: 286) remarks that “-able is the only affix with the meaning of the type ‘able to be Ved,’ and -ee is the only suffix with the meaning ‘person who is Ved.’” There are, however, a few exceptions to this; for example, the Latinate -and suffix (even if rare) has a similar meaning and use as -ee, that is ‘person who is Ved’ or ‘person to be Ved.’ Examples of -and words are analysand, confirmand, educand, graduand, multiplicand, operand, ordinand. The OED notes that “the meaning of these words is passive, thus ordinand ‘person to be ordained’ ” and does not ascribe any productivity to it: “this element has never been a living suffix, having no separate existence apart from the Latin gerundive form from which it is . From the way Portero Muñoz uses the term “semi-productivity,” it implies that the wordformation is neither fully productive nor fully unproductive. Other definitions of semi-productivity (e.g. Matthews 1974: 71) focus more on an element of unpredictability and idiosyncrasy, i.e., “idiosyncratic affixes which inexplicably fail to attach to apparently eligible forms. Furthermore, where such affixes are used, the meaning of the resulting word may be unpredictable.” 10. Note that Plag (1999: 13ff.) discusses a number of factors (e.g. unintentionality, rules and analogies) under the heading of “qualitative productivity.” These issues are dealt with here in the section on “morphological creativity.”

102 Heterogeneity in Word-Formation Patterns

derived. […]” (OED). Certainly, however, the twentieth-century formation analysand – (also as correlative opposite to agent noun analyst) must be seen as at least one instance of a creative application of the -and rule.11 An even more marginal contestant as a “passive noun” are the few cases (also cited in Section 2.8) where the -er suffix has a patient meaning (milker, cooker), none of which, however, applies to persons. If the -and suffix may be counted as a competitor to -ee formations, then the latter is clearly the winner or, as Aronoff & Anshen (1998: 244) call it, the “default, qualitatively unrestricted” option. This also has consequences for the quantity of the operations: “This qualitative difference is usually mirrored quantitatively as well: the qualitatively least restricted operation among a set of rivals will most often also be the quantitatively most productive […] It is also not always true that the least restricted member of a set of rivals will be totally unrestricted in its distribution” (ibid.), for instance, when a default (like -ness for de-adjectival nouns in English) does not attach to all words of the appropriate form. Qualitatively, -ee as an affix with hardly any serious contender is certainly the most productive and default option for the meaning ‘person who is Ved’ or, in a broader semantic description ‘sentient entity who non-volitionally participates in an event which is linked to the V.’ More accurately, the -ee suffix can itself sometimes act as a rival to another suffix, the -er suffix, as in the many examples of subject formations (parkee, standee, waitee, etc.). There are, of course, alternative syntactic patterns to express the idea of an -ee word (e.g. by means of past participle – employee – ‘X is employed’ or attributive adjectives ‘the employed X.’ As regards for noun derivations, -ee is certainly unchallenged in its qualitative productivity. There are a number of other issues which might be included in the question of qualitative productivity: Pounder (2000: 134–136), for instance, discusses both transparency and regularity in morphological productivity. In her definition, transparency “can be the simple recognition of the formal morphological units, i.e., stem and further elements, whose presence must be regarded as sufficient for the indication of the presence of word-formation meaning” (2000: 134). Baayen (1993: 199) notes that “semantic transparency, like phonological transparency, is a necessary but not a sufficient condition for productivity.” Regularity might be seen as equivalent to transparency. However, form-meaning relations can be 11. Analysand – OED: “The subject of, or patient in, psycho-analysis; one whose psycho-analysis is being attempted” (First entry: 1933 H. Crichton-Miller Psycho-anal. i. 105. ‘The transference.’ By this is denoted the relation of the analysand to the analyst. 1953 R. F. C. Hull tr. Jung’s Coll. Wks. VII. 136 Every analysand starts by unconsciously misusing his newly won knowledge in the interests of his abnormal, neurotic attitude).



Chapter 4.  Morphology and the lexicon 103

transparent without being regular, for instance, when a speaker forms a word by means of a foreign affix and a native stem – the new word may be transparent without being conventional and regular. In Pounder’s view (2000: 135), “regularity” merely implies that the formation is not an isolated case, i.e., that there are other formations of the same type. As we have seen in our discussion of syntactic and semantic features of -ee suffixation, the formation and its various sub-types (various syntactic types of deverbal derivation, noun-noun derivation12) are mostly regular – a possible exception being the odd de-adjectival noun derivation (e.g. absentee, deadee) – but not necessarily transparent. Rather, ambiguity of meaning, sometimes deliberate, is one of the characteristics of many -ee formations, making the lack of transparency the most severe restriction in qualitative productivity of this word-formation. The following example, a metalinguistic comment on the word dementee, once again illustrates some aspects of the ambiguity of meaning in -ee word-formation and the role of analogical coinage in the process of forming new words: dementee C 150: […] Probably there is an assumption that a “mentor” is one who ments, and a “mentee” is one who is mented. Would a “dementee” be one who is demented? www.halway.com/weblogs/english/mentee.html

The apparent puzzlement of the speaker is, on the one hand, based on the “nonunitary base,” i.e., the fact that -ee more regularly attaches to a verb but can also attach to other bases. The speaker’s first misconception is that there must be an underlying verb *to ment, which, however, does not exist – mentee is noun-derived, as a correlative opposite to the much older mentor. Dementee, on the other hand, is verb-derived, and there is no correlative *dementor to the dementee. The lack of transparency is then enhanced by the two possible meanings of the verb to dement, on the one hand, “to give the lie to; to assert or prove to be false,” and on the other hand, “to put out of one’s mind, drive mad, craze” (OED). In other words, dementee might refer either to ‘someone who is demented, gone crazy,’ etc. or to ‘someone who has been proven to be false’ (for instance, in a testimony).

(iii) Quantitative productivity As pointed out above, frequency of occurrence is often taken as the only or primary form of productivity in word-formation. Productivity is, by the nature of the term, related to growth. Usually, an analysis of quantitative productivity of a 12. Note that -ee formation is, in fact, a counter-example to the so-called unitary base hypothesis (UBH) in that the affixation operates on at least two, if not three distinct classes of bases. For an overview of the UBH, cf. Plag (1999: 47–48).

104 Heterogeneity in Word-Formation Patterns

morphological operation makes use of large computer-analyzable corpora. But how exactly such growth of, for instance, a particular derivation, can be measured involves a number of issues: Firstly, the growth rate of the items in question must be seen in relationship to the growth rate of the vocabulary of the language and, secondly, the ratio of hapax legomena formed by that operation must be seen in relation to the total number of items of that morphological type in the corpus. Therefore, the critical point in quantitative productivity analysis is the type of corpus used for the analysis,13 since the rate of occurrence of new words depends not only on the size of the corpus, but also on the types of texts (genre, spoken or written, etc.) collected here. One might get rather different results if different corpora or even dictionaries as a corpus are used (cf. Aronoff & Anshen 1998: 245 on differences in quantitative productivity of -ity and -ness in various sources). […] dictionaries are not always dependable indicators of actual usage, since the entries in a dictionary are selective rather than inclusive, and since hapaxes are less likely to be seen as meriting dictionary entries. Counts based on actual large corpora of the sort that Baayen employs are generally more reliable, since they measure actual use, rather than being filtered editorially.  (Aronoff & Anshen 1998: 245)

However, very recent formations may not be found in most of the established closed corpora either since they contain data collections from particular periods of time. Baayen (1993: 189) carefully distinguishes between hapaxes in the corpora, the number of which is used “to estimate the likelihood of encountering types that have not been registered previously,” and neologisms which, as he permits, will only be found in larger corpora: It is only for the larger corpora that neologisms will begin to appear, predominantly among the hapaxes. Even then, many hapaxes will be well-known lexical items. However, as the sample size increases, the proportion of the neologisms among the hapaxes will increase. Hence the probability of encountering neologisms is measured indirectly by means of the probability of encountering hapaxes. It is in this sense that P14 is a measure for the degree of productivity.

13. For a more thorough discussion of the types of computer corpora available, their uses, as well as on the question of alternatives to the established corpora for linguistic analysis – i.e., in my study: the World Wide Web – cf. Sections 5.1 and 5.2 14. Baayen’s formula for statistical measure of the degree of productivity: P = n1/N, i.e., P of a morphological process is the quotient of the number of hapax legomena n1 with a given affix and the total number of tokens N of all words with that affix.



Chapter 4.  Morphology and the lexicon 105

Aronoff & Anshen (1998: 244) note another statistical measure for quantitative productivity, i.e., what Baayen (1993) calls “global productivity.” This depends not only “on the likelihood of encountering new words of a given morphological type, but also on the number of words of that type that a speaker already knows.” In his discussion of quantitative productivity, Plag (1999: 22ff.) includes the dichotomy of actual versus possible words (cf. also Section 4.6), a distinction which also reflects on the empirical possibilities of quantifying productivity. After all, existing, lexicalized “actual words” cannot shed any light on the capacity of a word-formation to produce new lexemes. But how can one measure the number of forms which could potentially be derived with a particular affix? As a productivity index, Aronoff (1976) proposed the ratio of actual to possible words – a suggestion that Plag (1999: 23) refutes on the grounds that this ratio would be distorted both in measures of highly productive and of largely unproductive processes. For the historical data given in Chapter 3 and summarized below in Section (iv) on “diachronic productivity,” no such statistical measure will be given as the data is collected from dictionaries and cannot be compared to other corpus material. In my study of -ee neologisms on the web (cf. Chapter 5), quantitative productivity plays a role in that the neologisms found are classified according to their frequency on the web. However, as the data is not collected from a finite and coded corpus, Baayen’s statistical formula would not make any sense for this type of research.

(iv) Diachronic productivity As we have seen in Chapter 3, the number of new items found in a particular time period can differ greatly from that of another time period. The fact that this might also be dependent on the general development of the vocabulary has already been noted, especially for the development of new -ee words in the eighteenth century. In this period, the decrease of new -ee word production mirrors a general decline of “lexical activity” – in stark contrast to the early sixteenth to mid-seventeenth century where “the fastest vocabulary growth” in English (Nevalainen 1999: 336) took place (cf. also overview in Chapter 3). The fact that a particular affix occurs rather frequently may thus be a consequence of morphological productivity in the past. As Pounder (2000: 134) notes, however, frequency bears no direct chronological relation to productivity: Firstly, if an operation is productive, lexemes will be produced that are not ‘recorded’ or attested in the lexicon, and hence will not be counted. Secondly, the items in the list of attested lexemes will have been formed at different stages of increasing or decreasing productivity of the operation. […] Thirdly, there may be a good number of words of a given type retained in a language, even if the operation in question is no longer productive, that is, no longer used to create words.

106 Heterogeneity in Word-Formation Patterns

In his discussion of the problem of productivity in word-formation, Kastovsky (1986a) describes two different functions in word-formation patterns – labeling versus recategorization15 – and sees the productivity of the different types as “basically different phenomena” (1986a: 597). He shows that in the historical development of the suffix -ee, the word-formation underwent a process from the labeling type – e.g. in the early legal French loan words such as feoffee as a correlative noun to feoffor – to a “possibility of a reinterpretation as syntactic recategorization” (1986a: 598), seen especially since the seventeenth century with an increase of the possibility of verb-derivation (cf. historical overview Chapter 3): With the gradual increase of such deverbal derivatives, the function shifted more and more toward syntactic recategorization. But now, the more conspicuous the function of recategorization became, the more productive the type turned out to be.

Kastovsky notes that the change in productivity of the suffix -ee in English is thus as “a particularly striking example of the interdependence of function, scope of a rule, and productivity in the sense of actual application” (1986a: 599). For an analysis of diachronic productivity, the use of dictionaries as a source is inevitable, despite the restrictions – many successfully produced lexemes of that morphological operation might simply not be recorded – mentioned here and before. Based on my OED search of first entries of -ee words recorded over the centuries (plus various other sources for the twentieth century), the development of -ee formation can be shown in absolute figures as follows in Table 10. With the exception of the eighteenth century, the productivity of -ee words – as they are listed in the OED and other sources – has increased progressively, with the largest leap in the nineteenth century (for possible reasons and influencing sources, cf. Chapter 3). As Pounder (2000: 134) remarks on the use of frequency measures in studies which deal with historical material and focus on language change, “diachronically speaking, an increase in frequency may serve as a reliable indication of increasing productivity, as may a decrease indicate a relative decline of productivity. Finally, from a practical perspective, there are few other indicators of productivity in diachronic work […].”

15. According to Kastovsky (1986a: 595, 1986b), syntactic recategorization occurs when parts of sentences are transformed by processes of nominalization, verbalization, adjectivalization, or adverbialization.



Chapter 4.  Morphology and the lexicon 107

Table 10.  The growth of productivity of affixation with -ee, based on entries in the OED (14th to 20th century) & various sources for the 20th century (cf. Appendix 1) Century

Attested number of -ee items

14th century 15th century 16th century 17th century 18th century 19th century 20th century Total

   4   12   21   47   40 152 23516 511

250 200

14th 15th 16th 17th 18th 19th 20th

150 100 50 0

-ee words

Figure 6.  Increase of -ee word-formations (14th–20th century)

16. Based on the data material presented in Section 4.4. The numbers cited for the twentieth century are from a wider range of possible sources than those cited for the previous centuries, hence the comparability of sources is not absolute. Including other up-to-date sources may be justified, however, if we consider that the OED lists words in retrospect rather than in the moment when they first emerge. At the time of the research, the twentieth century had ended only five years before. To list only the 68 OED words which were already included in the dictionary for the twentieth century at the time of the research might be tantamount to putting the twentieth century at a disadvantage – after all, there are hardly any -ee words which have yet made it into the OED after the 1970s. It is clear that the number of attested words greatly depends on the recordings. As one anonymous reviewer remarked, “who knows what we might have found if we had access to fourteenth century blogs!”

108 Heterogeneity in Word-Formation Patterns

(v) Synchronic productivity It is evident that synchronic productivity of a morphological operation cannot be regarded as completely separate from its diachronic version, as the production of new words at a given point in time is also based on the quality and quantity of previous formations. Synchronic productivity, i.e., the quality and growth rate of a particular morphological process at a given point in time, is naturally not restricted to very recent operations. Yet, as we have seen in our discussion of quantitative productivity, the methods of data collection and analysis with large computer corpora allow a broader database to measure the synchronic productivity of, for example, a particular affix in comparison to other affixes. 4.4

Productive patterns and features of twentieth-century -ee words

Having discussed various notions of productivity, I will now turn to -ee formations and the patterns used in this morphological process in the twentieth century. As we have seen above, the question of productivity and the development of neologisms is not connected to the more recent formations or, indeed, the twentieth century. After the analysis of the syntactic and semantic constraints underlying the production of -ee words (Chapter 2) and the historical development of -ee words (Chapter 3), the theoretical issue of productivity will now be connected to an analysis of new words produced in the twentieth century, as far as they are documented in the OED, Bauer (1987, 1993, 1994), Barker (1998), as well as various other sources (cf. also Appendix 1). Bauer (1994: 40) notes that he had commented earlier (1983: 250): on the basis of very little data, that there seems in recent English to have been an increase in the non-passive meaning of the suffix. Subsequently, in Bauer (1987), I reported the arrival of a new meaning for the suffix, with nouns in -ee being used to denote non-humans, especially in technical terms in linguistics. We can thus observe a change in the meaning of the suffix going on in current English.

In the tradition of a diachronic perspective on productivity, he compares the syntactic patterns of 100 nineteenth-century words using -ee (based on OED1) and 60 twentieth-century words (which he collected himself) (Table 11). The category “ambiguous” was not given for the nineteenth-century words. In the twentieth century catalogue, he cites the words listed in this category as “all ambiguous between a subject and some other reading. They are charteree, retiree and returnee. It is not clear whether charteree should be glossed as ‘a person who charters a boat’ or ‘a person to whom a boat is chartered,’ since charter



Chapter 4.  Morphology and the lexicon 109

Table 11.  Comparison of 19th and 20th century syntactic patterns in -ee words according to Bauer (1994: 41, 46) 19th century Syntactic pattern Direct object Object of a preposition Subject None of these Ambiguous Total

20th century

N=%

Human N

Inanimate N

Total %

  54   28    2   16 – 100

27  7  6  6  3 49

10  0  1  0  0 11

  62   12   12   10    5 101

allows both uses. Is a retiree ‘a person who retires’ or ‘a person who has been retired’?” (Bauer 1994: 46). He concludes from the comparison of the syntactic patterns in nineteenthand twentieth-century words that “it is clear that the number of -ee words which act syntactically as the object of a preposition is falling in this century, while the number of subject formations is on the increase” (Bauer 1994: 46–47). The rise of direct object formations across the centuries has been substantiated, not only for the nineteenth and twentieth centuries but also for the preceding periods, as was discussed in Chapter 3. As we have seen from the detailed analysis in the historical part of this study, subject formations and non-human -ee words were already attested rather early (1537: absentee; 1602: co-patentee as agent nouns; 1691: patentee as a first non-human usage of a lexeme which had already been used earlier for person reference). Therefore, Bauer’s estimation, that “the use of -ee to derive subject nouns is basically a twentieth-century phenomenon [which] is confirmed by the fact that the two nineteenth-century subject formations are first attested in 1875 and 1880,” is not quite accurate (cf. Chapter 3). Equally, his claim that “the use of -ee to denote inanimate entities is not only a twentieth-century phenomenon, but a late twentieth-century phenomenon, dating from the 1970s”17 can be disputed on the grounds of the data presented in Chapter 3. Nevertheless, the rise of both phenomena in the twentieth century is unquestionable, as can be seen in the Table 12 below, with the formations collected according to date.18

17. He gives the earliest date for such a word as 1977 – from his list of twentieth century words (1994: 42–45) this would be ascendee (1977) and causee (1977). 18. Note that there are more attested -ee words in the twentieth century (cf. Table 11). However, at this point, only those items will be discussed for which a date and context is given.

110 Heterogeneity in Word-Formation Patterns

Table 12.  Twentieth century -ee words from various sources (OED, Bauer, Barker,19 Others),20 according to date 1900: 1902: 1910: 1911: 1916: 1916: 1917: 1918: 1918: 1922: 1925: @ 1927: 1928: 1928: 1932: 1934: 1934: 1934: 1935: @ 1937: @ 1938: 1938: 1939: 1939: 1940: 1940: 1940: 1940: 1940: 1940: 1941: 1941:

lecturee sayee (rare) amputee representee (Law) parolee rushee (U.S. college use) enlistee dilutee internee guidee absentee (absent voter) tutee directee detainee testee (person underg. tests) counsellee enrollee (N. America) evacuee mergee pledgee (sb who takes a p.) purgee wardee (nonce word) billetee, billettee releasee (from captivity) bombee sockee crackupee pollee quizzee selectee (U.S.) exploitee inductee

1971: 1971: 1971: 1972: 1972: 1974: 1974: 1975: 1977: 1977: 1977: 1978: 1978: 1978: 1979: 1979: 1979: 1980: 1980: 1980: 1980: 1980: 1980s: 1981: 1981: 1982: 1982: 1982: 1982: 1983: 1983: 1983:

narratee (chiefly literary theory) retardee (U.S.) @ adaptee rehabilitee muggee preceptee (U.S., mainly medical) drainee abductee @+ ascendee † causee pleasee @ sitee marquee/markee befriendee manipulee (earlier: manipulatee!) † deletee † reorderee secondee † determinee bumpee explainee @ waitee indemnitee (U.S.) aggressee erasee † benefactee † controllee † malefactee † possessee † dislocatee (2, cf. 19th C: person) communicatee conjuree

19. The particular source is made transparent by the choice of font, i.e., OED = bold type, Bauer = unmarked, Barker = italics, Others = underlined. 20. Only one source is given for each item, even though it may occur in more than one source (cf. Appendix 1). Thus, a word which is listed in the OED but was also cited by Bauer is marked here only as “OED entry.” Similarly, a word cited already in Bauer and reiterated in Barker is only given as “found in Bauer” (as the earlier source). For a more detailed documentary of citation for each word, cf. Appendix 1.



Chapter 4.  Morphology and the lexicon

Table 12.  (continued) 1941: 1942: 1943: 1944: 1945: 1945: 1950: 1950: 1951: 1952: 1952: 1953: 1955: 1956: 1956: 1957: 1958: 1958: 1959: 1959: 1960: 1961: 1961: 1963: 1963: 1964: 1965: 1966: 1968: 1968: 1968: 1969: 1969: 1970: 1976:

rejectee (Mil.) (@) abortee plannee @ returnee (orig. U.S.) @ forgettee @ retiree (U.S.) rapee rescuee missionee offeree socializee @ signee cohabitee co-optee laryngectomee @ recoveree (illness) manipulatee promotee franchisee (=franchised) restrictee contactee attendee (orig. & chiefly U.S.) tippee (hint, orig. & chiefly U.S.) placee appraisee (orig. U.S.) mergee mentee franchisee (=franchiser) † deadee @ embarkee tipee (money) honoree (†) acquiree (person or thing) @meetee leakee

1984: 1984: 1984: 1984: 1984: 1984: 1984: 1985: 1985: 1986: 1986: 1986: 1986: 1986: 1986: 1986: 1986: 1986: 1987: 1987: 1987: 1987: 1987: 1988: 1988: 1988: 1989: 1989: 1989: 1989: 1990: 1993: 1994: 1994: 1994:

†

†

† @

@

@ @ @ @ @

governee advancee constipatee exposee inquisitee interrogee likee eliminatee haulee controllee blackmailee followee fuckee harpoonee shadowee slaughteree slittee torturee cliticee auditionee bowee executionee tastee takee tailee mergee blind datee hittee usee (2) arrivee communicee relaxee returnee dinee offendee

Among these 134 twentieth century entries of -ee words, there are at least 22–23 subject formations (≈ 17%) and approximately 14–15 non-sentient and non­human references (≈ 11%), clearly more than was shown in my nineteenth­century data (cf. Chapter 3). Six out of 152 nineteenth-century words were subject formations (4%), and four out of the total 152 words had non-human references

111

112 Heterogeneity in Word-Formation Patterns

(2.6%). Despite the data from the two centuries not being exactly comparable, this comparison is in line with Bauer’s estimation given above. Unfortunately, date and context are not given for all twentieth-century -ee words attested in the various sources (e.g. American Speech (various sources); Barker 1998; Lehnert 1971; Muthmann 1999; Walker 1936; Wood 1943) (cf. Appendix 1). The lack of context makes it impossible to decide whether a new word has a non-human reference, and difficult to estimate whether it might have an agent meaning. Although it is highly possible that a derivation from an intransitive verb base, like for instance, standee, produces an “agent-type” word, this is not always the case. Besides, there are many ambiguous meanings (e.g. charteree, franchisee) in between where the context really determines the ultimate denotation of the word. Therefore, the following list of another 101 attested twentiethcentury -ee words without date or context is mainly given for the sake of comprehensiveness (Table 13). Table 13.  Twentieth century -ee words from various sources (OED, Barker, Others), not dated actee admittee affrontee allocatee asylee avowee basaree beateree best wishee bitee canee catapultee charteree citee cloutee compromisee conferree congratulatee conscriptee creditee deferree depurgee designee dictatee distinguee

dragee dumpee editee ejectee eliminee entertainee exchangee extractee feedee festschriftee fillee financee firee foolee forcee giftee hackee handshakee hazee head-bonkee huggee indoctrinee infiltree invadee knockee

libelee licencee lovee lunchee magnetizee markee moneylendee operee optionee parkee party nominee passee philantropee pickpocketee piedpipee politicee politico-politicee publishee punchee puntee queree raisee razee readee readjustee

recruitee reduncantee rentee retrainee return addressee roastee scendee separatee sharkee sittee smackee snatchee squeezee stagee stickee talkee teasee telephonee throwee toppee tryoutee vauchee venerealee wishee withstandee yellee



Chapter 4.  Morphology and the lexicon 113

Even without resorting to neat statistical measures, one can detect a number of phenomena when looking at both lists of twentieth-century -ee words: a. In comparison with the eighteenth and nineteenth century, the influence of borrowings from French has decreased (exception: distinguee). For the development of English vocabulary in general, one can observe a decline of lexical influence through borrowings from other languages, including French. Bauer (1994: 35) cites French as a source of new words in the period of 1880–1913 with 5.5%, for 1914–1938 with 3.6% and for 1939–1982 with 2.7%. If we take continued borrowings from French as a factor that enhanced the importance of the -ee suffix in earlier periods of time, then the twentieth century sees a decrease in this significance. b. During the periods of both World Wars, military- and war-related -ee items can be found in abundance, for example, enlistee (1917), internee (1918), releasee (from captivity, 1939), bombee (1940), sockee (1940), selectee (U.S.A., 1940), rejectee (Mil., 1941), returnee (orig. U.S.A., 1944). Dwight Bolinger (1941: 306) even associates the semantic characteristics of the word-formation with the spirit of the time and sees the wealth of newly formed -ee words in the 1940s in context with the “passivity” of the war years: Whether sped by times in which human beings are becoming accustomed to having more and more done to, for, and at them, it would be difficult to prove; but the fact is that the suffix -ee, denoting passivity, has gained a flexibility never before seen. Most, but not all, of the new derivatives have to do with war, or, in this country, with military description.



As an example, he cites newspaper sources from 1940 and 1941 where, next to the older draftee and trainee, selectee is cited for the first time; furthermore bombee, purgee (‘a victim of a political ‘purge’’), pollee (one polled by a public opinion ‘institute’’), sockee (‘one socked, i.e., struck’) and evacuee (which the OED lists for 1934) are also cited. Moreover, deportee was “revived” to new uses. c. As could already be observed in the nineteenth century, humorous formations – like rushee, crackupee, handshakee, quizzee and squeezee – still make up a considerable proportion of twentieth-century -ee words. d. As a rather new development in the twentieth century, compounds with -ee formations (for example, blind datee, handshakee, head bonkee, moneylendee,21

21. The compound base here is supposed to be understood as moneylender rather *to moneylend.

114 Heterogeneity in Word-Formation Patterns

party nominee, return addressee), as well as -ee formations with phrasal verbs (for example crackupee, tryoutee) can be detected.22 Generally, one can observe that the semantic field of -ee formations has become more diverse still, and that twentieth-century -ee words appear in legal contexts (e.g. admittee, libelee), in the context of crime (e.g. muggee, rapee), interpersonal relations (e.g. separatee, cohabitee), in business relations (e.g. optionee, franchisee), war items (cf. above), in humorous environments (e.g. lunchee, tryoutee), as well as in registers of highly colloquial speech (e.g. head-bonkee) or even taboo words (e.g. fuckee).

4.5

Among the new words: Nonce words, neologisms and processes of establishment

Given the relatively large number of new -ee formations in the twentieth century, it is rather striking that there are hardly any examples given in a number of latetwentieth-century “Dictionaries of New Words,” such as Green (1992), the Oxford Dictionary of New Words (1991), or even Barnhart et al. (1990): while Green does not offer a single example of an -ee word and the ODNW notes only one addition for UK English (muggee, 1981), Barnhart et al. have a separate entry for the suffix as well as a number of twentieth-century -ee formations, including one example not found in any other sources: deadee:23 (1968) n. portrait of a dead person painted from a photograph. [Source:] N.Y. Times (1968): “There are ‘deadees’ (painted after their lifetimes) and portraits from life.”

One can assume that the policy of including a new word in such a dictionary rests on somewhat subjective criteria (e.g. on “whether or not the general public was made aware of the word or sense,” ODNW 1991: v). Does this low visibility suggest that the twentieth-century formations are made up mainly of nonce words? 22. This, in fact, counters one of the “generative” morphological principles, i.e., the so-called “no phrase constraint” (NPC), for examples, cf. Bauer (1983: 164). 23. Note that this is also an interesting example of a word-formation which Barnhart et al. (1990: 152) attribute to adjective derivation in their entry on the -ee suffix: “a very productive suffix meaning ‘one who is ___ed,’ freely added to verb stems to form nouns paralleling agent nouns in -er; but also added to intransitive verbs and meaning ‘one who ___s’ (as in escapee); and sometimes to adjectives (as in deadee).”



Chapter 4.  Morphology and the lexicon 115

One of the reasons for the lack of attention in dictionaries of neologisms may be that derivations – and especially when they are seen as part of a rather regular and productive word-formation pattern – receive less note than other types of neologisms (as, for instance, eponyms, blends, acronyms, compounds). Derivations are also less likely to result in a process of what Lipka (1975: 200, 212, 2002: 22–23), following Leisi, calls “hypostatization,” i.e., that the existence of a word implies the existence of a single entity denoted by it.24 Despite the fact that a smaller number of twentieth-century than nineteenthcentury -ee words are openly classified as nonce words – this may be attributed to the fact that there are fewer OED listed words, where such a classification is generally made – quite a few of them can be assumed to be nonce formations in the sense of the OED definition. Here, the meaning of nonce – ‘for the occasion, for the time being, temporarily’ is extended to lexical usage: […], the term used in this Dictionary to describe a word which is apparently used only for the nonce […], similarly nonce-use, etc.; similarly nonce-borrowing, -combination, -form, -formation, -meaning.

In the discussion of the distinction between morphological productivity and creativity (Section 4.2), I have already briefly touched upon the phenomenon of nonce words or ad-hoc creations, which Bauer (1983: 45) describes as “a new complex word coined by a speaker/writer on the spur of the moment to cover some immediate need. This definition admits new words as nonce formations even when they are totally regular, and even if they go on to become accepted in the language community […].” The last assertion Bauer makes seems a little puzzling, in that here neither a creative aspect (“extending the rules of morphology”) nor a temporary and purpose-restricted nature appear to count as a characteristic of the nonce formation. According to Bauer (ibid.), not even the fact that the lexeme is coined independently by different persons prohibits his understanding of the term: Bauer (ibid.) states that “It is perfectly conceivable that the same form should be coined by different speakers, either at different times, or very close together in time (as when some new object appears which requires a name) without affecting the item’s status as a nonce formation.” This certainly cannot imply that, once a nonce formation is made, it can never again lose this status. In Fischer’s words (1998: 3), a nonce formation may move on to become a neologism, “[…] a word which has lost its status of a nonce-formation but is still one which is considered new by the majority of the members of 24. Hypostatization is particularly characteristic of (but not restricted to) creative neologisms in science fiction (cf. Hohenhaus 1996: 317–322; Kastovsky 1978) where fictional entities (matter converter, space warp, etc.) are described as “given” in the fictional world.

116 Heterogeneity in Word-Formation Patterns

a speech community.” Bauer (1983: 45) bases the change in category on speaker awareness: A form ceases to be a nonce formation as soon as the speakers using it are aware of using a term which they have heard already: that is to say, virtually immediately. Despite this fact, there are large numbers of nonce formations which are used on very few occasions (perhaps no more than once) and, in the cases where they do appear more than once, they are used by different speakers, so that their status as nonce formations is not affected. Typically this is the case where the “immediate need” which gives rise to the formation is unique or extremely rare.

Ad-hoc creations are a rather frequent and everyday phenomenon and we may regularly hear quite a number of nonce words without being aware of them, or hearing and using them again. For example, Bauer (1983) cites a study by Thiel (1973: 379) who analyzed 1,331 compound nouns from a 1970 issue of the German magazine Die Zeit and found that 37.9% were listed in dictionaries while 62.1% were not. However, not all of these unlisted and new words are necessarily nonce words, and “although it might be expected that a larger proportion of the derivatives in such a corpus would be listed in dictionaries, the figures are still suggestive” (Bauer 1983: 46). The road towards lexical establishment may thus be described as starting with an ad-hoc creation out of an immediate need. If speakers become aware of the lexeme while it is still considered “new” by the speech community, it will then become a neologism. Elements such as “frequency, neighborhood density, and semantic priming” (Fischer 1998: 5, following Lehrer 1996) influence a further institutionalization. Increased productivity and institutionalization may then have a furthering influence on one another: An increasing degree of productivity can be viewed as a mark of institutionalization that is either currently taking place or has already been repeated. This does not mean, however, that productivity will necessarily follow. In turn, institution(Fischer 1998: 180) alization is positively effected by productivity. 

The reverse process can also take place and a word may become obsolete after a period of regular or frequent usage. Ultimately, there are many reasons why a lexeme becomes established and another one does not. It is clear, however, that classifications such as “nonce word” or “neologism” are not static categories but part of a process in lexical formation and establishment. Certainly, the classification found in a dictionary such as the OED must therefore be regarded with caution and a word classified there as nonce word might, in fact, be an institutionalized word by now.



4.6

Chapter 4.  Morphology and the lexicon 117

Actual words and possible words: Towards an empirical study

In the discussion of productivity above, the traditional separation between actually attested words from non-attested words – which, nonetheless, are deemed well-formed and possible – plays a considerable role. But what exactly are such “actual” versus “possible words”? The definition of a possible word seems rather straightforward: Plag (1999: 7), for instance, notes that a possible word can be defined as “a word, existing or non-existing, whose morphological or phonological structure is in accordance with the rules of the language.” In his words, then, an existing word can also be a possible word. But what gives an actual word its status as an actual word? A simple characterization would be ‘those words that are in use.’ But when precisely is a word “in use”? Is the specification based on the usage of a single speaker or a number of speakers? Is the specification based on the majority of the speech community or the fact that it is listed in a dictionary? When does a possible word become an actual word? Dictionaries, it seems, are still the default proof of the status of actual words, even though the weakness of this interpretation is very obvious (cf. Baayen & Lieber 1991: 803) and has been acknowledged even by researchers relying on the dictionary as the primary or only source: One of the problems concerning -ee nominalizations is the apparently limited productivity of the process. Bauer (1983: 250) points out that while some types are not very productive, at least one type seems to be increasingly productive. However, as soon as one starts to look for evidence of the alleged productivity one gets thwarted, for the primary source to find new cases is dictionaries, where (Portero Muñoz 2003: 130) not many examples of nouns in -ee are attested. 

As we have seen in Chapter 3, there are a few, but it is true that new cases are few and far between: as far as first entries of -ee words in the OED are concerned, the most recent dates of occurrence are now in the 1980s. One should therefore avoid the surprisingly common equation of productivity with listings in dictionaries. References like dictionaries of neologisms (e.g. Barnhart et al. 1990; Green 1991; ODNW 1991), but even standard reference works like the OED are rather idiosyncratic in their inclusion of new words and, certainly, many words which are somehow “in use” are excluded from the listings in such dictionaries. Therefore, the first problem with the definition of an “actual word” is one of documentary evidence and the selection of criteria of how widespread the knowledge of an actual word has to be. The latter point is related to the second problem in the definition of an actual word: a speaker may not be able to decide whether a given word actually exists or not. As experiments with informants have shown (Aronoff 1983: 166; cf. also

118 Heterogeneity in Word-Formation Patterns

Di Sciullo & Williams 1987), speakers tend to judge potential words as actual words even if they are not (see also Schröder 2008). Plag (1999: 8) concludes from this result that “First, it simply shows that the speakers’ intuitions cannot be used to tap this distinction. This does not at all imply that the distinction is non-existent. Second, it tells us that highly productive processes create words that are so similar to many actual words that the two are no longer distinguishable.” The last point seems almost self-evident; after all, one of the characteristics of a potential word is that it is morphologically well-formed and regular. Coming back to the distinction between morphology and the lexicon, this means that even if it were an actual word, it would not get a “full entry” (Jackendoff 1975) in the mental lexicon but the speaker would be more likely to generate it from the set of morphological rules which fit the word-formation pattern (cf. also Aronoff & Anshen 1998). This is not always the case for “real actual words,” which may be much more irregular and idiosyncratic – due to processes of language change or to creative expansions of the rule – than potential words are. Thus, the difference between an actual and a potential word is that a potential word could always be an actual word, whereas an actual word, if not listed somewhere, might not be included in the list of potential words. Yet, as we have also seen above, it is those actual words which deviate from the established patterns that are the most influential in changing the morphological patterns over time, since they are the basis of new words formed by analogy which might then result in new rules. Potentiality is one of the fundamental features of morphological productivity, as Plag (2006: 127) also observes: “The productivity of a word-formation process can be defined as its general potential to be used to create new words and as the degree to which this potential is exploited by the speakers.” But how do we go about finding “potential words”? Portero Muñoz (2003: 130) points out that […] the task of morphology is the enumeration of the class of possible words of a language, which means that we need an overgenerating morphology, that is, one that generates not only words that are familiar to any particular individual, but also unfamiliar ones, and not only words that are to be found in any particular reference book […]

This is an empirical problem which cannot even be solved by searching for individual neologisms – which are, if produced, actual words – in one of the established large computer-searchable corpora. In order to find evidence for the productivity of contemporary patterns of -ee formation, we would need to find out if and how potential words are actually produced across a large range of natural data.



Chapter 4.  Morphology and the lexicon 119

The next chapter will therefore be devoted to an empirical test of the productivity of 1,000 potential -ee words, i.e., unattested and unlisted words in the largest corpus of contemporary (written) text, the World Wide Web. The focus of Chapter 5 will be on a discussion of using the web as a corpus, on the methodology and procedure of designing the test sample and on the results of the test sample.

chapter 5

A corpus-based analysis of 1,000 potential new -ee words

So far, this study has been concerned with the production and productivity patterns of earlier -ee words. This chapter will now investigate the question of how to determine the potentials of new -ee word realizations. The first issue that needs to be addressed in this context is the question of data source. As pointed out before, dictionaries notoriously lag behind current language usage. Many of the twentieth-century neologisms were found in rather random sources and by chance (e.g. newspapers, literature, other personal and public forms of communication) and are not listed in the OED – thus, dictionary listings do not reflect the presence of a word in public usage. A further problem is that dictionaries of neologisms often do not include derivations if their meaning is predictable. While there are quite a few nineteenth-century nonce words in OED listings which, apart from citations in linguistic studies, might have been really used just once – e.g. the jocular shavee – others are not – e.g. admittee (‘a member admitted to the legal bar,’ attested first in Lehnert 1971), optionee (‘someone who has been granted stock options,’ noted in Barker 1998), or honoree (‘a recipient of honors in recognition of noteworthy accomplishments’ attested first in Bauer 1983 for the year 1969); all of these have rather wide and long-standing records of usage. For instance, admittee, a word from the legal “home-turf ” of -ee words, could be found in more than 22,000 English language webpages in January 2006, mainly in legal texts, like in the following webpage on “Rules of the Supreme Court of the State of New Hampshire:” admittee: Each person seeking to practice law in New Hampshire is required to attend a practical skills course to be presented annually by the New Hampshire Bar . As will be explained in Section 5.2, the number of hits given, for instance, in an English language websites google search, cannot be taken as absolute figures, as there are many sources for errors. However, one might make allowances for inaccuracies of the total figures, given the sheer dimension of the potential hits found for these examples.

122 Heterogeneity in Word-Formation Patterns

Association. The course will assist new admittees in developing basic lawyering skills and in gaining practical knowledge of New Hampshire practice and procedures. Attendance is required and each new admittee will be required to execute an affidavit stating that he or she has attended each session of the course unless otherwise excused by the Supreme Court. (www.courts.state.nh.us/rules/scr/scr-42.htm)

Similarly, optionee is now a very common word in the language of the stock market and more than 50,000 English language websites have been found to contain this item. There are in excess of 1.5 million hits for honoree, many like the example below, from the website of the American Football Coaches Association: honoree: Please enter the last name of the honoree you are trying to find. You may also enter the first initial to filter your results: however, the first initial is not required. (www.afca.org/afcf/searchForm.cfm)

Notwithstanding the fact that entries can now be found in online dictionaries, for this particular item (honoree), lexical innovations and ad-hoc creations in spoken or written usage are scarce in established dictionaries due to selecting, editing and filtering processes in these publications. When we want to explore the possibility of new -ee words, the first issue to consider is the data source. Rather than relying on chance finds in random sources, an obvious choice to search for new words would be the use of larger language corpora, such as the British National Corpus with its 100 million words (cf. Aston & Burnard 1998). While any of the established corpora have advantages with regard to easy accessibility and simple searchability according to linguistic criteria (word-class categorization, subcategorization, context, collocations, etc.), there are also restrictions with regard to language varieties – the British National Corpus (BNC), obviously, is only interesting for varieties of British English – genre and topic (e.g. specific newspaper corpora) and up-to-date material (the Brown corpus,

. Accessed January 2006. . Accessed January 2006. . Corpora with other geographical varieties of English are, for instance, the Macquarie Corpus of Australian English (Collins & Peters 1988) or the Kolhapur Corpus of Indian English (Shastri 1988). The International Corpus of English project (ICE) is an ongoing venture. Scholars from different parts of the world are here creating a corpus of each of the major international varieties of contemporary English with similarly designed patterns for each variety (Greenbaum 1992; cf. also www.ucl.ac.uk/english-usage/ice/).



Chapter 5.  A corpus-based analysis of new -ee words 123

for instance, is from the 1960s). Some of the standard corpora contain only written texts (e.g. Brown, written American English – Kucera & Francis 1961; LOB – London-Oslo-Bergen corpus – written British English, Johansson et al. 1986), others are restricted to spoken material, i.e., transcriptions of conversational data (e.g. London-Lund Corpus – Svartvik 1990). For the search of neologisms, and especially recent neologisms, even large computer corpora are not representative enough and, like dictionaries, lag behind current usage. A search of -ee words in the BNC, for instance, merely produces the most established and standard words with this suffix, for example, employee, referee, refugee, trainee and trustee. The one large body of text collections which is unbeatable with regard to size, diversity of Englishes, varieties of users (i.e., here producers of text), text types and recentness is the World Wide Web, a corpus which has gained in importance in innovative linguistic research in the last few years (cf. Hundt et al. 2007; Kilgarriff & Grefenstette 2003; Resnik & Smith 2003). While there are also shortcomings in the uses of the Web for particular types of linguistic analyses (cf. discussion in 5.1), the advantages easily outweigh them. The adequacy of this source ultimately depends on the research question and the procedures involved in resolving it. In the following Section (5.1), the debate on the validity of using the Web for linguistic data search will be followed up and supplemented with the specific conditions of Web usage in this study. The second concern is the selection of the search criteria. Due to the high frequency of ‘double e’s’ that are not suffixes in English – for instance, in words like leek, peer, etc. – a simple search in uncoded Web material for the suffix -ee would result in an impossible amount of rather chaotic data and would be unanalysable. Unlike in established and coded corpora, a search for *ee# is not possible. An alternative option is the search for specific lexemes, with their frequency and context of occurrence. The problem with prospective neologisms is that the searchable items cannot be found in any dictionaries or word lists – this would negate the whole point of the endeavor. Therefore, a set of 1,000 potential search words – non-existent in common dictionaries or scholarly articles – had to be created as a test pool for -ee neologisms (see also the discussion of actual versus possible words in Chapter 4). This approach has the advantage of leaving ample room for an observation of word-formation processes at work in contemporary and recent -ee formations. It also avoids the danger that Bauer (1983: 290–291) pointed out, which lies in a post-hoc explanation of the patterns of selected cases: “it is far easier to justify the use of a particular word-formation process post hoc than it is to predict which process will be used before the event […]”. He concludes:

124 Heterogeneity in Word-Formation Patterns

[…] it is by no means always possible to predict which word-formation process will be used to form a new lexeme in any particular instance, but that the formation of new lexemes is, nevertheless, far from random. Despite the dangers of posthockery implicit in the approach that has been taken here, it can cast some light on word-formation which other approaches would not be able to.

The present study is an attempt to evade such “posthockery” and to provide the groundwork for a large data analysis without selecting just a few cases. A more detailed account of the specific research methods and choices this investigation has adhered to will be given in Section 5.2.

5.1

Web data for linguistic purposes: Searching corpora and searching the Web

Computer-based language study can be dated back to the early 1960s, when the Brown corpus opened up the opportunity for computer-assisted analysis of what was then considered to be a large body of text: one million words. Since that time, the size of corpora has greatly increased, as has its significance for particular types of (quantitative) linguistic analyses which require large and quantifiable sizes of data. It was not until the late-1980s and early 1990s that the use of corpora became common for a wider scholarly community but ever since, large text collections like the BNC with its 100 million words have become a standard tool for linguists. However, the question of what exactly makes a corpus a corpus, is subject to ongoing debates. McEnery & Wilson (1996: 21) contend that, in principle, any collection of more than one text may be called a corpus; but they also draw up the boundaries of the definition according to a set of criteria: But the term “corpus” when used in the context of modern linguistics tends more frequently to have more specific connotations than this simple definition provides for. These may be considered under four main headings, sampling and representativeness, finite size, machine-readable form, a standard reference.

If we regard these conditions – sampling and representativeness, finite size, machine-readable form and a standard reference – as vital principles, then many text collections linguists have used as their corpus would not fit. After all, the question of what is representative or what is a standard reference is rather multifaceted. . Cf. one of the earliest volumes on corpus linguistics, Aarts (1984). . Cf. also Meyer (2002: 142–150) for an overview of standard corpus resources.



Chapter 5.  A corpus-based analysis of new -ee words 125

Therefore, one of the main issues in corpus linguistics is the question of what the collections of texts in question really represent. Biber et al. (1998: 246) note: […] a corpus seeks to represent a language or some part of a language. The appropriate design for a corpus therefore depends upon what it is meant to represent. The representativeness of the corpus, in turn, determines the kinds of research questions that can be addressed and the generalizability of the results of the research. For example, a corpus composed primarily of news reportage would not allow a general investigation of variation in English. Similarly, research based on a corpus containing a single type of conversation – such as conversation between teenagers – could not be generalized to conversation overall.

We can see here that most of the established corpora are rather limited in their capability to represent “the English language” (or any other language) but, rather, the usefulness of the data collection has to be seen in the light of the research scope. While McEnery & Wilson’s (1996) definition of what counts as a corpus is rather narrow in its approach, others (e.g. Manning & Schütze 1999) have presented a broader view on data collection and corpora: In Statistical NLP, one commonly receives as a corpus a certain amount of data from a certain domain of interest, without having any say in how it is constructed. In such cases, having more training data is normally more useful than any concerns of balance, and one should simply use all the text that is available.  (Manning & Schütze 1999: 120)

The question of whether or not the World Wide Web can be regarded as a corpus must be seen in the light of these conflicting viewpoints. In a recent article on the subject, Kilgarriff & Grefenstette (2003: 334) note that there is a certain amount of confusion in setting criteria for what makes a corpus. They plead for a definition of corpus according to its usefulness when they state that “McEnery and Wilson (following others before them) mix the question ‘What is a corpus?’ with ‘What is a good corpus (for certain kinds of linguistic study)?’ muddying the simple question ‘Is corpus x good for task y?’ with the semantic question ‘Is x a corpus at all?’” Their stance regarding the inclusion of the Web as corpus is patent: We define a corpus simply as “a collection of texts.” If that seems too broad, the one qualification we allow relates to the domains and contexts in which the word is used rather than its denotation: A corpus is a collection of texts when considered as an object of language or literary study. The answer to the question “Is the web a (Kilgariff & Grefenstette 2003: 334) corpus?” is yes. 

Ultimately, it might be said, there is no “general” usefulness or uselessness of the Web or any other corpus but, rather, the effectiveness of the text collection in

126 Heterogeneity in Word-Formation Patterns

question is determined by the specific research question or problem. As Lüdeling, Evert & Baroni (2007: 7) summarize the value of Web uses, for instance, in cases: […] in which the data needed to answer or explore a question cannot be found in a standard corpus because the phenomenon under consideration is rare (sparse data), belongs to a genre or register not represented in the corpus, or stems from a time that the convenient corpus data do not cover (for example, it is too new).

For the purpose of the present study, which, after all, focuses on neologisms and ad-hoc creations, the most obvious asset of the Web is its size, its inclusion of very recent texts and its wide range of unfiltered texts of all types. It is therefore both more useful and more representative than a standard corpus in that a delimitation of time and text type/genre/register is not given here in the same way a corpus of, for instance, British news reportage of the 1980s would be. On the question of “representativeness of the Web,” Kilgariff & Grefenstette (2003: 43) remark that “the Web is not representative of anything else. But neither are other corpora, in any well-understood sense.” The fact that the Web is neither edited nor coded may be seen as an advantage or disadvantage, depending on the research question. The pre-selection of text types in certain corpora, for instance, can be seen as an ambiguous format in that it removes the responsibility of defining criteria for text types from the researcher: ‘Text type’ is an area in which our understanding is, as yet, very limited. Although further work is required irrespective of the Web, the use of the Web forces the issue. Where researchers use established corpora, such as Brown, the BNC, or the Penn Treebank, researchers and readers are willing to accept the corpus name as a label for the type of text occurring in it without asking critical questions. Once we move to the Web as a source of data, and our corpora have names like ‘April03-sample77,’ the issue of how the text type(s) can be characterized de(Kilgariff & Grefenstette 2003: 43) mands attention.

Similarly for other types of coding in established corpora versus the Web: whereas with standard corpora the researcher has to trust and rely on the selection, editing and classification processes of other persons, he or she has to do it themselves when using the Web as a data source. As pointed out before, the size of the Web beats any other corpus. Exact or even approximate figures are difficult to obtain. Kilgariff & Grefenstette (2003) give estimates for the number of words that were available in 30 Latin-script languages in March 2001. They found, not surprisingly perhaps, that English . The search was conducted by means of the search engine AltaVista. Kilgariff & Grefenstette (2003: 339) point out some difficulties which arose with this procedure: (a) only a fraction of



Chapter 5.  A corpus-based analysis of new -ee words 127

is the most frequently used language with approximately 76.5 billion words, followed by German (7 billion), French (3.8 billion), Spanish (2.6 billion) and Italian (1.8 billion). It may be safe to assume that these figures from 2001 have multiplied in the last years but, nonetheless, they illustrate the proportion of languages used. In another estimate by Xu (2000) non-Latin-script languages are also included. Xu calculates that approximately 71% of the pages (453 million out of 634 million Web pages indexed by the Excite engine at that time) were written in English, followed by Japanese (6.8%), German (5.1%), French (1.8%), Chinese (1.5%), Spanish (1.1%), Italian (0.9%), and Swedish (0.7%), clearly making the Web a multilingual corpus (cf. Kilgariff & Grefenstette 2003: 337). Using the Web for linguistic research offers new and stimulating research possibilities. In their work on multilingual natural language processing, Resnik & Smith (2003: 350) state: People tend to see the Web as a reflection of their own way of viewing the world – as a huge semantic network, or an enormous historical archive, or a grand social experiment. We are no different: As computational linguists working on multilingual issues, we view the Web as a great big body of text waiting to be mined, a huge fabric of linguistic data often interwoven with parallel threads.

Clearly, linguistic work with data from the Web is still in its experimental phase but the wealth of research possibilities makes it almost inevitable that “this great big body of text” will find many more miners in the future.

5.2

Searching the Web for -ee words: On methods and procedures

(i) Creation of a test set of search words A Web search for possible -ee neologisms requires the creation of a set of searchable items. For the present analysis, a set of 1,000 potential -ee nouns was created by going through the OED and selecting those (mostly verbal) bases for which, according to my sources (see Appendix 1) no -ee derivation had been listed. Each one of them was then tested for its actual realization by speakers/writers in any of the English language webpages. The criteria for the list of potential -ee words was as follows (cf. also Mühleisen 2007): the indexical Web pages are covered; (b) there is a certain bias towards North American webpages; and (c) this search does not cover the “hidden Web.” . Admittedly, this may not be the most sophisticated way to create a set of test words but, at the time, it seemed the most feasible procedure and the closest one can get to a random selection.

128 Heterogeneity in Word-Formation Patterns

1. No occurrences in any of the established sources, such as the OED, selected other dictionaries, or in scholarly texts on -ee words (cf. Appendix 1). 2. Recognizable derivation of either a contemporary English verb (the majority) or -er noun (less frequent). Apart from the above criteria, the choice of bases was alphabetically ordered, yet random, choices from the dictionary (OED), starting with the following ten lexemes: Example: First ten test words No.

Lexeme

  1.   2.   3.   4.   5.   6.   7.   8.   9. 10.

abhorree abidee abjectee abolishee abrogatee absorbee acceptee acclaimee accommodatee accursee

[…]

and ending with the ten items below: Example: Last ten test words   990.   991.   992.   993.   994.   995.   996.   997.   998.   999. 1000.

wonderee workee woundee wrappee wreckee wrestlee wrongee yammeree yearnee yieldee yodellee

(For a complete documentation, cf. Appendix 2).

. At least not by the time of the data collection, i.e., February–June 2005.



Chapter 5.  A corpus-based analysis of new -ee words 129

Factors such as syntactic features (for instance, transitive/intransitive patterns), phonological and orthographic shape,10 apparent meaningfulness/meaninglessness of the derivation or likelihood were not taken into account in the creation of the list, so as not to preselect any features which determine a successful realization or not. It is nonetheless clear that by creating a search set of 1,000 possible words – on whatever random and “impartial” criteria – one potentially excludes other lexemes which might equally be realized (but were not searched for). Again, the advantages of the method chosen – creating a definite set of possible search words as opposed to collecting new -ee words by the chance of their occurrence in random sources (as has been done in previous work on -ee words) – outbalances its weakness of quite a few new -ee words in all likelihood being missed because they were not searched for.

(ii) Using data from the Web There are several choices of how to use data from the Web (cf. Lüdeling, Evert & Baroni 2007): First, by use of a commercial engine for searching the whole Web – Either directly (with, for instance, Google, Metacrawler, AltaVista, Yahoo, etc.) – Or by the use of additional pre- or post-processings to the engine (e.g. WebCorp) (cf. Kehoe & Renouf 2002) to refine the results. Secondly, by creating one’s own corpus by collecting pages from the Web – by downloading pages directly from the Web, for instance, by running Google queries. An alternative would be to use one’s own Web crawler (cf., for instance, Ghani, Jones & Mladenic 2001). The corpus data might then be processed in any way necessary. – “One can collect a corpus by manual or semi-automatic selection of pages downloaded from the Web, according to precisely specified design criteria. This procedure is not different in principle from building a corpus such as the BNC or Brown Corpus, and has the same advantages and disadvantages as these (except that there is much more material without strict copyright on the Web […])” (Lüdeling, Evert & Baroni 2007: 8).

10. Where there were spelling alternatives, such as honoree versus honouree or analysee versus analyzee, both versions were tested. In cases where only one spelling variant was successful, this was used in the list (e.g. organizee versus *organisee).

130 Heterogeneity in Word-Formation Patterns

Direct use of search engines is, evidently, the most practical and uncomplicated approach and the one that has been chosen by most researchers. In this study, the use of Google with some refined search criteria was chosen (e.g. the use of English as the language of the websites). For Chapter 6, the query was further refined to explore websites from specific countries only, by searching in specific domains (for instance, .au; .nz; .uk; .ie, etc.). With regard to access to the Web by means of search engines such as Google, Kilgariff & Grefenstette (2003: 344–345) state that although the search engines are “dazzlingly efficient pieces of technology and excellent at the task they set for themselves,” they also pose some problems for the linguist: a. they do not present enough instances of the search items (1,000 or 5,000 maximum); b. they do not present enough context for each instance (Google provides a fragment of around ten words); c. they are selected according to criteria that are, from a linguistic perspective, distorting (with uses of the search term in titles and headings going to the top of the list and often occupying all the top slots); d. they do not allow searches to be specified according to linguistic criteria such as the citation form of a word, or word class; e. the statistics are unreliable, with frequencies for “pages containing x” varying according to the search engine load and many other factors. The difficulties cited in (a) and (e) mean that Google searches which focus on absolute frequency of an item are of limited value. In my study, however, the search finds will be classified according to approximate frequency categories (cf. below). It is clear that in this type of research, exact figures above a certain frequency are not the main aim. As is pointed out in Biber et al. (1998: 5), “corpus-based analyses must go beyond simple counts of linguistic features. That is, it is essential to include qualitative, functional interpretations of quantitative patterns.” Problems (b) and (d) suggest that the researcher has to make sure (manually) that the search item in question fits the context and word class (cf. description below). As my study deals with neologisms with (usually) a limited rate of occurrence on the Web, the obstacle cited in (c) was not prevalent. However, one characteristic of the World Wide Web which differs from standard corpora and which might be seen as a problem is the fact that the Web is not a finite body of texts. Thus, the whole makeup of what can be found on the Web is in constant flux. The only way to address this obstacle, in my view, is by means of temporal limitation. My research was conducted between February 2005 and June 2005 and the figures and results count for this particular period; it is clear that this



Chapter 5.  A corpus-based analysis of new -ee words 131

will already have changed by the time of writing (or publishing). For practical and technical reasons, not all websites found which contained the search items, i.e., new -ee words, could be downloaded to create a reproducible result. However, a sample website for each search items was downloaded. For the items with only one or two occurrences (classified as Hapax Legomena for the time of the search), a reproducible corpus was created which can furthermore be searched for specific features such as text type, collocations, ambiguities or special characteristics. Again, the very size of the corpus gives rise to some inaccuracies and questions: we do not know who the writers are, what their linguistic background is, or whether they are native speakers. However, in most other standard corpora we also do not have detailed information on the language history or the linguistic influences on the speakers. Above a certain size of results, the hits cannot be checked for accuracy and the results bear many erroneous items. In my search, the results were verified up to the first 10 pages (= 10011 webpages), and had to fulfil a number of criteria:

(iii) Qualitative criteria in the test a. Misspellings or typos were not counted. Typos, like an involuntary adding of an -e to a word which already ends on -e (as, for instance, ‘He communicatee with his friend’ which seems to happen particularly often with some verbs), are probably the largest source of errors but they are also easily distinguishable from “purposeful” -ee words. b. The -ee word had to be part of a coherent English-language text. Even in webpages which are classified as English language webpages, French-language texts are not uncommon (sometimes occurring as part of a translation exercise). For the -ee suffix in particular, French past participle formations can be a source of errors. The item disputee, for instance, was found on both French and English websites. Others, for example quittee, redressee, regardee, were only produced in French texts (at least on the first 10 pages = 100 webpages). If something like that was the case, a cross-check with -ee words in connection with other English language word forms was conducted (e.g. common English words like ‘that’ or ‘quittee, quitting’ or ‘quittee, quits’) to verify whether the word appears in English language texts. In the above cases this did not produce a meaningful English result either, so that they were not counted as a successful English word-formation.

11. New -ee items which occurred in 100 or more webpages were categorized as “frequent” (cf. below) – whether they occurred, for example, 101 or 10,000 times. The results were therefore not verified for accuracy above the rate of occurrence of 100.

132 Heterogeneity in Word-Formation Patterns

c. Many personal names (e.g. first names ‘Desiree,’ ‘Hailee,’ numerous last names), nicknames (e.g. ‘Robbee’ or ‘Hidee’ – the latter as a curious spelling version of ‘Heidi’), or brand names (e.g. ‘Chuckee Cheese,’ ‘Scrubbee Doo’) end on -ee. These had to be carefully eliminated from the result list. d. As has been pointed out in Section 2.6 (iii), diminuitive and hypocoristic meanings, usually represented in writing as -ie (‘cabbie,’ ‘bookie,’ etc.) sometimes seem to merge with the homophonous -ee suffix meaning, especially when the semantic feature ‘non-volitional’ is jocularly combined with a hypocoristic meaning (i.e., as a pet name or term of endearment). After all, the non-volitional position of the referent might inspire a tender or condescending attitude towards him or her (or it), depending on the situation (e.g. in some cases of the use of failee or druggee). Whenever there was overt ambiguity or a multiple interpretation (e.g. surfee, blue rinse; cf. also (i), below), the item was counted. In cases of clear, unambiguous, diminuitive meaning, the items were not considered. e. The -ee lexemes had to be recognizable as nouns. Apart from typos, there are also -ee formations with verbs which jocularly seek to imitate a stereotypical feature of ‘Chinese pidgin’ of the type “Softly, softly, catchee monkey.” Some verbs are particularly affected by this: apart from the catchee of the popular motto cited above, there are also widespread uses of workee, payee, (for instance, ‘no workee no payee’) or lookee, touchee (e.g. ‘lookee but no touchee’) in this manner. Because they were easily recognisable as verbs, their elimination from the result list did not pose any problem. f. The -ee word had to appear in some meaningful context, at least in a coherent sentence: pure strings of unrelated words – which can often be found, especially in commercial webpages – were not considered. g. As could be seen in the diachronic treatment of -ee words (Chapter 3), word play has become characteristic for some types of -ee formations. It is clear that the meaning of the verb base of the new -ee word is often extended, either metaphorically or to adapt to particular new uses. At the time of the research, the meaning of pimpee, for instance, was often not related to the original verb meaning in common dictionaries (‘to pimp – to act as a pimp or pander,’ OED) but was connected to the name of a popular MTV car show called “Pimp my ride”: the pimpee here is someone whose car has been ‘pimped out,’ i.e., given a stylish make-over: pimpee C 559: MTV’s Pimp My Ride – Behind the Scenes – Motor Trend “My car attracts a lot of attention,” admits “Big Ron” Felder, 23, a registration



Chapter 5.  A corpus-based analysis of new -ee words 133

clerk at a Los Angeles college and another lucky pimpee. His radical, sunsetorange 1984 Eldorado, he says, is a far cry from the bucket whose fake vinyl roof blew off on the freeway while he was on a date. (motortrend.com/features/scenes/112_0505 _pimp)



However, no word plays with a meaning completely detached from the base of the derivation were counted; for instance, rackee did not produce a ‘proper’ -ee word result, but there were many puns based on homophony, like eyerackee for Iraquis. h. Repetitions of a word in the same website were counted as only one occurrence. If the same text appeared in more than one website it was also counted as one occurrence. i. -ee words which appeared as part of a compound were counted as valid. Examples of this in my test result include fellow rehearsee (as part of the result for rehearsee, etc.), blue rinsee, road ragee, guest housee, page turnee, lap dancee, party ruinee and home-wreckee: road-ragee C 654: Civility By Nicholas Von Hoffman and Alan Ehrenhalt You say that road rage derives from the reification or what you call “the dehumanization” by the road rager toward the road ragee and that this seldom happened a generation ago. Being almost as old as the Dalai Lama himself, I confess that I was driving a vehicle a generation ago and the biggest change I’m aware of from then to now is a helluva lot more cars. (slate.msn.com/id/3675/entry/24076) party ruinee C755: Northeastern University Times New Roman An Open Letter From Party Ruiner to Party Ruinee. William Bonner. I know your party was very important to you, and I know that you invited me on the on the condition that I remain on my best behavior. (www.nutimesnewroman.neu.edu/v2i3/soapbox_main.html-)

Similarly, derivations from phrasal verbs like followee-up were also counted as valid.

(iv) Quantitative categories The study does not aim at eliciting the exact figures of -ee word occurrences but, rather, approximate counts were categorized according to the following scheme (cf. detailed documentation in Appendix 2):

134 Heterogeneity in Word-Formation Patterns

1. 2. 3. 4. 5. 6.

Not found (No word-formation of that pattern was found on the Web) Hapax legomena (1–212 entries) Rare (3–20 entries) Established (21–100) Frequent (more than 100) Not quantifiable (Many entries found but not quantifiable, for instance, because of competing French lexemes, names, brand names, etc.)

It is important to note that, within the first 100 pages, the validity of the -ee item was ensured according to the above criteria, i.e., only those words were counted which appeared as an -ee noun in some meaningful context. Items which are categorized as “frequent” might vary between just over 100 entries to several tens of thousands – but for more than 100 pages the “meaningfulness” of the entries of each item could not be assured any further. For the purpose of this study it should be enough to verify the significantly widespread use of the -ee words classified as “frequent.” The overall approximate frequency analysis shows that, among the 1,000 words tested, there were: – 252 non-successful ones (i.e., the extended Google query produced either no results at all or no meaningful result in the first 100 pages) – 214 Hapax Legomena (1–2 occurrences) – 267 rare items (3–20) – 97 established items (21–100) – 120 frequent -ee words (over 100) – 50 non-quantifiable items (i.e., large numbers of entries found, among them also meaningful ones but for reasons of, for example, competing French lexemes, names, brand names, etc., the result is difficult to quantify) The proportions may best be illustrated in the diagram in Figure 7. Approximately one quarter of the randomly created test words were not used – which, on the other hand, makes for almost three quarters (748) being successful new -ee words, a considerably higher proportion than expected.

12. The vast majority in this category are “true hapaxes,” i.e., they occur only once. A small quantity of double entries was also included, however, for a number of reasons: (a) to also allow a close and contextualized analysis for these (rare) double entries, (b) to provide a direct contextual comparison for these cases, and (c) an absolute quantitative productivity analysis in the sense of Baayen (1993) is not possible in a Web search, as an absolute figure of words in the corpus cannot be established.



Chapter 5.  A corpus-based analysis of new -ee words 135

Table 14.  Test word analysis according to frequency categories Category Not found Hapax Legomena Rare Established Frequent Not quantifiable

Number of words

Percentage

252 214 267   97 120   50

25.2 21.4 26.7   9.7 12.0   5.0

Not found H. L. Rare Establ. Frequent Not quant.

Figure 7.  Proportions of test words acording to frequency patterns

Among the successful words, the category with the highest frequency is Rare, i.e., the category where between 3 and 20 words were found on different websites. This is followed by single (or double) occurrences (Hapax Legomena); after that, by a considerable margin, is the category with the highest frequency of word occurrence (Frequent). The lowest number of entries among the quantifiable results could be noted in the Established list (the words appear in 21–100 different websites). Finally, the Not quantifiable category leaves 50 test words without approximate frequency results.

5.3

Syntactic and semantic patterns of successful neologisms

As we have established in Section 2.1, the verb is seen as the contemporary base of -ee word-formation – despite many counter-examples given both in the historical development and for contemporary word-formation patterns (cf. Chapter 1; Section 2.1; Chapter 3). Most of the 748 out of 1,000 words which produced a result are verb- derived. We have to note here, however, that this a feature preconditioned by the selection of test words and does not necessarily reflect whether or not there might also be a substantial number of new noun-derived -ee words. There were various reasons for choosing mostly verbal bases for the test words.

136 Heterogeneity in Word-Formation Patterns

One reason concerns the number of possibilities. Even though the -er deverbal noun-derivation pattern is a highly productive word-formation, there are, of course, still more verbs than -er nouns. Secondly, as we have seen in previous sections, there are quite a few examples of -ee word-formations where either no verb exists (e.g. astrologee C 46, therapee C 903, therapeutee C 904) or where a verb as well as a corresponding -er noun exist and the contextualized meaning of the -ee noun suggests noun-noun derivation (e.g. computee C 107, elevatee C 238, professee C 621) (cf. also Section 2.1). In other words, the existence of a potential base verb does not exclude the possibility of the -ee word being noun-derived. A last consideration was the notion that by choosing -er nouns as a base, one might miss potential -er neologisms as well as the new -ee words, i.e., there might also be verb-derived -er words which are not in the dictionary yet. As a consequence of these deliberations, the verb was first taken as the basis for the creation of the test-word pool. The results were then tested for -er (or -or) collocations. The underlying assumption was that the existence (or also new formation!) of an -er/-or word makes an -ee formation more likely.

(i) Collocations Collocations, i.e., the habitual juxtaposition or association of a particular word with other particular words in a particular text unit, usually a sentence (cf. Firth 1957) have also been called “lexical solidarities” (Coseriu 1967) and are semantically motivated. In the present study, the 748 -ee words which produced results in the Google query were subsequently tested for co-occurrence with the corresponding -er word within one text on a website. There were several types of collocation results in this search: Examples (a): Cases in which a direct opposition or even mutual exclusion of the two roles becomes obvious (of the type “I’d rather be the x-er than the x-ee”): anointer/anointee C 33: The question here is who is the anointer and who is the anointee. Certainly the one who does the anointing wouldn’t anoint himself. Why should he? (scatteredsheep.com/anointed_one.htm) bruiser/bruisee C 74: Got bruised on my shoulder, thigh and shin but what’s most important was that I was more of a bruiser than a bruisee LOL. (www.weedisgewds.blogspot.com/-)



Chapter 5.  A corpus-based analysis of new -ee words 137

fusser/fussee C 330: I tend to lead my partner’s suit. If it works badly, I can fuss at him for bidding such poor suits. If I lead my own suit and that’s wrong, he gets to fuss at me. It’s better to be the fusser than the fussee. (www.washingtonpost.com/wp-dyn/articles/A19570-2004Jul27.html) gripper/grippee C 346: President Polk considered handshaking an art to be developed, both for his image and self-preservation. When a man approached, he’d mentally assess his strength and quickly beat a strong man to the shake, ensuring that he was the gripper, not the grippee. President Hoover never did master the art of handshaking. (w3.ouhsc.edu/ehso/saf-t-gram/summer05.pdf-) investigator/investigatee C 420: To the contrary, he is an investigatee, not an investigator, and one who has stubbornly resisted the disclosure of official information […] (www.fas.org/sgp/news/secrecy/2002/11/112702.html) prodder/proddee C 619: The only problem was that we couldn’t decide who should go first. Who should be the prodder and who the proddee. The whole time I kept noticing this group of men who were standing at a distance just watching us. (www.clifftemple.org/sermons/26-Sep-99.html) Examples (b): Cases where both forms occur in one sentence and are clearly related: interpreter/interpretee C 417: So, also, are the differences between the nature of the interpretee as it is already integrated into the interpreter on the one hand and as it stands outside him as the object of (and the objection to) an interpretation on the other. (www.uta.edu/english/rcct/5340luannecrit.html) smirker/smirkee C 812: Smirks and sneers are A needless irritation, Mr. Rogerson, and usually say more about the ‘smirker’ than the ‘smirkee.’ (www.virtuallystrange.net/ ufo/updates/2004/dec/m03-009.shtml) whistler/whistlee C 987: I even discovered that not all whistlers are creeps. One former whistler insisted he never did so with the intent of actually embarrassing the whistlee. Being attracted to young girls is “an affliction,” he explained. (www.ocregister.com/ocr/2005/03/16/ sections/life/life/article)

138 Heterogeneity in Word-Formation Patterns

Examples (c): Where the two forms are confused or not quite clear: auctioner/auctionee C 48: I had never considered the psychology of the auctioner/auctionee – and this guy has captured the online (and live, in-person) auction dynamic and describes it pretty thoroughly […]. (www.amazon.com/exec/obidos/ tg/detail/-/0764505785?v, May 2000) fainter/faintee C 298: Eventually the aforementioned fainter (faintee?) regained consciousness, was helped off the train by a couple of staff members and we proceeded on our way. (www.roberthampton.me.uk/news/archives/2004_07, July 2004) navigator/navigatee C 493: You can’t do a search for files on your PPC (can you???), but I found a post a few pages back (24 pages is rather long) which says that the file data.chk can be located in Program Files/Navigator … and so it was. I replaced it and everything works fine and dandy. One happy navigator (or is it navigatee?). PS All I need to find now is how to create my own sounds/voices. (www.pocketgps.co.uk/modules.php?name=Forums&file=viewtopic&p= 124549, June 2005) slipper/slippee C 807: I too was a slipper or slipee for about eight years before I had that last drink. The days have turned to weeks, the weeks to months and the months to years. (stayingcyber.org/scarchive/dm020302) stinker/stinkee C 855: My associate here says he’d do precisely that, but only if it was a restaurant or other public place where a quick escape was impossible for the stinker (or is it the stinkee?), and thus massive embarassment [sic] was assured. (www.benrik.co.uk/content/blog) Examples (d): Collocations of new -er words along with new -ee words (here, -er words which were not found in the OED): alerter/alertee C 25: Computing.Net – Re-install obviously, but how now? It is just to let the alertee know that the alerter wants more help. All you do is check back on your post to see if the alertee has responded. (www.computing.net/office/wwwboard/forum/1991.htm)



Chapter 5.  A corpus-based analysis of new -ee words 139

amazer/amazee C 27: British Chess Magazine: Chess Book Reviews: July 2000. The author is to be commended for his remarkable modesty; of the examples taken from his own play, he was the ‘amazee,’ rather than the ‘amazer,’ in all three cases. (www.bcmchess.co.uk/reviews/bcmrev0007.html) attracter/attractee C 47: there something about the taken individual (i’m not thinking gender makes things different, but it may) that makes them more appealing? i’ve observed that individuals who have significant others seem to attract others more readily, even if said attractee has no idea that the attracter is seeing someone and even if the attracter is not necessarily trying to attract the attractee. is it the confidence, the lack of really trying to impress whoever and what not, the loss of the “let me get into your pants” ulterior motive? (www.palesky.com/ life/archives/000097.html) demoter/demotee C 154: This would send a text message to the demoter and demotee stating why the demotion was reversed and would return the demotee to his normal rank discussions. (playnet.com/ viewtopic.php?p=107736) disinheritor/disinheritee C 199: The Project Gutenberg eBook of Kenny, by Leona Dalrymple. If anything, the disinheritor was sulking. And the eyes of the disinheritee were intelligent and disconcerting. “Well?” said Garry, amazed. www.gutenberg.org/files /16040/16040-h/16040-h.htm embarrassor/embarrassee C 245: Jazzcorner’s Speakeasy Archive. I can tell you frankly that I’ve never enjoyed being either The embarrassee* or The embarrassor, So when I say that whichever it was that I ended up being was entirely inadvertant, you must try to believe me. (www.jazzcornertalk.com/speakeasyarchive/thread.php? forumid)

The above examples are not exhaustive. Altogether, there are quite a number of -er words in the -ee/-er collocation test which were not so established as to be listed in the OED. They may also be seen as either ad-hoc creations, nonce words or, at the very least, unusual formations. Apart from the examples given above (alerter, amazer, attracter, demoter, disinheritor and embarrassor), the “new -er words” appeared in the collocations test in Table 15. Some of them may be explained by a need to name a new or recent practice, such as in emailer or the possibly more ephemeral pimper (as opposed to pimp,

140 Heterogeneity in Word-Formation Patterns

Table 15.  New -er words (as a result of the collocation test) Besmircher Damager Emailer Empowerer Entruster Escorter Fantasizer Follower-up Funder Glarer

Hollerer Humiliater Leerer Obsesser Opinioner Orienter Outcaster Pimper Prejudicer Pressurer

Proder13 Pummeler Relocater Satisfyer Spiter Subjecter Subsumer Suffocator/-er Tenurer Trick-or-treater

Collocation No collocation

Figure 8.  Collocations versus no collocations with -er words

cf. explanation of pimpee above); others may be simply gaps in the established dictionaries that have yet to be filled. Examples (e): Cases where -er word and -ee word have (usually deliberately) slightly incongruous meanings (e.g. instrument versus person): elevator/elevatee C 238: With the Secret Floor Ballot, each employee in a large office building has a small remote control in his pocket, with which he can signal “3” without the elevator button lighting up. By the time the thing stops at the third floor, he is out the door before he can see the grimaces of the elevatee he left behind. www.halfbakery.com/idea/Secret_20Floor_20Ballot

Of the 748 successful -ee lexemes in the test, 552 items (73.8%) occurred in at least one instance of “lexical solidarity” with an -er word, whereas 196 items (26.2%) produced no such collocations (see Figure 8). 13. Proder and prodee might be taken as a misspelling of prodder/proddee but both versions occurred with equal frequency. The pattern of final consonant doubling in -ee word derivation is surprisingly heterogeneous.



Chapter 5.  A corpus-based analysis of new -ee words 141

100 Frequent

80 60

Established

40

Rare

20

Hapax Legomena

0

Collocations

Figure 9.  Collocations with -er according to frequency categories

Such a high frequency – almost three quarters of all successful -ee words in the test – may suggest that the co-occurrence of an -er word plays a role in the formation of the -ee word. Collocations could proportionately more often be found in items in the Frequent category (112 out of 120, or 94%) than in those categories with a lower frequency rate (Established: 87 out of 97, or 90%; Rare: 211 out of 267, or 79%). The words classified as Not quantifiable produced 41 collocations (out of 50 words = 82%). However, even in the Hapax Legomena list, the proportion of those one or two items which appeared in collocation with an -er word was almost half (101 out of 214, or 47%) (see Figure 9). As we have seen above, collocations often occur even in cases where the -er word is not an established word, as in orienter versus orientee C 511 – after all, the same limitations apply to the usefulness of dictionaries with regard to -er ad-hoc creations and nonce words as they do with our -ee words in question. We might thus state that the existence of a conventional -er word can be advantageous in the formation of an -ee word but its absence is not an obstacle either. Even so, the existence of recognized -er words was tested for various categories of the test results: – Of those 552 successful -ee words which occurred in collocation with a corresponding -er word, the overwhelming majority (93.3%) of -er words are listed in the OED. 37 of the -er words (or 6.7%) found in the collocation test are not listed (cf. example list above). – Of those 196 successful test words which produced no collocation with an -er word, 77.6% display an -er derivation in the OED, whereas 22.4% have no -er correspondence (e.g. *addicter, *dementer, etc.). – Of the 252 “unsuccessful” test words, 70 (27.7%) did not have a corresponding -er word in the dictionary (e.g. *ashamer; *exculpater/or; *overdresser; *rehearer or *untamer).

142 Heterogeneity in Word-Formation Patterns

The proportion of the non-existence of a traditional -er words among the failed words (27.7%) does not seem to differ very much from the successful words which did not produce a collocation (22.4%). In all, however, there were only 10.8% (or 81) among the total of the successful words for which no recognized -er word exists. If we compare this figure with the 27% of the unsuccessful words, we can confirm the hypothesis of the supporting role of a possible correlative opposite.

(ii) Prototypical characteristics In Chapter 2, we have seen that, while the word-formation is diachronically and synchronically ambiguous and incorporates a number of syntactic and semantic features, there are some more “typical” examples of -ee and other words that are on the margins. It could now be assumed that more typical -ee items (like, for instance, interviewee) also have a higher rate of recurrence than those which do not share the most prototypical features (for instance, festschriftee). From the list of successful new -ee words, we will now take a look at those in the category Frequent (= more than 100 attested entries) to analyze their compliance with the characteristics described in Chapter 2: 1. 2. 3. 4. 5. 6. 7.

verb-derived (i.e., verb exists), with existing correlative -er noun, direct object relation to the verb, sentient and probably human, role participant, non-volitional and non-active part in the event, can be used in a legal as well as more general contexts.

Ad (1) Possible verb bases were attested for all (100%) of the new -ee words in the category “frequent,” i.e., there is an established corresponding verb from which a derivation could have been performed.14 However, there are a few cases where the meaning of the verb listed in the OED differs from the meaning of a potential verbal base used in the actual test words, e.g. ‘to cast’ in the new -ee words primarily related to hiring practices in the film or theatre business rather than other attested meanings (e.g. ‘to throw,’ ‘to mould,’ etc.). Ad (2) 112 out of 120 (94%) words have potentially corresponding established -er correlatives. Note that here, the established dictionary (OED) entries were tested rather than the actual -er collocations on the Web (cf. above), even though the figures correspond. The reason for this is that a few new (not OED listed) -er

14. It has to be noted, however, that this is also determined by the fact that verbs formed the basis of derivation for the test words.



Chapter 5.  A corpus-based analysis of new -ee words 143

Table 16.  120 ‘frequent’ new -ee words accusee acquittee adaptee admiree assuree attractee awardee blamee blowee bouncee castee checkee circumcisee claimee confrontee convertee corruptee coveree demotee dependee disciplinee disclosee discriminatee displacee druggee elopee emailee entrustee escortee evadee

evaluatee executee exoneree floatee forgivee forwardee givee greetee gropee harassee hatee healee helpee hurtee infectee influencee inspectee instructee insultee investigatee invokee launchee lendee lickee linkee listee lynchee maskee observee obsessee

paintee persuadee pickee piercee playee pointee pollutee prankee probee protestee punishee raidee ratee recommendee recruitee relocatee remandee removee replacee reportee requestee respondee resurrectee retainee revengee reviewee revokee rulee screwee searchee

servee shaggee shockee slappee smilee spankee speakee stalkee stingee strandee strikee submittee subscribee subsumee suckee supportee taggee teachee temptee textee ticklee timee touchee tracee trackee tradee transmittee tutoree votee wrappee

words (e.g. demoter, obsesser, relocater, subsumer) could also be found in the collocations test. On the other hand, not all co-existing -er words produced an actual collocation in the test, e.g. corrupter, eloper, evader, remover, were found in lexical solidarity with the frequent new -ee words corruptee, elopee, evadee and removee. Ad (3) As has been discussed in Chapter 2, the exact pattern of deverbal derivation according to the classification of Bauer (direct object – DO; indirect object – IO; prepositional object or object with preposition – OP; subject – S) is somewhat tricky and often depends on the meaning of the derivation (cf. discussion on multiple patterns or change of pattern, Chapter 2). Thus, reportee might mean ‘someone to whom something is reported’ (indirect object), ‘someone who

144 Heterogeneity in Word-Formation Patterns

Table 17.  Patterns in ‘frequent’ new -ee words accusee acquittee adaptee admiree assuree attractee awardee blamee blowee bouncee castee checkee circumcisee claimee confrontee convertee corruptee coveree demotee dependee disciplinee disclosee discriminatee displacee drugee elopee emailee entrustee escortee evadee evaluatee executee exoneree floatee forgivee forwardee givee

DO DO OP DO DO OP DO DO ? ? DO DO DO IO/S OP DO DO DO DO OP DO IO OP DO DO S IO IO DO S DO DO DO ?? DO IO ??15

hatee healee helpee hurtee infectee influencee inspectee instructee insultee investigatee invokee launchee lendee lickee linkee listee lynchee maskee observee obsessee paintee persuadee pickee piercee playee pointee pollutee prankee probee protestee punishee raidee ratee recommendee recruitee relocatee remandee

DO DO DO DO DO DO DO DO DO DO DO DO IO DO DO DO DO DO DO OP DO DO OP DO OP OP ?? OP DO IO DO IO DO DO DO DO DO

requestee respondee resurrectee retainee revengee reviewee revokee rulee screwee searchee servee shaggee shockee slappee smilee spankee speakee stalkee stingee strandee strikee submittee subscribee subsumee suckee supportee taggee teachee temptee textee ticklee timee touchee tracee trackee tradee transmittee

IO OP DO DO OP DO DO DO DO DO DO DO DO DO OP DO OP DO DO ? OP IO IO OP DO DO DO DO DO DO DO DO DO DO DO IO IO

15. Givee C 336 is one of my new corpus words. However, a potential rival in the form of giftee was already listed (in Appendix 1). Even though ‘to gift’ is also a verb in Scottish English, there is strong evidence that the formation giftee is derived from the Standard English noun ‘gift’ (see examples below). In a number of cases, an -er derivation ‘gifter’ was coined despite the existence of the established -er verb derivation ‘giver’.



Chapter 5.  A corpus-based analysis of new -ee words 145

Table 17.  (continued) greetee gropee harassee

DO DO DO

removee replacee reportee

DO DO DO

tutoree votee wrappee

DO OP DO

Distribution: DO = 83, IO = 12, OP = 17, S = 3, ?? (Uncertain) = 5.

is reported’ (e.g. to the police, direct object) or ‘someone who reports’ (perhaps in a non-volitional situation, subject). Because of the sheer number of items found for each test word in this ‘frequent’ category, it was not possible to ascertain all possible patterns for each one of the test words. Direct object relation to the verb is the most prototypical pattern. I therefore examined the above words with regard to whether or not they can potentially have been formed in this pattern (regardless of other possibilities). Ad (4) None of the -ee words in the ‘frequent’ category had a purely technical reference, i.e., all of them (100%) can refer to sentient and mostly human entities. However, there are some cases (e.g. dependee, launchee, transmittee) where technical reference is frequent but sentient and human reference is nonetheless possible, i.e., ‘someone who is dependent of something/somebody,’ ‘someone who is launched (i.e., initiated) to something,’ ‘someone to whom something is transmitted (including disease).’ Ad (5) Role participation is also given in 100% of the cases, i.e., all of the new -ee words can be somehow linked to an event. Role participation is all the more likely because of the co-existence of -er words for almost all -ee words in the frequent category (cf. Section 2.6 (iii) on episodic linking and -er words). Ad (6) Most of the words in this category can be assumed to have non-volitional and non-agentive functions in the event in most of their concrete realizations. There are at least two new frequent -ee words which are, in fact, subject formations with arguably the semantic feature “lack of volitional control”: elopee (C 239) and evadee (C 272). Here, I would place the non-volitional character on a par with that of established words like standee or retiree in the sense that Bauer (1983: 290) suggests – that they are “affected by circumstances rather than acting of their own free will.”

(1) Epinions has the best comparison shopping information on Epinions.com – First Know your “Giftee”!, You are the “Gifter”! Now You are ready for gift giving! www.epinions.com/ gift-review-6E20-139E0D30-3A0B6191-prod2 (2) Gift Certificate Recipient (called Giftee) Required fields in Red. USA residents, be sure to type in your City and State […]. www.garlanddrake.com/205/au06CF.htm

146 Heterogeneity in Word-Formation Patterns

elopee C 239: Whenever the police are accused of custodial rape, they immediately set to work. Not to prove that she wasn’t raped. But to prove that she wasn’t nice. To prove that she was a loose woman A prostitute. A divorcee. Or an Elopee – i.e.: She asked for it. (Arundhati Roy on Sheik Kapoor’s Bandit Queen) (www.sawnet.org/books/writing/roy, August 1994) evadee C 272: Regulations prohibited an evadee from flying combat missions again for fear of his revealing, under torture, information on the French resistance if shot down again. […] His stubbortn determination to finish what he started as an evadee and combat pilot never weakened during his distinguished career as test pilot, commander, and – always – fighter pilot. (www.afa.org/magazine/valor/0291valor.asp, February 1991)

An additional possible subject formation where the “non-volitional” feature is not quite as clear is convertee (C 123): convertee C 123: Sounds fascinating! I am a recent convertee to Morrissey and am smitten by his lyrics. However, I think it would be very difficult to construct a linear storyline around them. If it’s not a musical and it hasn’t got a story what the hell is it?? A revue? (www.alisonmoyet.co.uk/cgi-bin/yabb/YaBB.cgi?board=you;action=display; num=1100183271 Jan 2005) I have only so far played it for 5hrs solid and not yet got pass half way in a season but even so I am hooked so much so that my CM 01/02 programme and my saved games has be finally removed from my hard drive. This latest version is not so bewildering for any new convertee to this series of management games as to cause them to lose interest after a short period nor does it fail to provide something new to Old timers like myself. First impressions count and for me at least this is the game that CM4 should have been. (www.pcgamereviews.co.uk/reviewsallamzn.aspx?a=15186&p=2, November 2003)

The question of whether this is a subject formation is, once again, dependent on whether one interprets the word as “someone who converts to sth” or “someone who is converted to sth.” So why not use convert which has, after all, a similar meaning? The (co-)existence of the older noun (first entry: 1561, OED) leads us to suspect that convertee either stresses the non-volitional aspect (as an “addict of X”) or that it marks a widening of the traditional domain of conversion (beyond religion and politics).



Chapter 5.  A corpus-based analysis of new -ee words 147

Ad (7) Quite a few of the new -ee words in this category can be used in legal as well as general contexts (e.g. accusee C 11, aquittee C 14, claimee C 89, confrontee C 111, insultee C 414, investigatee C 420). Others are more specific in the environments in which they might occur. These domains include, for instance, sex (e.g. blowee C 69, screwee C 776, shaggee C 790, spankee C 826), military (recruitee C 673), professional interpersonal relationships as in counselling (harrassee C 354, hurtee C 376).

5.4

On ‘failees’: Ineffective patterns of -ee words

It is always more difficult to establish why words are not formed according to a certain pattern than to explain the ones that were successfully created. Lexical gaps exist and there does not always have to be a specific reason for that. Why was bereavee formed but not bewailee? Why well-wishee but not ill-wishee? In the list below there are some items which could make perfectly reasonable and even prototypical -ee words, such as luree (verb-based, direct object, human reference, episodically linked to the verb, non-volitional character, etc.), but were nonetheless not found. Table 18.  252 unsuccessful -ee test words abhorree abjectee accommodatee accursee acerbatee aggroupee agonizee amendee annectee ashamee batee besmearee bewailee bewitchee buskeree calmee capitalizee caree clappee claspee

disuseee domesticatee eludee enactee endearee enfeeblee engrossee ennoblee eschewee espionagee exacerbatee excitee exculpatee eyee falteree fetteree floutee foretellee frequentee furtheree

misplacee misquotee mistrustee modernizee moralizee mystifyee namee nativizee naturalizee onlookee opportunee ostracizee outsizee outwittee ovationee overbiddee overcompensatee overdressee overnightee overpoweree

prognosee prolongee proscribee proselytee publicizee puffee punnee quackee quailee quarrellee quellee quittee rackee ravee reaffirmee reassertee reassuree recoilee recompensee recomposee

snivellee sparee spicee spiritualizee spongee sputteree startlee stonee strappee surfeitee surpassee swaggeree swankee televisee testifyee tranquilizee treacheree triumphee tyrannee tyrannizee

148 Heterogeneity in Word-Formation Patterns

Table 18.  (continued) coloree combatee concealee consideree contradictee contritee dashee defyee demoralizee denominatee

gawpee gigglee glorifyee groinee hagglee hamperee heedee hidee hissee hospitalizee

overpraisee overrulee overseee overwhelmee perfumee perishee persiflagee personatee pindownee plaitee

recreatee recriminatee recuperatee redressee regardee rehearee rejuvenatee remarryee remissee remorsee

unbelievee uncaree uncomfortee unconfirmee uncoveree underestimatee underminee underscoree understatee undeservee

deprecatee deridee despoilee dickeree disapprovee disavowee disbelievee discommodee discomposee discouragee discreditee disgracee disheartenee disintegratee dispiritee displeasee disploree disprovee disregardee dissociatee distortee

huddlee ill wishee immunizee imploree inclinee ingratiatee insinuatee inspiree jeopardizee laudatee leechee lieee lodgee loiteree lookee luree maledictee mincee misguidee misinterpretee misnamee

plantee plasteree platformee pleadee plightee policee ponderee posee positionee postponee practisee precognizee preconceivee preconsultee predestinee predeterminee predominatee preparee prepossessee presumee

renunciatee reprisee reproachee reprobatee reputee retractee retrogressee revitalizee robbee roughee sailee saturatee securee segregatee sermonee shruggee slobbee slouchee smokee sniggeree

uneducatee unforgivee unobligee unsuspectee untamee usheree valedatee valedictee vengee vexee vindictee vituperatee vociferatee wateree weepee wheedlee wincee workee yammeree yearnee

On the other hand, there are quite a few test words where it immediately becomes clear why they are not very good candidates for -ee words: – It is rather striking that many of the unsuccessful words are based on complex verbs with prefixes, such as dis-, mis-, over-, pre-, re- and un-, three of which (dis-, mis-, and un-) carry a negative meaning. This ties in well with the notion of -ee words as episodically linked: the prefixed verbs in question here often cannot be linked to an event because they point, in fact, to a non-event. This is most evident in cases like *uncomfortee, *uneducatee, and *untamee,



– – – –

5.5

Chapter 5.  A corpus-based analysis of new -ee words 149

where the verbal bases (‘to uncomfort,’ ‘to uneducate,’ ‘to untame’) are, indeed, non-events. However, it is not quite as convincing for most of the verbal bases with a dis- or mis- prefix: after all, potential formations like *discouragee or *misguidee are not linked to a non-event but to a negative event. Besides, there are quite a few successful formations where the verbal base was prefixed with dis- (cf. Appendix 2). Yet, in many of these (e.g. disappointee C 178, discardee C 184, dismemberee C 203, disownee C 206), the meaning of the complex verb (e.g. ‘to disappoint’) has lost its ties to the meaning of the base verb (‘to appoint’) and the status of the prefix is somewhat questionable. Some of the words in the list have other meanings: lookee, workee (cf. explanation in Section 5.2). Some of the words may be too close to an obvious hypocoristic meaning or homophonous adjectives (for example, robbee, smokee). Phonological limitations may also be a reason for the failure of some of the words (for instance, eyee C 295; cf. also Section 2.1). In addition to phonological limitations, orthographic constraints may also apply to some cases (for instance, in overseee C 528).

Hapax Legomena and their production in various text types

A single – and in the context of this study we also including double – occurrence of a word in a corpus like the World Wide Web is particularly interesting because the assumption is that we are dealing with either a very new formation or a nonce word which has not been taken up by anyone else outside the one special occasion when it was formed. It also has to be noted, however, that not everything that has ever been written and posted on the World Wide Web remains there; many entries on websites or whole websites are deleted after a certain period of time. Thus, the limitation of the single (or double) occurrence is connected to the period of research, i.e., February to June 2005, and some of the hapaxes from this period may now have developed to become simply new words with a broader range of users. In this study, a large proportion of the overall test words (214 items or 21.4%) produced Hapax Legomena, making up almost one third (28.5%) of all successful test words (see Table 19). The large proportion of hapaxes among the overall successful words in the corpus might be explained by the fact that there are quite a lot of “informal” or “semi-private” text types in which ad-hoc creations and nonce words might be more readily used than in more formal registers and text genres. Some of them, such as chats, are rather “oral” in character in that they are dialogic and

150 Heterogeneity in Word-Formation Patterns

Table 19.  214 Hapax Legomena abidee abolishee abrogatee acclaimee allegee amazee analyzee animatee anticipatee applaudee astrologee awaitee badgee bellowee bereavee beseechee besmirchee boastee boxee burglaree butcheree caressee cheeree chuckee clubbee clutchee commandee compactee congratulee consortee contaminatee cullee dancee decodee defloweree degradee demeanee demolishee denigratee denouncee depravee despairee

disownee disparagee dispossessee disrobee dreadee emaciatee encouragee enragee (loan) enthralee envyee erodee estrangee eulogizee excusee exercisee exhumee facee fancyee fantasizee fencee fetchee finee forgee forsakee frownee fumblee fussee glancee grippee gullee hailee hearee hobnobee housee hoveree huffee hushee imaginee immunee impairee incarnatee inconveniencee

maimee manicuree masteree meddlee medicatee mimicee mimickee misdirectee navigatee neckee omittee opposee outcastee outplacee overridee overthrowee pacifyee panicee pedicuree peepee peltee penalizee perfectionee perplexee persistee placatee polarizee postulatee prejudicee preselectee preventee proclaimee procuree professee prohibitee promptee pronouncee propagatee proposee puzzlee quenchee quibblee

repatriatee reprievee respectee retortee retracee reveree reversee revilee ridee rinsee riotee rowee ruinee rumouree satisfyee savee scoutee scowlee screechee scrubbee sedatee self-deprecatee self-sacrificee self-servee sequencee sketchee sledgee snarlee sobbee soothee spuree squealee squelchee stakee sterilizee stinkee subordinee suffocatee surmountee tamee terrorizee therapeutee



Chapter 5.  A corpus-based analysis of new -ee words 151

Table 19.  (continued) detectee detestee detractee devastatee disbandee dishonouree disinheritee dislikee dislodgee dismemberee disobligee

incriminatee indebtee jeeree kiddee lamentee leeree liaisee lockee lubricatee lurkee lullee

rebootee rebuffee reclaimee refinee refusee regeneratee rehearsee reincarnatee rejoicee rekindlee rememberee remonstratee

thrillee toutee traitee unauthorizee unitee upliftee vanishee wageree wailee well-wishee wonderee yodellee

frequently imitate speech by using highly colloquial vocabulary, oral features of syntax such as half-finished sentences, additive rather than analytic structures, etc. (cf. Oesterreicher 1999 for a discussion of ad-hoc formations and parameters of oral versus written language; Ong 1982 for features of orality in general; Rehm 2002 for orality on the Web). On the other hand, the diversity of different text types is enormous and ranges from the most informal chat/netforum via semi-private and public weblogs, professional online newspapers and journals (or online versions of printed papers and journals), advertising sites, business reports/minutes of business meetings, to legal documents/regulations and academic/scientific publications (cf. also Rehm 2004 on classification of hypertexts on the Web).16 Furthermore, not everything which appears on the Web is recent material: there are by now many archives of historical documents and literature on the Web, from some of which the results are also automatically taken. Besides the alleged “newness” factor of hapax legomena, it will be just as interesting to see where they have been created. In order to gain a better picture of the types of texts and contexts, the list of hapaxes has been analyzed according to the following criteria: 16. So far, search machines do not specify text types, a deficit Rehm (2004: 121) seeks convey in his project HYPNOTIC (“Hypertexts and their Organisation into a Taxonomy by means of Intelligent Classification”). “Robuste Methoden zur automatischen Bestimmung von Hypertextsorten ermöglichten eine völlig neue Funktionalität für Suchmaschinen, indem der Benutzer die Möglichkeit erhielte, neben verschiedenen Stichwörtern auch die gewünschte(n) Hypertextsorte(n) der zu findenden Dokumente zu spezifizieren, beispielsweise “Texttechnologie” und “XSLT” in den Hypertextsorten persönliche Homepage eines Wissenschaftlers und Wissenschaftlicher Artikel.”

152 Heterogeneity in Word-Formation Patterns

i. ii. iii. iv. v. vi.

Text type Context/domain Collocation with -er Web domain Date of entry (if given) Specifics

Ad (i): Text types on the Web – hypertexts – are a highly interesting subject which has recently garnered attention in a number of publications (cf. Asirvatham & Kumar Ravi 2001; Brandl 2002; Rehm 2004). Despite some heterogeneity, the characterization and labeling of text types and genres provide a useful framework for exploring the contexts in which ad-hoc creations of new words might occur. For the purpose of my study, I have selected seven categories, which differ in mode of interaction (category 1, for instance, is always interactional and dialogic), range of intended audience, level of formality and subject matter: 1. 2. 3. 4. 5. 6. 7.

chat/netforum/message boards 80 weblogs, private websites, fictional or factive report 66 newspaper report/ review (e.g. book/film review) 51 advertising website   3 business report/minutes on business meetings   2 legal documents/regulations 19 academic paper/scientific or technical report 25

Not unexpectedly, the text type in which such ad-hoc creations were formed most frequently was category one – chats, forums or message boards – where users exchange their views and information on any given topic. A total of 80 out of 246 entries (this figure is higher than the 214 new -ee lexemes in my hapax legomena list because some lexemes occur as double entries) appeared in participatory and interactional texts like in chats, in net forums and message boards (category 1). A typical example of an ad hoc -ee (C 845 squelchee) word creation in this text type can be found in the following discussion (Topic: “Do low class Southerners lack intelligence”): Text type 1: […] Writer A: Quite the opposite, really. Everyone is born with an intellectual curiosity, but all people, from eventual geniuses to dimwits, have to build their intelligence from interaction with the world around them. If their environment squelches their desire to learn, then this is where willful idiocy begins. In this case, you could make an argument that more environments in the



Chapter 5.  A corpus-based analysis of new -ee words 153

South kill intellectual curiosity than others, but it’s still unfair to make that generalization. Writer B: Sounds as much a paragraph about knowledge as about intelligence. But, anyway, if we get into a nature/nurture argument we’ll be here forever with no gain. And being “squelched” wouldn’t be very willful on the part of the squelchee – maybe the squelcher? (www.rottentomatoes.com/vine/showthread.php?t=225073&page, December 2003)

Text type category 2, i.e., private or semi-private websites, weblogs, diaries and other types of narratives, saw the second largest number of -ee words (66 items), followed by text type category 3 (51 items), i.e., professional reports, for instance, in online newspapers. The following examples will illustrate the environments in which -ee words such as scowlee (C 772), guest housee (C 370) or hushee (C 377) were created: Text type 2: Tuesday 2nd November I wasn’t quite as grumpy today because I had a bit of special “me” time when Peter went to school. I spent the time perfecting my two new looks. One is “the scowl” – this means tilting the head down but keeping the eyes pointing at the scowlee. I use it for people I’ve only just met and therefore don’t trust and need to fend off. I’m really good at it now. The second look is my cheeky skywards glance which I use when I’m pretending I haven’t noticed Dad. Do do this look I tilt my head upwards (similar to the scowl) but this time I look up too. I allow a little smile to play around my mouth before I do a really quick scowl and then go back to the cheeky look. (www.babyblog.co.uk/04_november_ajm.htm-29k, November 2004) Text type 3: We managed to climb through the guest house doorway that was blocked by tables and chairs and checked in Nathan’s room (a fellow guest housee) as we went by. It was quite scary as his room was a terrible state and at a first glance it was difficult to make out whether he was in there or not – he wasn’t (www.bbc.co.uk/suffolk/content/articles/2005/02/21/nick_rowland_ february_update_f, February 2005) Especially now that the leader of the real-live “Rambos” hauled into an Afghan court insists that he has been in regular contact, “five times a day, every day,” with the office of none other than Donald Rumsfeld himself, and that he has tapes to prove it.

154 Heterogeneity in Word-Formation Patterns

Apparently the Pentagon’s hush-hush apparatus is not hushy enough when your hushee is left twisting in the wind. (www.pww.org/article/articleview/5585/1/148/ - 38k, September 2004)

The other text types did not produce -ee nonce words to the same extent. Category 7 – scientific or technical report, academic paper or abstract – was the most prolific among the remaining four types and 25 examples were found in this classification: applaudee C 38, badgee C 54, commandee C 102, decodee C 143, denouncee C 158, devastatee C 173, erodee C 266, exercisee C 281, facee C 296, fetchee C 306, grippee C 346, hoveree C 371, lamentee C 434, meddlee C 465, prejudicee C 605, prohibitee C 625, propagatee C 630, quenchee C 649, refinee C 677, refusee C 679, repatriatee C 705, respectee C 722, sedatee C 782, sequencee C 787, therapeutee C 904. Text type 7: While women are the instruments of the keen and therefore the holders and keepers of memory, Airt Uí Laoghaire is nevertheless the subject of the lament. As is the Caoineadh, The Lament for Arthur Cleary is the tale of one man whose eighteenth-century doppelgänger makes him an everyman. In Bolger’s poem, however, and even more so in his play, the lamentee is given a voice; Arthur Cleary voices his own particular plight which represents in a more general way that of a returned working class emigrant. (www.lib.unb.ca/Texts/ JSW/ number25/ grubisic. html, July 2003) To illustrate the problem with closure replacement, here is the summary of the fetcher protocol: Fetcher Fetchee

--------- Come here! ---------> <----- I challenge you! ---------- I accept the challenge ----> <--------- Coming… ----------

Consider an implementation of the fetcher protocol, where the fetcher uses operations to send the “Come here!” and the “I accept the challenge” messages and the fetchee uses an operation to send the “I challenge you!” message. Even if the fetchee correctly copies the closure from the “Come here!” message into the “I challenge you!” message, its use of an operation results in the closure being replaced and fetcher’s operation never seing an appropriate closure. (www.cs.nyu.edu/rgrimm/one.world/doc/one/world/util/Operation. Chaining Closure.html, date not given)



Chapter 5.  A corpus-based analysis of new -ee words 155

Legal texts or regulations, the original domain of -ee words, are responsible for 19 -ee words in the hapax legomena list: abolishee C 4, compactee C 105, consortee C 117, disownee C 206, finee (2 x) C 308, forgee C 318, grippee C 346, liaisee C 441, lockee C 448, opposee C 508, outplacee C 515, preselectee C 608, procuree C 618, reclaimee C 664, reprievee C 711, subordinee C 868, unauthorizee C 938 and unitee C 952. Text type 6: NORQUEST COLLEGE FACULTY ASSOCIATION BYLAWS 8.2.2 The Faculty Association Grievance Committee (FAGC) […] will offer support to an abolishee by accompanying the member to a meeting with Human Resources to discuss employee reduction transitional support services as may be stated in college policy. (www.norquest.ab.ca/pdf/faculty/bylaws.pdf, May 2003) A reprieve, unlike a parole, is a sentence suspension, but the Board usually credits reprieve time to an inmate’s sentence if he obeys all reprieve conditions. However, a reprievee returned to prison as a reprieve violator may receive no credit on his sentence for any time spent on reprieve. (www. pap.state.ga.us/other_forms_clemency.htm)

The moderately high proportion of hapaxes in text types reports, regulations and scientific papers (text types 3, 7, and 6) demonstrates that these new formations can not all be attributed to ad-hoc creations in an interactional format. One can also assume a relatively high level of linguistic competence in such specialized registers. The fear that a Web search for neologisms might produce a high number of formations that were created by non-native speakers/writers with a very limited competence in English seems to be unjustified. Category 4 and 5 – advertising and business texts – are negligible, with only three and two items respectively. It is worth noting that in the cases of a double entry, the text genres in which they appeared are often very different: thus, fetchee appears in a private weblog (2) as well as in a scientific article (7), similar to prejudicee (1 & 7), procuree (1 & 6), proposee (5 & 7) and refusee (2 & 7). proposee C 631 Text type 5: Bridal Showcase 2003 Keep the proposee in mind. Planning a proposal he or she will love is all about tailoring your approach to what will surprise and delight the person … (www.theeagle.com/announcements/ bridalshowcase/2003/27.htm)

156 Heterogeneity in Word-Formation Patterns

Text type 7: Topography of a GIS Proposal The listener (proposee) interacts with the proposer in a communication process … The proposer approaches the proposee’s office with project in hand (www.vectorone.info/prop1.htm)

Ad (ii): The context or subject domain in which the entries occurred are, necessarily, rather diverse. Yet, as we have seen in the development of -ee words, domain restriction was for a long time a characteristic feature of this suffix and it was considered interesting to see in what context new -ee words have predominantly emerged. Therefore, eleven different subject domains have been selected for analysis of hapaxes: a. b. c. d. e. f. g. h. i. j. k.

private – interpersonal (incl. fictional accounts, literature) 113 media   37 religion   10 professional – interpersonal (incl. self-help groups, etc.)   22 legal    7 education   10 science   17 sports   11 business    4 politics   13 military    1

As can be seen in the figures above, the largest number of subject domains by far was private-interpersonal matters, followed by media (books, films, etc.) as a subject, followed by professional-interpersonal matters (including counseling, selfhelp groups, etc.) and scientific subjects. Sports, education and religion as topics were in the middle to lower region; legal, business and military matters were the least found in hapaxes. Ad (iii): The collocation of -ee words with -er words was already described in Section 5.3, including the percentage (47%) of collocations occurring in the hapax legomena list. Ad (iv): Unfortunately, Web domains (such as ‘.com,’ ‘.org,’ ‘.uk,’ ‘.nz,’ ‘.edu,’ etc.) do not always point to the origin of the website. Yet, in some cases they can provide clues as to the affiliation of the writer/creator of the -ee word, i.e., either institutional (e.g. ‘.ac.uk’ or ‘.edu’) or national (e.g. ‘.uk,’ ‘.au,’ ‘.de’). Information on this was collected. The vast majority was .com, followed by .org and .edu. However, no conclusive results could be gained from this classification. In Chapter 6,



Chapter 5.  A corpus-based analysis of new -ee words 157

a more detailed search for (attested) -ee words according to various geographic environments will be conducted. Ad (v): Equally, the date of writing or publishing was collected if provided on the website. As mentioned above, most of the examples in my Web research are from recent text material but some cases are also historical: The earliest date was found in a legal text from 1643: procuree17 (C 618) published in House of Lords Journal, Volume 5: 21, January 1643: Blake granted One Hundred Pounds per Annum out of the same Office, he had made a former Assignment thereof as aforesaid, Rolfe being a party to the one and yet a Witness to and Procuree of the other; and the said Rolfe also, pretending to increase and amend the Petitioner’s Jointure, assigned to the Petitioner’s Friends some of the same Lands which were formerly conveyed her, whereby it appeared he had been a Means to deceive her at first; and the said Rolfe further assigned to the Petitioner’s Use a House in Fleetestreete, formerly mortgaged by himself. (www.british-history.ac.uk/ report.asp)

Among the “historical archive” words is another seventeenth-century find (all are, of course, not in the OED): disobligee (C 205) occurs in a fictional text from 1673, “The gentleman dancing master” by William Wycherly. The historical examples18 in the hapax legomena list include

1643 1673 1839 1890 1899 1917 1930 1930 1968 1977

procuree (legal) disobligee (fictional, William Wycherly) revilee wageree (fictional, Australian poet Barcroft Boake) enragee (fictional, Kate Chopin) disinheritee (fiction) neckee (media) unitee (legal) squealee (fictional, Frank Zappa) unauthorizee

17. Note that there is also a contemporary entry (March 2004) for procuree in my hapax legomena list (it is thus a double entry h.l.): “On Sunday we decided this was our best option for procuring beer and I was dragged along to be the procuree as the role(s) of driver and amused passengers had already been filled.” valerieledbetter.blogspot.com/2004_03_ 01_valerieledbetter_archive.html – Note the agent meaning and the lack of volition in this example. 18. Here, “Historical examples” is also meant to include twentieth-century examples which were produced before the 1990s, i.e., the time when the internet became a popular medium.

158 Heterogeneity in Word-Formation Patterns

1980 preventee 1985 terrorizee 1985 erodee

While these twelve “pre-1990s” instances are the exception – the vast majority of all words found in the hapax legomena list date from the end of the 1990s and early 2000s – they nevertheless illustrate that a Web search can also be a helpful means in historical research. It is interesting to consider whether these words might be classified as nonce words, given the time lapse between the production and the research and the fact that they are still less frequently used, yet, established words. For seventeenth-century legal and literary language, however, such a conclusion would be volatile, since we can not assume an ad-hoc creation in the legal context; furthermore, a lot of the contemporary written material might be lost, so that other examples of usage might have existed. Enragée (C 257), marked graphically as a loan word with French spelling, curiously occurs in Kate Chopin’s 1899 “Lilacs” from A Re-Awakening, as well as one hundred years later, in a 1999 film review in the newspaper The Nation: (1) It seems to me they are richer than ever this year, Sister Agathe. And do you know, I became like an enragée; nothing could have kept me back. I do not remember now where I was going; but I turned and retraced my steps homeward in a perfect fever of agitation: ‘Sophie! My little trunk – quick – the black one! A mere handful of clothes! I am going away. Don’t ask me any questions. I shall be back in a fortnight.’ And every year since then it is the same. At the very first whiff of a lilac blossom, I am gone! There is no holding me back. (Kate Chopin, A Re-Awakening. Lilacs. Electronic Library) (2) Bonier than Isa, more conventionally pretty but so pale that she looks translucent, Marie is your basic 20-year-old enragée. When she meets a luxury car, her first response is to kick in its taillights. (Film review, the nation, March 1999) (www.thenation.com/doc/19990322/klawans-30k)

Similarly, neckee (C 494) appears both in a 1930s humorous university newspaper for Engineers (“The Tech-Engineering News,” November 1930): […] asked the neckee. “Isn’t he healthy?” – Virginia Reel. “I’ll have you know-hic-hic-hic-that I’m part of the Standard Oil Company.” “What part are you?” (web.mit.edu/voodoo/www/archive/pdfs/1930-Nov.pdf)

and in a contemporary (2002) news forum:



Chapter 5.  A corpus-based analysis of new -ee words 159

Hey, Jody what’re you talking about? I stopped necking and jumped off the couch to get my head in the floor model radio’s speaker when Joaquin took his ride on “Just A Pair of Blue Eyes” On Tex Williams record the first time I heard it. (my neckee at the time had a pair of blue eyes & the goldest hair!) I thought he was great with that western swing on the steel! Does Herb play like that? Wow! I’m glad he’s keepin’ that tradition alive! Western swing has always been some of the finest American music ever played! (steelguitarforum.com/Forum15/HTML/001730.html)

It is questionable whether the latter creations follow the earlier ones. Rather, we seem to have the phenomenon “that the same form should be coined by different speakers, either at different times, or very close together in time” (Bauer 1983: 45) – an incident which does not affect the item’s status as a nonce formation. Ad (vi): Finally, comments on syntactic or semantic specifics of the word-formation (agent type, non-human character, ambiguity) are included in the analysis, as are speaker comments on the term itself. One of the qualities of a nonce word, according to Bauer (1983: 45), is that speakers are aware of using a term which they have not already heard. Often speakers express insecurity about the status of the word by either presenting it in quotation marks or by making explicit comments, such as adding, “is that a word?” in parentheses. In my hapax legomena list, 14 out of the 246 words (including the double entries) are written in quotation marks (amazee C 27, analyzee C 30, anticipatee C 35, burglaree C 76, cheeree C 86, detractee C 172, devastatee C 173, disinheritee C 199, erodee C 266, hoveree C 371, manicuree C 462, masteree C 464, panicee C 536, rekindlee C 689), and two are commented on directly (envyee C 265, incriminatee C 395). Other specifics include subject patternings, as in the following examples where the disrobee (C 216) is actually ‘someone who disrobes’ and the yodellee (C 1000) is essentially the yodeller: If child molestation has been claimed because of an adult having disrobed in the presence of a minor, and the minor was never touched, spoken to, or otherwise directly engaged by the disrobee, then Sue the Child’s Parent and/or Legal Guardian for “Parental Negligence” (“consideration of the law,” personal website, Oct. 2004) I am a 19 year old yodellee. However i refuse to conform to the constraints of the traditional ‘del…i’m more of a street style ghetto freestyler … (pub33.bravenet.com/guestbook/show.php, date not given)

Altogether, 14 items in this hapax legomena list have an agent function, i.e., 6% of the 246 examples (abrogatee, acclaimee, awaitee, disrobee, lurkee, reclaimee,

160 Heterogeneity in Word-Formation Patterns

rehearsee, self-deprecatee, self-sacrificee, self-servee, sobbee, vanishee, wonderee, yodellee). In the case of those formations which are prefixed with self-, the very meaning does not permit a formation with an object but, instead, clearly points to a subject formation. Thirteen of the hapax legomena -ee words have non-human references (abrogatee, commandee, decodee, erodee, facee, fancyee, fetchee 2 x, hoveree, propagatee, rebootee, [fellow] reincarnatee, tamee), either referring to animals (e.g. the second C 306 fetchee entry – dog or C 371 hoveree – butterfly, C 893 tamee – lion) or technical components (e.g. C 630 propagatee, C 266 erodee): (1) Rico knows the names of more than 200 toys, in English and German, which suggests he is not only precocious but bilingual. Rico’s owner has tried to teach similar tricks to her other pooch, a Rhodesian ridgeback who is far closer to normal. He hasn’t got a clue. That’s a relief. As for Rico, you have to wonder how long it will be before he starts thinking very seriously about who’s the fetcher and who’s the fetchee. (June 2004) (www.fortwayne.com/mld/newssentinel/news/ editorial/ 8926919.htm) (2) In your data, record whether the individual was just sitting (S), visiting flowers or artificial nectar (N), laying eggs (E), or hovering above other butterflies or being hovered over (H1 = hovering; H2 = “hoveree”) or attending pupae (H. charitonia is a pupal mating butterfly!) (www.sbs.utexas.edu/bio373l/docs/ butterflymrr/MRR%20handout%20Sp05 %20Rev1, May 2005) (3) Finally we listen to a conversation between a lion tamer (Max Moore) and his tamee. It’s a familiar love story, really (you need me – you’ll be sorry if you try to get rid of me!), only spectacular because the tamer speaks it from inside the animal’s belly. Such is Chapman’s excellent craftsmanship at getting us to suspend our disbelief in just about anything. (www.offoffoff.com/theater/2002/bigtop.php3, June 2002) (4) The additional propagatee parameter on the stack specifies the id of the propagatee thread. If propagatee and current sender or propagatee and receiver belong to the sametask [sic], propagation is always permitted. (l4hq.org/docs/manuals/l4-86-x0.pdf, September 1999) (5) The continuous removal of a solid ‘erodee’ is effected by gradual consumption of a solid ‘eroder.’ In the process under study, the erodee melts and the (liquid) melt diffuses to the eroder, where an exothermic, indefinitely rapid, diffusion-controlled surface reaction occurs. (adsabs.harvard.edu/abs/1985aiaa.meet, 1985)



Chapter 5.  A corpus-based analysis of new -ee words 161

Quite a few of the hapaxes (15) are formed as part of a compound or fixed collocation:

“[…] As a first time Bonaire “anticipatee” […]” C 35; “The lap dancee accused the lap dancer […]” C 136; “Word to that, fellow despairee” C 167; “[…] the title emaciatee is played by a popular Glasgow disc jockey Flash” C 241; “[…] (a fellow guest housee)” C 370; “That always annoys the diplomatic immunee” C 384; “[…] a fellow auto-impairee” C 386; “[…] my nine sprays from the 110st truth omittee/” C 504; “My fellow rehearsee also got a slap […]” C 683; “[…] a fellow reincarnatee and cat of questionable character” C 684; “A woman repatriatee told HRW/Africa […]” C 705; “[…] from showing the beam retracee” C 730; “One blue rinsee with a budgerigar that never ceased to talk” C 749; “An open letter from party ruiner to party ruinee” C 755, and “Fat joins fellow rumouree Keith Richards” C 758.

It is worth noting that the modification with “fellow” was chosen rather frequently, also in cases where the agent/patient status of the referent is not quite clear, as in “fellow rehearsee” or “fellow despairee.” This is yet another indicator of the role of the semantic feature “non-volitional” in -ee formation: in the apparently reluctant situation, the common fate is thus emphasized. In addition, some of the other compound formations are not quite clear in terms of their agency or possible hypocoristic meaning, for instance, anticipatee or rinse; others evidently have non-volitional and patient meaning, such as dancee (in collocation with dancer) or repatriatee. Omittee is certainly a rather special pun in a lyrical format (part of a rap song) and its meaning is meant to be ambiguous. In this experimental field of nonce words, an ambiguity of meaning is rather frequent. C 35 anticipatee, C 61 bereavee, C 167 despairee, C 230 dreadee, C 270 estrangee, C 283 exhumee, C 464 masteree, C 508 opposee, C 549 perplexee, C 605 prejudicee, C 618 procuree, C 683 rehearsee, C 729 retortee, C 746 ridee, C 749 rinsee, C 775 screechee, C 777 scrubbee, C 868 subordinee were all found to have a somewhat unclear meaning, such as in the following examples: (1) I hope Miss Skinner verified the good opinion I so rashly expressed of her, but I shall never know. The next scrubbee was a nice looking lad, with a curly brown mane, and a budding trace of gingerbread over the lip, which he

162 Heterogeneity in Word-Formation Patterns

called his beard, and defended stoutly, when the barber jocosely suggested its immolation. (www.etext.lib.virginia.edu/.../modeng&data=/texts/english, February 2001) (2) […] People, the rider and the ridee, wears unproper helmets. […] To all the motorbike riders and ridee out there, please be careful. (lemontee3.tripod.com/2004_05_01_archive.html, May 2004)

In the first example, the choice of an -er agent form could have been prevented by the existence of scrubber as an instrument – this would be a case of blocking. Blocking is also a likely reason for preventing *awaiter, as it is too similar to waiter. At least one of the formations is formed not by the general -ee formation rule, but by analogy, i.e., the new formation is clearly modelled on one already existing lexeme (cf. Bauer 1983: 96): dismemberee is created in opposition to rememberee. If we compare the features of -ee words in the hapax legomena list with those in the ‘frequent’ category it becomes clear that the syntactic and semantic diversity of these nonce words is far greater than that of the much more frequent items: Among the hapax legomena, there are some of those examples which are without doubt noun-derived (astrologee, professee, therapeutee); many show a deliberate or accidental ambiguity of meaning, with quite a few having non-human references or an agent meaning. In quite a few cases, the speakers show insecurity about the formation or playfully undermine the more established patterns of -ee formation and its interpretation. We can conclude that, indeed, those neologisms that comply with prototypical features of -ee words19 are also the ones which become more easily established and widespread.

5.6

New productivity? On data collection and productivity measurement

The overall result of the search of new -ee words on the Web raises a number of questions. Is the -ee formation pattern much more productive than previously assumed? How does this evidence of productivity relate to the medium of analysis? If we only consider the first question, the rather striking results seem evidence enough to assume a growth rate of -ee formations which goes far beyond earlier postulations: not only is the ratio between potential words (my list of previously unattested search words, formed largely as deverbal -ee nouns: 1,000 words) and 19. The fact that the words in the ‘frequent’ category comply largely with prototypical features of -ee formation patterns does not mean that individual formations might deviate from it (cf. example of reportee above).



Chapter 5.  A corpus-based analysis of new -ee words 163

new -ee words found on the Web (748 words) remarkable and exceeds all expectations. But the relationship between previously attested words (in my count: 511 words) and new words (748 words) is also significant if we consider that the approximately 500 listed words were attested in a period of six centuries and the new words – one and a half times as much – are mostly very recent and contemporary formations, collected in a short period of time. The problem is, however, that we do not know whether similar figures might also have been found earlier had a data source the size and versatility of the Web been available. To some extent, productivity measurements have always to be seen in relation to the data source – which is why absolute quantitative growth rates have to be taken with considerable caution. Commonly, the productivity of an affixation process is compared to the growth rate of another affix in the same data source, a procedure which was not within the scope of the present study. Nor were the occurrences of new -ee words compared to the occurrences of already attested -ee words on the Web. Therefore, a conclusive result of the quantitative productivity of the -ee formation pattern was not intended and cannot be given in this empirical study. Rather, the procedure of investigating productivity here was based on the ratio of actual to potential words (see also Schröder & Mühleisen in press). The results of the present investigation illustrate a number of points worthy of note, both about the nature of this word-formation pattern and about the methods of collecting data to describe the pattern: – The sheer size of the new data on -ee words and the ratio between potential words and actually realised words leads us to cautiously consider a higher rate of productivity for the word-formation pattern than previously assumed. – The analysis of the new data provides evidence about the dominant syntactic and semantic patterns in -ee word-formations. The role of collocations with -er words in the production of -ee words, for instance, leads to question the previously assumed unique role of the verbal base in -ee formation. It also shows the distribution of prototypical semantic and syntactic features in the various categories of frequency of occurrence in the data. – In a comparison with earlier patterns of -ee formation (Chapters 3 and 4) it shows both the continuities and processes of change in the new and recent -ee formations. With regard to the semantic domains in which the most frequently occurring new -ee words appear, we can note additional fields (e.g. professional interpersonal relations, sex, military usage) to the already established domains of usage (e.g. legal, interpersonal, humorous). Furthermore, we can observe a growing diversity and ambiguity of meaning especially in those words classified as hapaxes. Here, the close analysis of the context in which

164 Heterogeneity in Word-Formation Patterns

–

– – –

these words are formed sheds light on the processes of creating neologisms and on the role of analogical coining in the formation of new -ee words. By classifying the new words according to their frequency of occurrence, we can establish differences between more creative nonce words and more prototypical frequent words, offering conclusions on routes of establishment. The empirical data given here also offers a basis for future research on the career of now nonce words. It offers insights on the context-based production of nonce words and the role of “oral” or dialogic characteristics in ad-hoc creations. It critically reviews the relationship between data sources, i.e., listed words (in dictionaries) and actual words (actual usage). It explores the potentials of using the Web as a corpus.

Furthermore, this empirical study also offers new perspectives for future research, such as a follow-up research of the hapax list (to see which ones will remain nonce words, and which ones become established) or a comparison with another process of affixation. In Chapter 6, some of the established and new -ee words will be considered in their geographical distribution, i.e., whether or not they are especially prevalent in particular varieties of English. For this purpose, websites from the U.S.A, the U.K., Australia, New Zealand and India will be analysed according to particular -ee occurrences. A discussion of local versus global uses of -ee words will finally be included.

chapter 6

-ee words in varieties of English

The history of -ee words is ultimately connected to the history of the English language with all its developments, changes and diversifications. We have seen in the diachronic account in Chapter 3 that by the 1760s and 1770s, -ee words specific to American usage seem to have emerged. Thus, separate first entries for words of U.S. origin, such as petitionee or conferee, can be found in the OED (cf. also discussion of regional vocabulary entries into the OED in Price 2003). By the eighteenth century, the spread of English as a world language had begun, due to processes of colonial expansion, migration, and settlement, as well as to the rise in the political and economic significance of Britain (cf. Bailey 1992; Crystal 1997 for a description of the globalization of Englishes). While French was still the major language of international communication, the firm establishment of the English language in the United States ensured its stability and growth. In 1780, John Adams made his proposal to Congress for an American Academy: English is destined to be in the next and succeeding centuries more generally the language of the world than Latin was in the last or French is in the present age. The reason for this is obvious, because the increasing population in America, and their universal connection and correspondence with all nations will, aided by the influence of England in the world, whether great or small, force their language into general use, in spite of all the obstacles that may be thrown in their way, if (as quoted in Crystal 1997: 66) any such there should be. 

The political independence of the United States brought the first and conscious language-political separation of a variety of English distinct from the British motherland and mother tongue. But Noah Webster’s dictionaries and the language-political choices established there (e.g. distinct spelling system) only made visible the development of language divergence in the lexicon and pronunciation which had already begun by the time the first English settlers arrived in the New World. While American English clearly became the most important result of the geographical dispersal of the English language, other distinct varieties of English also started to emerge in different parts of the British Empire. As Horvath (1985: 26–27) points out with regard to the historical reconstruction of varieties

166 Heterogeneity in Word-Formation Patterns

of English, “there is a period after the standardization of spelling and before the beginnings of the scientific observation of dialect variation during which it is difficult to reconstruct varieties of English. It is during just this period that several major regional dialects of English arose, including American English and Australian English.” By the nineteenth century, the status and future of the English language as a world language was hardly debated and its legitimacy in this role barely questioned. In his Images of English, Richard W. Bailey cites a number of nineteenthcentury sources which illustrate the attitudes towards the growing influence of English around the world: And this, in our tongue, is a very hopeful feature not observable in any other language that has fallen within our observation. Wherever it may go, the force of our institutions, our character, our literature, and policy accompany it; the vigor of the race that uses it, almost as surely triumphs over all opposing obstacles as do their arms over all opposing nations […] The language of the seas is already our own. Nine tenth of the commerce of the ocean is transacted through the copious and flexible medium of our tongue, and claims the protection of the Anglo-American fraternity. The barbarism of Australia, the heathen institutions and worn-out languages of India, the superannuated hieroglyphs of China, and the rude utterances of important parts of Africa and of numberless islands in the Eastern seas, are fast giving way to the institutions and the languages of our race. But the great field for its most splendid and extensive development, we believe, must be looked for in our own youthful and magnificent republic [i.e., the United States], and the supremacy she is yet destined to exercise over the whole of this (“Our Language is Destined to be Universal” 1855: 311, Western world.   as cited in Bailey 1992: 109)

As a consequence of the further establishment of different Englishes with distinct norms of usage in different parts of the world, new -ee words with a separate geographical attribution can be detected in the OED for the nineteenth century, for instance, the Australian expiree (1802) – “one whose term of punishment has expired; an ex-convict” and the Scots geggee (1855) – “based on the Scots English noun gegg: ‘hoax, trick, practical joke’ and the verb to gegg: ‘to hoax, play a trick on’” (OED). Their inclusion into the OED illustrates, particularly with regard to the first example, that some new words from the fringes and (then) colonies were noticed and acknowledged beyond the specific localities of their usage. In Jennie Price’s (2003) account of the gradual inclusion of varieties of English to the OED and the preparation of a New English Dictionary (NED), she cites a plea to the readers and contributors from 1879 to also include citations from the colonies:



Chapter 6.  -ee words in varieties of English 167

In order that [the Dictionary’s] progress may be certain, and that it may have that complete and representative character which has been its aim from the beginning, and be a lasting monument of our language, the Committee want help from the readers in Great Britain, America, and the British Colonies, to finish the volunteer work so enthusiastically commenced twenty years ago, by reading and extracting the books which still remain unexamined […] American and Colonial readers we ask, besides sharing the general work, to read for us those recent books which show the additions made to English in their respective countries, as received names for physical features, productions, &c.  (as cited in Price 2003: 120)

Despite this gradual recognition of regionalities, many newly coined words in separate varieties of English might not be found in the dictionary, as their inclusion depended entirely on the awareness of individual readers and contributors. As already commented on in the discussion of neologisms in the twentieth century (Section 4.5), derivations are less likely to be noted as neologisms than other types of word-formations (such as compounds, blends, eponyms, etc.). With regard to -ee formations in varieties of English, I will distinguish two different phenomena. On the one hand, there are some words that were only coined in a particular variety of English, usually for a specific localized purpose. The aforementioned expiree is one example of this: the term relates to the historical role of Australia as a penal colony where British convicts were sent. On the other hand, there are also various formations which have existed in British – and later, international – English for some time but have taken on a different meaning in a specific variety. Assignee is one such word, which has been in English usage since the early fifteenth century with the meaning “one who is appointed to act for another; a deputy, agent or representative” (OED 1419). In Australian English, its denotation changed to “a convict assigned as unpaid servant to a colonial settler” (OED 1843). The meaning of deportee in various geographical and historical contexts is more divergent still: its first denotation seems to have emerged in a colonial context: One who is or has been deported; spec. in Indian use, = DÉTENU (OED). 1895 Westm. Gaz. 13 Dec. 5/1 One party of fifteen deportees from Constantinople having been massacred. 1909 J. MORLEY Recoll. (1917) II. 309 The failure to tell the deportee what he is arrested for. […] 1959 Times Lit.

. An Appeal to the English-speaking and English-reading Public to Read Books and Make Extracts for the Philological Society’s New English Dictionary (Philological Society, 1879).

168 Heterogeneity in Word-Formation Patterns

Suppl. 6 Mar. 126/3 In that sparsely populated, primitive and depressed region … the deportees were more or less abandoned to fend for themselves.

More than a century after the first entry of deportee in this sense, Jean Aitchison (1999: n.p.) notes a change in the meaning of this item in British usage: as she argues here with reference to recent use of the word in the media, the word has now taken on the meaning of refugee: But why deportees? To a native speaker of English, anyone deported is (usually) someone put back where they originally came from, perhaps because they are illegal immigrants, or because they have done wrong, as when British football hooligans were deported from Holland. Yet this is the label which the exiled Kosovans desire, it seems, even though most of them have lived in Kosovo all their lives, and have committed no crime.

For Jamaicans, however, a deportee has a completely different meaning (cf. also Christie 2003: 3), as can be seen in these two items from the World Wide Web: Sunshine Auto Parts is an accredited distributor of Genuine and NonGenuine parts for Toyota, Isuzu, Mitsubishi, Lexus, and also used parts for Toyota Corolla and Camry (deportee). With over 12,000 sq. ft. of storage space, the company stocks a full inventory, including hard to get parts and parts that are not regularly requested. (www.sunshineauto.com.jm/our_company, not dated) A DAY or so before Christmas I was driving eastward nearing Ocho Rios when a motorist in a line of traffic going in the opposite direction kept overtaking the bumper-to-bumper vehicles in that line. I soon realised that this suicidal or myopic moron hurtling towards me was not about to allow my vintage Japanese deportee to impede his progress either, and he did not seem at all concerned about ending mine. (www.nwa.gov.jm/News/whatourcustomerssay/2005/jan)

As might be gathered from the context in which the word occurs, a Japanese deportee here has nothing to do with an illegal immigrant or a refugee but simply denotes a car imported from Japan. Another remarkable example of a meaning divergence of an -ee word in different varieties of English is transportee. On Australian and also British .uk websites, the meaning occurs largely within the context of Australian history, i.e., a transportee is ‘someone who was transported to the country (as a convict),’ whereas in the U.S., a transportee seems to denote ‘someone who uses public transport’:



Chapter 6.  -ee words in varieties of English 169

Australian English The following study sets out to sketch the profile of the average ribbon transportee to Tasmania by investigating those convicts who advised the Tasmanian authorities that they had been transported for offences relating to “ribbonism.” (www.jcu.edu.au/aff/history/articles/kennan, May 1996) American English The district may contract with a parent or guardian for individually furnishing transportation for an eligible transportee to an authorized district bus stop. (www.townsend.k12.mt.us/policy/policy8000/8140)

Unlike in Chapter 3, where the diachronic development of patterns and meanings of -ee words was considered within a perspective of language change, this chapter takes into account a simultaneity of patterns and meanings of -ee words in different geographical contexts, i.e., -ee words within a perspective of language variation. In the following, I take a closer look at -ee suffixation in various varieties of English. The first part is devoted to this word-formation in American English and its possible influence by language contact. Both historical and contemporary examples will be used to illustrate the development of the suffixation. This will be followed by a discussion of -ee nouns in Australian English usage, especially in view of the question, previously raised in this study, of a merging of the hypocoristic suffix -ie and the -ee suffix. Some -ee words will then be tested according to their occurrence in websites of different geographical origin, in order to gain a more complete picture of the distribution of -ee words in the different varieties of English. A discussion on the sources of variation and change of -ee words across Englishes concludes this chapter.

6.1

Suffixation with -ee in American English: A comparison of historical and contemporary usages

Some scholars have claimed (Mathews 1945; Marchand 1969) that the formation of -ee nouns is more prevalent in the U.S.A. than in Britain and that the wealth of -ee words created in the nineteenth and twentieth centuries is really an American “fashion.” This stance is also reflected in the fact that many of the scholarly articles (and anecdotes) on -ee formation can be found in American Speech, an academic journal on linguistic issues in American English. Why -ee formation should have come to be more popular and widespread in American English than in, for instance, British English is subject to speculation. As noted in Section 3.5, a political and growing linguistic separation between British English and American

170 Heterogeneity in Word-Formation Patterns

English became manifest in the eighteenth century, a time when the word-formation pattern had already diversified extensively and various syntactic and semantic models coexisted. At the same time, new lexemes with -ee endings from various languages around the world came into the English language as a result of colonial contacts (e.g. from Indian or Australian Aboriginal languages, cf. examples in Section 3.5). Both factors may have led to an increased popularity in the formation of new American English words. In addition, Mathews (1945: 106) sees a connection between the -ee suffix under discussion here and other, similar endings which are, however, not necessarily in the same category. He sees this “terminal element that has often been utilized in making American words” as influenced by the contact with other languages: It is possible that this ‘-ee’ fad may even have been inspired, in part at least, by the common Dutch diminuitive seen in cooky, and by the further fact that in some of the African languages words ending in consonsants are likely either to add a vowel or drop a consonant.

Mathews does not really explore such alleged influences further, and his list of words ending on -ee (or -y) is a rather crude mixture of patient nouns (e.g. devisee, contestee), diminuitives and hypocoristics (Dutchee, blackee, shirtee), particular African-American lexical items (Cuffee) and infinitive verbs to which an -ee ending is added (makee, workee). While he sees the latter as associated with African-American eighteenth-century speech, such words are also, in fact, associated with Pidgins on a more international level (Chinese Pidgin English stereotypical words like catchee, sabbee, makee; cf., for instance, Sebba 1997: 68), and are sometimes imitated in a “jocular” or patronizing fashion. But the general idea – that this word-formation may have gained in popularity through other, at first unrelated, word-formation patterns – is nonetheless compelling. We have seen in Chapter 3 as well as in our analysis of the new -ee words in Chapter 5 that there is occasionally an ambiguity of meaning between a non-volitional and a hypocoristic interpretation of the suffix. In Section 6.2, I will discuss this in more detail with regard to the Australian examples. The question remains whether -ee words are in fact more prominent in American English usage than in other varieties of English. A comparative analysis in Section 6.3 will show whether this is the case for the contemporary situation, or whether some particular lexical items may only be seen as “Americanisms.” As . He cites a passage written by Benjamin Franklin in 1787 to illustrate this point (1945: 107): “They are pleased with the observation of a negro, and frequently mention it, that Boccarorra (meaning the white man) make de black man workee, make de horse workee, make de ox workee, make ebery ting workee; only de hog.”



Chapter 6.  -ee words in varieties of English 171

far as the historical development of -ee words is concerned, there are some items marked as “U.S. origin” or “chiefly U.S.” in the OED: Marked as U.S.-specific (18th–20th century) in the OED

1764: petitionee 1779: conferee 1856: standee 1866: draftee 1870: contestee 1891: usee 1916: rushee 1934: enrollee 1940: selectee 1944: returnee 1945: retiree 1961: attendee 1961: tippee 1963: appraisee 1971: retardee 1974: preceptee 1980s: indemnitee

These 17 items, however, are not quite as many as one might suspect if the claim was true that -ee formation is a marker of “American-ness.” In Mathews (1951) Dictionary of Selected Americanisms, -ee-suffixed nouns are very scarce; there is only one example of this type – draftee. Panten’s (1959) study on Americanisms in a British newspaper (the Manchester Guardian Weekly) between 1948 and 1954 classifies a few -ee derivations as “Americanisms,” i.e., absentee, appointee, conferee, draftee, trainee but does not really explain his classification criteria. A preliminary test of the frequency of occurrence of some -ee words according to their geographical use should therefore provide some first evidence of the popularity of these words in particular geographical domains. In October 2002, I conducted a Web search of some -ee items according to regional domain criteria as part of a morphology project with students. Among other things, I searched for specific -ee words in websites marked with the domain specification .uk, and for the United States, .us and .edu. In comparison with these, .com and .org will . Unfortunately, domain specifications vary rather widely for the United States. The domain .us is not as commonly used as .uk would be for Britain or .de for Germany. Rather, alternative domains such as .edu for websites from the educational sector or .gov for official government sites are employed. I have therefore also included .edu as reflecting American usage.

172 Heterogeneity in Word-Formation Patterns

Table 20.  Domain-specific Web search 10/2002 .uk trainee retiree grantee advisee mentee deportee addressee refugee

85,600 964 1,280 120 1,380 271 8,200 128,000

.us 20,000 21,600 38,200 1,920 1,470 22 4,930 16,400

.edu

.com

.org

54,000 32,100 17,900 9,190 6,810 659 19,100 106,000

113,000 79,700 36,700 1,880 4,340 3,730 57,300 307,000

56,200 52,700 59,700 1,950 6,110 1,300 31,600 381,000

Table 21.  Domain-specific Web search 03/2006 trainee retiree grantee advisee mentee deportee addressee refugee

.uk

.us

.edu

.com

.org

1,850,000 26,000 23,600 417 42,100 884 139,000 2,250,000

190,000 286,000 397,000 11,500 11,300 573 52,200 185,000

489,000 309,000 178,000 124,000 110,000 642 166,000 924,000

3,540,000 1,790,000 567,000 51,500 106,000 114,000 1,130,000 5,270,000

1,000,000 840,000 1,460,000 25,400 143,000 37,400 536,000 6,790,000

also be considered – these are the most widely used domains which are, however, not geographically specified. Tables 20 and 21 present the distributional result of some of the words from the 2002 research in comparison with a follow-up search three and a half years later (March 2006). Firstly, the contrast between the 2002 search and the 2006 search documents the enormous growth of the Web during this relatively short time span. The comparison of the frequency of the different lexical items between domains illustrates on the one hand, the overall popularity of the word and, on the other hand, the specific frequency of occurrence in the separate domains. As already pointed out, the American-specific domain distribution is somewhat more dispersed than the British one. Nonetheless, we can observe a few interesting phenomena when we relate the results of the different items. For both 2002 and 2006, the hits for trainee – an item characterized as an “Americanism” (Panten 1959) – are remarkably high for the .uk domain, proportionally even more so in the earlier results. If we take the figures for the . The figures given here are absolute numbers. A comparison of some of the percentages follows below.



Chapter 6.  -ee words in varieties of English 173

Table 22.  Development of trainee and retiree in % (.com = 100) trainee 2002 retiree 2002 trainee 2006 retiree 2006

.uk

.us

.edu

.com

.org

75.7 1.2 52.2 1.4

17.7 27.1 5.3 15.9

47.8 40.3 13.8 17.3

100.0 100.0 100.0 100.0

49.7 66.1 28.2 46.9

“international business domain” .com as a point of reference (= 100%), we can calculate that the .uk domain reaches 75.7% of this figure, a result the .us (17.7%) and the .edu (47.8) domain do not achieve if taken together. A comparison of the figures for 2006 (.uk: 52.2% of the .com figures of 2006; .us: 5.3%; .edu: 13.8%) shows an even more unequal relationship. This could be evidence that the proportions of the domain specific websites on the Web have changed, a hypothesis which can be further discussed if we take a look at the distribution of other search words together with the follow-up results. Retiree, another -ee word of “U.S.-specification” (OED), seems to be a more likely Americanism if we look at the figures given for both periods: in 2002, the usage of this word on UK websites seems almost negligible in that it makes only 1.2% of the frequency of occurrence on .com websites, while the .us domain achieves 27.1% and .edu websites 40.3%. The highest figure here – after our reference point .com (100%) – is .org with 66.1%, a fact which demonstrates that the usage of different words is also context specific. The comparison with 2006 shows similar figures for .uk (1.4%), and considerably lower ones for .us (15.9%) and .edu (17.3%) (see Table 22). The figures presented in Table 22 are somewhat rudimentary but we can nonetheless draw some tentative conclusions: (a) retiree is a much more convincing Americanism than trainee. The latter seems to be, in fact, remarkably widespread in the UK if we take the usage on .uk websites as an indicator. Retiree, on the other hand, seems to be hardly ever used on .uk websites; (b) in the comparison of the 2002 and the 2006 search we can observe a frequency growth of the reference figure in .com domains in contrast to the more geographically specific domains. However, the proportion of the rise is not comparable, i.e., the increased frequency of an item on .com websites may also be due to the fact that the frequency of an item varies according to context. In analogy to this, we can also make further observations: advisee is one more word which seems to be rarely used in British English as opposed to American English. In 2002 there were only 6.3% UK-specific uses with regard to the reference figure in .com, a percentage which dropped to 0.8% in 2006. In comparison, the usage in .us was 102% (2006: 22.3%) and in .edu even 488% (2006: 240%) – no

174 Heterogeneity in Word-Formation Patterns

Table 23.  Development of mentee in % (.com = 100) mentee 2002 mentee 2006

.uk

.us

.edu

.com

.org

31.7 39.7

33.8 10.6

156.0 103.7

100.0 100.0

140.0 134.9

Table 24.  Development of refugee in % (.com =100) .uk refugee 2002 refugee 2006

41.6 42.6

.us 5.3 3.5

.edu

.com

.org

34.5 17.5

100.0 100.0

124.1 128.8

doubt due to the fact that there are more advisees in education than in business. The first entry of advisee is listed as 1824 (OED), some years earlier than trainee (1841). The date of occurrence in nineteenth and twentieth-century words is therefore not necessarily an indicator of whether or not a word is a likely Americanism. If we take a look at the development of the twentieth-century word mentee in the different domains, we notice that, among the geographically specific domains, the American education sector scores highest. We can also see that the only domain which marks an increase in relation with .com is, in fact, .uk – all other domains show a decrease. Therefore, it seems as if mentee is gaining ground in Britain in comparison to the United States (Table 23). Finally, the long-established word refugee has the highest overall score of the words researched above. The distribution of its usage on different websites can be illustrated as in Table 24. In relation with the usage in .com websites, the .uk figures and the .org figure increase slightly, whereas the .us and the .edu percentages decrease in 2006. After trainee, the figures for refugee show the widest distribution on .uk websites. In sum, we have seen that there are Americanisms among the -ee words: retiree and advisee, for instance, are hardly ever used in British English, or in a number of other varieties of English (cf. below). On the other hand, there are also -ee formations that seem to be particularly widespread in British English, such as trainee or refugee, as well as some where an increase in British English usage (mentee) can be observed. It must therefore be concluded that the question of “Americanism” depends very much on the specific lexical item and not on the type of word-formation itself. It is clear that a Web search on the geographical distribution of lexical items has its flaws: we don’t know whether the 2,250,000 .uk websites which contain the item refugee were all created by native speakers of British English, and the same



Chapter 6.  -ee words in varieties of English 175

limitations apply to the American websites. But they equally apply to all larger corpus material: we also have no record of the language history of the speakers/writers in the established corpora. However, we can see that it is possible to establish certain trends with regard to the regional distribution of lexical items so as to draw putative conclusions on their usage in varieties of English.

6.2

Australian English usage: On hypocoristics and -ee words

In the above comparison, only British and American English were considered. But how do other specific varieties behave with regard to their employment of -ee words on the Web? In my 2002 search, websites of Australian, New Zealand and Irish origin were also considered and these particular searches have been followed up in March 2006. To correspond to the above results on U.K. and American Web usage of particular -ee words, we can document some comparable results of the search on .au (= Australian), .nz (= New Zealand) and .ie (= Irish) websites (see Table 25). By and large, the varieties discussed here behave more like British English than American English in their usage of retiree, advisee and trainee. We can see that the numbers of retiree and advisee in New Zealand and Ireland are insignificant and do not even reach 0.1% of the .com figure. This might be attributed to the small size of population and, consequently, small number of websites. However, we also notice in the comparison with the figures of trainee that a higher percentage can be reached with a more popular item – again, the analogy to British English usage is compelling for both Irish and New Zealand English. The Australian result is somewhat more complex: for both the 2002 and the 2006 results, trainee is also much more frequently used than retiree and advisee. However, we observe a slight increase in the proportion of usage for the latter two items and a decrease for the former. Is Australian English increasingly influenced by American English Table 25.  Distribution of retiree, advisee and trainee on Australian, New Zealand and Irish websites (.com = 100%) retiree 2002 retiree 2006 advisee 2002 advisee 2006 trainee 2002 trainee 2006

.au

%

.nz

%

.ie

%

.com

2,490 65,900 36 1,820 42,400 623,000

3.1 3.7 0.2 3.5 37.5 17.5

110 526 8 52 3,900 77,500

0.13 0.029 0.04 0.01 3.4 2.2

99 275 8 30 4,170 131,000

0.12 0.015 0.04 0.005 3.7 3.7

79,700 1,790,000 1,880 51,500 113,000 3,540,000

176 Heterogeneity in Word-Formation Patterns

usage? Does Australian English have a significant idiosyncrasy in its usage of -ee words? After all, some of the earliest examples of -ee words in a variety of English other than British or American – expiree and assignee, (cf. above) – occurred in Australia. For these two nineteenth-century examples one might suspect that the legal register and the association of the word-formation with borrowings from French may also have aesthetic reasons and result in a euphemistic connotation of the words: after all, expiree sounds better than ‘ex-convict,’ assignee more worthy than ‘unpaid servant.’ In her investigation of the morphology of Australian English, Roswitha Dabke makes particular mention of this word-formation (1976: 33–36). With regard to historical examples like assignee, she notes that in the first centuries of its usage, derivation with the suffix -ee was “productive mainly in the language of jurisprudence. As Australia was a convict colony during the first decades of European settlement and daily talk centered around the convict and the convict system, a few -/’i/-derivatives of the language of the law courts were in common use during the first half of the nineteenth century” (Dabke 1976: 36). These examples from the legal context are well known and almost stereotypically Australian. As Baker (1945: 44) notes, “if our authors are to write about Australia’s early days, it is essential that they learn something of the language of the times. It is not sufficient merely to drag in expressions like assignment system, assigned servants, assignees, free-grant system, ticket of leave man […] on the ground that Australia was the first country to give these words currency and meaning.” Indeed, the history of Australia and, consequently, Australian English should not be reduced to the narrow context of its role as a penal colony. In her Variation in Australian English, Barbara M. Horvath (1985: 29) sums up the two basic hypotheses by earlier scholars of how Australian English came to be a distinctive variety of English: either Australian English is “simply the continuation of changes already in progress in England,” or it is seen as “basically London Cockney and was transported here by the first people to arrive from England.” While the London Cockney influence is certainly most evident in phonological patterns, such a view neglects other (mainly lexical) contact influences from the “many voices” (Leitner 2004) of Australia, influences both from other immigrant languages and . Such euphemistic connotations of -ee formation might also have played a role in other words – not limited to the Australian context – which denote a person’s status and which might be interpreted as negative, for instance, divorcee, retardee, deportee, etc. . Note, however, that one of the earliest dictionaries on Australian English slang, James Hardy Vaux’s New and Comprehensive Vocabulary of the Flash Language (1812) consisted mainly of English criminal cant, making the convict circles of Australia one of the earliest sources of record of Australian language usage; cf. Ramson (1966: 10–11).



Chapter 6.  -ee words in varieties of English 177

indigenous languages. In Australian Aboriginal languages, for instance, word endings on ‘ee’ are not uncommon, such as corroboree, 1793; Baker (1945: 5) cites coee and budgeree as “words we filched from the natives.” Direct links between such Aboriginal word endings and an increase of the -ee formation pattern cannot be made but it is plausible that it could be one of the sources of the alleged popularity of -ee formation in Australian English (cf. below). Relating to the different syntactic patterns of -ee suffixation in general and in Australian English in particular, Roswitha Dabke (1976: 33–34) notes that there are some coinages which are particularly common: employee, followed by trainee, “a word which, in Australian English, may refer to any kind of training; thus the Age writes ‘Trainee priests want the right to wed’ (10 April 1969).” In her observation, words like divorcee or interviewee are “neutral in meaning” in Australian English in that they can be found in casual conversations as well as in other text types, i.e., they “may be found in texts of a scientific nature; they are also firmly established in everyday speech” (Dabke 1976: 34). Dabke points out that this stands in contrast to American usage where these words “belong to officialese” (ibid.). Although she reports that she has come across a “reasonably large number of nonce words of this type in Australian texts,” Dabke (ibid.) does not see Australian English using this word-formation as extensively as – in her estimation – American English. The ad-hoc formations she cites are mainly verb-derivations, as for instance, boree, cheatee, knucklee. While the two former ones are documented elsewhere (cf. Appendix 1), the latter is not and will therefore be presented here (cited in Dabke 1976: 34): knucklee […] “the young gentleman who had been knuckled – the knucklee so to speak – was thoughtfully trying to throttle the knuckler” (Nation Review, 6–12 July 1973)

Another interesting nonce word is derived from a noun (Dabke 1976: 35): camp-fringee […] ‘person from the fringe of a camp’: “a quarterbodgie from a Berlin slum, a camp-fringee, New Australian fireman” (Hal Porter, A Bachelor’s Children, 1962: 286)

Besides -ee derivations, Dabke (1976) devotes as substantial chapter of her Morphology of Australian English to hypocoristic formations, i.e., alternative forms of words or names which are usually formed by clippings or by clippings plus suffix.

. One of Australia’s national newspapers.

178 Heterogeneity in Word-Formation Patterns

Like Dabke (1976: 39), Jane Simpson (2004: 643) notes that the usage of hypocoristics is particularly widespread in Australian English – a notion which has already been commented on by Gunn (1972: 60): Aren’t we reaching the inane, when we Australians start accepting beddie, cardie (cardigan), Chrissie pressie, ciggie, habbie (haberdashery), leckie (lecture), prossie (prostitute), sandie (sandwich), tabbie (tablet), weepie, and yewie (U-turn)? Of course, these and other habits are not restricted to Australia, but the increase in their popularity here is phenomenal.

In her study on hypocoristics in Australian English, Simpson (2004) analyses a body of 1740 hypocoristic words collected by herself and David Nash, as well as from written sources and other authors’ works. She classifies the data into nine different templates for forming hypocoristics in Australian English, the most popular forms being “syllable plus /i(s)/” forms (824 items or 47% of her data), of the type coldie for ‘a cold beer’ or gladdie for ‘gladiolus.’ The second most popular form is “syllable plus /o/” as in, for instance, prawno for ‘prawn seller’ or journo for ‘journalist’ (333 items or 19% of her data). Simpson (2004: 647–648) notes that the most common form of “syllable plus / i/,” often associated with “baby talk” in international English contexts, has been in usage in Australia for derived words, and in normal adult conversation and writing, since the nineteenth century. Generally, the most common patterns of forming this type of hypocoristic word would be /i/ suffixing of proper names (first names, last names), brand names, place names, nouns, especially nouns forming names of occupations (e.g. speechie/speech pathologist) and adjectives (e.g. marvey/marvellous), often also used to create a de-adjectival noun (e.g. oldie, toughie). As Simpson observes, verbs can sometimes be the basis of such derivations – a fact which might lead to ambiguity with the -ee suffix form (2004: 648): As with adjectives, /i/ is occasionally added to verbs to create nouns: clippie ‘ticket examiner,’ twisty/twist (a brand of savoury snack). But the stressed suffix -ee which forms verbs from nouns is probably a different suffix (escapee, refugee, absentee; Dabke 1976). These forms are not alternative words, and are often more than two syllables long. However, the two syllable forms are only distinguished by stress (grantee), and some forms have two interpretations, thus both blockee and blockie appear for ‘someone farming on a block of land.’

While in oral pronunciation the two forms are, in fact, distinguishable by different stress patterns, the spelling is sometimes less conclusive. Is the blockee a less . These are, in fact, not the very best examples for verb-noun derivations (cf. Chapter 3 on the history and possible bases and sources of these formations).



Chapter 6.  -ee words in varieties of English 179

volitional farmer than the blockie? We have already noted on such ambiguities in earlier chapters (cf. Section 2.3). Due to the attested popularity of hypocoristic formations with /i/ in Australian English, such multiple interpretations seem to be particularly applicable to the Australian context. This seems to be highly prevalent where the main criteria for a hypocoristic formation are also observed: for instance, failee (C 297), surfee (C 881) or faintee (C 298) are sometimes possible “alternative words” to failer, surfer or fainter, i.e., they are subject formations, if non-volitional ones. Similarly blue rinsee (C 749) might not be a subject formation (*blue rinser) but has both a patient and a (if somewhat patronizing) hypocoristic meaning. Simpson (2004) stresses the fact that hypocoristics are usually monosyllables (plus suffix). However, in our observation on multiple (hypocoristic and non-volitional) interpretations we can also find examples which consist of polysyllabic stems plus suffix, like rehearsee (C 683) or convertee (C 123). We can thus summarize that there are two possible sources for an increased popularity of forming new nouns with an -ee suffix: on the one hand, there is a long established tradition of separate -ee words in a legalistic context due to Australia’s particular history as a penal colony. It is quite possible that a euphemistic connotation of -ee formations plays a role in this context. On the other hand, we have seen that the popularity of -ie hypocoristics in Australian English can also result in an increased use of -ee forms with ambiguous meanings. Then again, what may be an advantage for the creation of some -ee words may be a disadvantage for other, already existing ones: after all, the Australian love for hypocoristics also results in forms such as reffo for refugee (Simpson 2004: 649). In the following comparison of the domain-specific use of some of the new -ee words in American, British, Australian, New Zealand and Indian Englishes, we will be able to investigate whether some of the new lexemes show a higher rate of occurrence in particular varieties of English such as Australian English.

6.3

A corpus-based analysis of some new -ee words in American, British, Australian, New Zealand and Indian Englishes

In my Web search for new -ee lexemes documented in Chapter 5, I observed that some of these new words often occurred in specific geographical contexts and at times, with rather specific meanings. I have therefore selected nineteen of these new -ee words and have conducted a follow-up search (March 2006) in websites . Note that for all of these -ee words, various interpretations have been found. A faintee therefore might be ‘someone who faints’ (by nature unvoluntarily) or ‘someone on whom somebody faints’ (cf. also Chapter 2 for more details on various interpretations of surfee and failee).

180 Heterogeneity in Word-Formation Patterns

Table 26.  Selected new -ee words in varieties of English C 45 C 53 C 85 C 123 C 159 C 209 C 343 C 354 C 357 C 440 C 553 C 586 C 701 C 703 C 727 C 739 C 744 C 848 C 926

assuree awardee checkee convertee dependee displacee greetee harassee healee lendee persuadee pollutee removee renouncee retainee revertee revokee stalkee transmittee

.uk

.us

.edu

.au

.nz

.in

2 344 8 107 28 6 6 18 104 12 28 12 1 12 138 14 11 163 12

1 675 1 9 1 101 1 13 15 34 4 1 0 1 28 0 25 20 2

1 787 22 13 48 17 15 71 36 57 107 59 5 0 30 1 52 32 6

25 607 1 49 14 0 2 5 35 8 16 2 27 0 2 1 7 58 194

0 124 0 7 1 1 1 3 16 7 2 0 1 0 0 0 0 6 3

0 394 0 0 0 0 0 0 0 2 0 0 0 10 0 0 2 0 1

with specific regional domains. In order to ensure a reasonable rate of occurrence beyond the more widespread international .com and .org domains, most of the words chosen are from the “frequent” category. The procedure of searching, selecting “countable” examples and eliminating nonsense cases from the websites followed the pattern explained in Section 5.2. As can be seen in the results in Table 26, the rate of occurrence of these words varies considerably in the different varieties of English or regions. Results that are considered significant for a particular regional domain are marked in bold type. (1) .au: Following our discussion in Section 6.2, we can observe with regard to the Australian domain .au that there are some -ee words which seem to be used extensively in an Australian environment: Some of them (assuree C 45, removee C 701, transmittee C 926) denote legal terms in the context of immigration and employment and can be found on many Australian government sites:



Chapter 6.  -ee words in varieties of English 181

assuree C 45 Definition: assuree The assuree is a new resident covered by an AoS. That is, an assuree is a person who was required to obtain an AoS before his or her visa could be granted. There are 3 kinds of assurees: – PRIMARY ASSUREE – the person who was granted the visa on the basis of meeting primary criteria for grant of that visa, – DEPENDENT ADULT ASSUREE – an adult who is granted a visa on the basis of meeting secondary criteria for grant of that visa, Example: A dependent spouse or aged parent of a person who is granted a visa on the basis of meeting primary criteria. – DEPENDENT CHILD ASSUREE – a dependent child who is granted a visa (www.facs.gov.au/guide/ssguide/11a310.htm, not dated)

That this meaning of assuree is specific to the Australian context can be seen when comparing it to the following example from an .edu site where the word is used in a much more general sense of the speech act of assuring: […] the assurer probably used the imperative voice, giving a command to the assuree. Promise, in contrast, does not show. (www.stanford.edu/~coppock/promise.pdf) removee C 701 (1) If the applicant is a person who has left Australia (otherwise than as a removee or deportee) following: a. compliance action by Immigration resulting in the detection of the whereabouts of the person, being an unlawful non-citizen, illegal entrant or prohibited non-citizen; […] (www.austlii.edu.au/au/au/legis/ cth/num_reg/mr1994n268254/sch5.ht)

As with assuree, the Australian-specific domain .au is the only one which shows a significant usage of removee. While in the Australian context it appears to be used synonymously with deportee, removee has a less specific meaning in other geographical environments: There was too much stuff to fit in honest Ronnies van so they used the back of the removee’s vehicle to save a second trip. I was tempted to feel sorry for a split second til I realized he was most likely signing on. I’ve met one or two of these light removals fellows since – when moving myself. (fearandloathinginlondon.blog.co.uk, August 2005)

182 Heterogeneity in Word-Formation Patterns

With 194 websites, transmittee occurs in the “Australian” category much more frequently than in any other context. This legalistic term might therefore be taken as a “true” Australianism, i.e., the word does not really occur in other contexts with a different meaning (as assuree and removee do): (g) […] Where a business is before or after the date of this Agreement transmitted from the University (in this clause called the transmittor) to another employer (in this clause called the transmittee) and an employee who at the time of such transmission was an employee of the transmittor in that business becomes an employee of the transmittee: (www.cpsu.unsw.edu.au/Enterprise/EA, not dated)

(2) .in: Awardee C 53 generally has a high frequency of occurrence on all regional websites. On Indian websites, the comparison with the frequency of other lexical items makes it stand out considerably. There seems to be no difference in the meaning of awardee (‘someone who has been awarded a prize/scholarship, etc.’) in the different geographical environments. The other remarkable item found on the .in domain was renouncee C 703 – ‘somebody to whom something is renounced’ – and it seems to be rather specific to Indian English. Of course, the cost of the rights share acquired by the original shareholder is the price actually paid by him to the company for acquiring the rights share. But where the rights renouncee acquires the rights share, the cost of the rights share is equal to the cost incurred by him for purchasing the rights entitlement plus the price paid by him to the company for acquiring the rights share. (incometaxindia.gov.in/circulars/1994/Cir684)

(3) .uk: UK websites document a number of remarkable -ee word usages: convertee C 123, healee C 357 and stalkee C 848 have the same meaning as in other regional contexts but seem to be much more widespread in Britain than elsewhere. One of the reasons for this could be a special popularity of the underlying verbs in British English – a possibility which could not be substantiated in a comparison of Web hits of the verb stems (convert, heal, and stalk) on British and American websites.10 It follows that the particular lexemes stalkee, convertee and healee are more common in British English usage than in American English. The meaning of the formations does not differ in the different geographical contexts. Retainee C 727 and revertee C 739, on the other hand, appear to relate to particular Britishmedical and legal-linguistic practices, as the following examples illustrates:

10. March 2006: stalk: .uk 815,000; .us 109,000; .edu 620,000; convert: .uk 6,000,000; .us 1,510,000; .edu 16,700,000; heal: .uk 2,210,000; .us 263,000; .edu 1,330,000.



Chapter 6.  -ee words in varieties of English 183

How many sessions can a GP retainee work in general practice? On average four per week. This can be averaged over a quarter as a total of 52 sessions (including leave), with a maximum of six sessions per week and a minimum of one session per week. How can GP retainees increase their commitment to general practice? If an increase in sessions is desired, the retainee should write to the patch Associate Director [link to contacts] requesting permission. The agreement to increase should be copied to the health authority. (www.sevwesdeanery.nhs.uk/Default.aspx) The intention was that, if the land ceased to be used as a poor school, then the legal title of the school trustees would come to an end, and the land would ‘revert’ to the original donor (or his successors) – ‘the revertee.’ The 1987 Act was brought in to simplify the position. Firstly, if the land ‘reverts’ (because the school has closed) and the trustees then sell, the revertee only has a claim on the proceeds of sale. Thus, the revertee cannot claim the property off the person who has bought it from the school trustees. Secondly, the Act established the right of the trustees to claim adverse possession against the revertee, so, if the use as a poor school ends, the trustees who remain in occupation can claim title after 12 years and so remove all rights of the revertee. (www.practicalconveyancing.co.uk/content/view/9784/1)

Both retainee and revertee therefore appear to be true ‘Britishisms.’ (4) .us and .edu: For the American context, some linguistic items from the search list may be highlighted: Dependee C 159, displacee C 209, greetee C 343, harassee C 354, lendee C 440, persuadee C 553, pollutee C 586 and revokee C 744 all show a higher frequency on these American domains than in any of the other regional sites. With the exception of perhaps revokee – a legal term for ‘someone whose business license has been revoked’ – none of them are as specific to American (legal or other) practices as transmittee is to the Australian, renouncee to the Indian and retainee to the British contexts. revokee C 744 […] (2) “Revokee.” – “Revokee” means a person whose business license has been revoked, denied or withheld under § 2124 of this title; (3) “Substantially same business.” – A business is substantially the same as a revokee’s business if any of the following conditions exist: a. The business location is substantially the same as the revokee’s; b. The customers are substantially the same as the revokee’s; c. The employees are substantially the same as the revokee’s; or

184 Heterogeneity in Word-Formation Patterns



d. The physical assets of the business are substantially the same as the revokee’s. […] (www.delcode.state.de.us/title30/c021/index.htm, not dated) Lutgen’s expanded temporary restricted license was intended to facilitate his attendance at AA meetings and therapy, functions that have been deemed unnecessary for a revokee’s present employment. See id. (citing Iowa Dist. Court for Greene County, 458 N.W.2d at 2). (www.judicial.state.ia.us/ appeals/opinions/20021211) […] He said he had been told that as a revokee with less than six months to serve he had no need of sentence planning. He asked to be considered for category D […] (inspectorates.homeoffice.gov.uk/hmiprisons/thematic-reports1/recalledprisoners.pdf, not dated)

(5) .nz: No remarkable deviations can be noted for the New Zealand websites. These examples of an analysis of the distribution of -ee words in varieties of English deal with just a fraction of my entire corpus (19 words = almost 5 per cent) and consist characteristically of the more frequently attested new -ee words. This brief exploration should illustrate that there is, indeed, a diversification of meaning or a creation of individual new lexemes in different varieties of English, in addition to the diversification of -ee formation due to language change and the consequent simultaneity of various syntactic and semantic patterns. There is much room for further investigation, both of geographically specific uses of long-established -ee words and of the distribution of new -ee words in varieties of English. Because of the nature of these regional websites – they are usually official websites of the government, or educational and other institutions – ad-hoc creations are not likely to be found there. The international business .com and organizations .org sites, which also host many of the genres and text types in which nonce words are to be expected (cf. the analysis of hapax legomena in Section 5.5), are far more widespread than regional domains and hence more important for the production and further development of neologisms.

6.4

Global meets local: Variation and change of -ee word production in varieties of English

One might lament the fact that in these international websites the regional background of the writers/creators remains largely anonymous – at least at first sight – and that a regional relationship cannot be established with an easy search



Chapter 6.  -ee words in varieties of English 185

mode for most of the new -ee words found. However, I would argue that this is symptomatic of other processes of globalization in an international language like English. In this sense, the anonymity of the linguistic background of the writer interacting with readers and other writers of indefinite linguistic extraction might be a more accurate picture of the state of English in its global interactions than a straightforward regional classification would propose. Is it the case, as John Algeo (1991: 3) writes, that “all linguistic varieties are fictions”? He suggests a rather idiosyncratic view of language variation when he continues that, “Because language is constantly changing, adapting to the circumstances of its use and the moods of its users, every instance of use is unique and different from every other use.” While a very broad view of linguistic heterogeneity can undoubtedly be extended to the individual speaker, such a leveling of any regional, social or contextual differences of language use among clusters of speakers is not helpful for either the description of synchronic use or the observation of language change phenomena. In my view, an increasing globalization of English language usage among a certain group of speakers does not necessarily contradict a continued local practice. In his article “Beyond homogeny and heterogeny” (2003), Alistair Pennycook examines the “dichotomization between local, multiple vernacular languages and the monolingualism of the language of power” as well as the “ways in which English is as much a language of global discommunication as it is a language of global communication” (2003: 6). He argues against both what he calls a “homogeny position” on the diffusion of English, i.e., the view that the global spread of English is leading to a “homogenisation of world culture,” and a “heterogeny position,” i.e., the idea that English is becoming increasingly pluricentric, with new norms of usage and performance emerging in different parts of the world. Pennycook asks: […] how can we start to account for the constant reciprocity between globalization and localization? […] How can we account, to take an example from a different domain, for the processes by which beer production became centralized in large breweries in many parts of the world, was then decentralized after beer drinkers protested against homogenized beer, and has now become part of a globalized process of localization, in which heterogeneous beer production is being (Pennycook 2003: 9) pursued globally? 

Applied to a global usage of English by either native speakers of a particular regional variety of English or by speakers of English as a second and international language of communication – which is what we encounter on the internet – we can note that the course of language influence is not necessarily unidirectional. Depending on the type and context of contact, we can see the following processes

186 Heterogeneity in Word-Formation Patterns

operating simultaneously in the interplay between localization and globalization of English language usage. a. Non-participation versus active participation: Not all speakers participate in the processes of globalization in the same way. Just as there are beer drinkers who have always stuck to their local brewery or have been brewing their own beer, there are speakers whose communicative practices do not reach out beyond a very narrow local social network. On the other end of the scale there are those participants who very actively engage in wider communication practices, have rather wide social networks, including international communication contacts. Especially in the case of users of English as a second and international language, the level of education and social mobility can be assumed to be higher than that of non-participants. b. Local usage persists despite globalization: The fact that speakers of British English have come across the words advisee or retiree does not necessarily result in an adoption of these words into British-English usage, perhaps because they are seen as “Americanisms.” Similarly, speakers of Jamaican English may continue to use words like deportee with a locally specified meaning even if they are aware of the meaning of the lexeme in other geographical contexts. c. Local and global norms of English usage influence one another: There is a constant interplay between the local and the global. Just as new elements such as neologisms originate from a particular locality and may or may not spread to wider circles, trends in international language usage are taken up in localized contexts. Lexical influence and influence on word-formation by nonnative speakers can also play a part in the adaptation and transformation of English. In the first instance, we have a range of possibilities from no contact with wider/ global communication to extensive contact. In the second development, we have contact without change In the third possible course of action, we have contact with the result of multidirectional change. Rather than contradicting or excluding one another, the three processes, I would argue, exist side-by-side. The suffixation with -ee arose out of a language-contact situation and has continually been influenced by language contact – be it continued borrowings from French until well into the nineteenth century, or through some other influence of borrowed words from international contexts or particular preferences of this word ending in contact varieties such as Pidgins or early, arguably creolized, stages of African American English. The word-formation will persist to be influenced by the growing globalization of English and Englishes, and particularly through uses in electronic international communication, such as on the internet.



Chapter 6.  -ee words in varieties of English 187

The use of the World Wide Web in researching specific word-formation processes and neologisms has the advantage that one can find many text types which can be seen as primary sites of language adaptation, new creation, and transformation. On the other hand, we have also observed that other text types do not have the same interactional and innovative potential, and it is in these that regional coinages like renouncee or preference for a particular type of meaning in -ee words – such as hypocoristic or euphemistic meaning – are less likely to spread beyond the specific locality of origin. Thus, while global contact on the World Wide Web results in diffusion and change in one instance, variation between different meanings and usages of -ee words will also persist in different varieties of English.

conclusion

On the study of an individual word-formation pattern General and particular implications

This study has presented an in-depth look at a particular type of word-formation in English. Such a rather intense treatment of one particular suffix was prompted by the fact that the ambiguities and the simultaneity of various patterns in the suffixation with -ee have often been cited as a source of puzzlement in synchronic morphology: Plag (2003: 45), for instance, uses the variability of the -ee suffix as an example of the possible contradictions between word-formation rules and actual formations, concluding that “ideally, one would find an explanation for these apparently strange conditions on the productivity of these affixes.” A quarter of a century earlier, David Gold (1977: 160) had already proposed that “a study of nouns ending in -ee, which have generated some prescriptivist wrath, would be useful.” Beyond the study of a particular suffix in English, this investigation has also attempted to highlight a number of wider theoretical and methodological concerns in the field of word-formation. By first investigating the various synchronic syntactic and semantic patterns of -ee formation in Chapter 2 and then undertaking a close and carefully documented analysis of the history of this word-formation in Chapter 3, this study has attempted to bridge the gap between the often all-too sharp dichotomy of synchronic and historical perspective on word-formation patterns. Chapter 4, with its focus on twentieth-century formations, has largely been concerned with the question of productivity versus creativity in word-formation and the specific conditions in which an ad-hoc creation or a nonce word becomes established as a new word. This is also fundamentally associated with the relationship between morphology and the lexicon, i.e., possible words versus actual words. Yet, how are actual words attested? The empirical question was taken up in Chapter 5, where a substantial test of 1,000 possible new -ee words was conscientiously conducted on, arguably, the largest and, for the purpose of the study, most useful text corpus, i.e., the World Wide Web. Here, the question of productivity and methodology of data collection were fruitfully linked to investigate new -ee words in various

190 Heterogeneity in Word-Formation Patterns

text types and in different forms of establishment. Unlike all previous treatments of the suffix, this study is not content to rely on isolated examples as a data base; rather, it is the first study to undertake a structured empirical approach to contemporary productions of this type of word-formation. The final chapter then looked at -ee formation in a more globalized perspective, as a word-formation which arose out of a contact situation and is continually influenced by the contact between different varieties of English as well as influences from other languages. Here, the perspective of language variation and change also offered insights into the different distribution of -ee words in global and local contexts. The new insights gained from the thorough research of the -ee formation pattern are varied and concern both the particular and immediate investigation of the suffixation with -ee and the more theoretical issues in word-formation theory as outlined above. In Chapter 2, it was demonstrated that none of the existing syntactic or semantic models are capable of describing the multifarious ambiguity of the -ee formation patterns and the simultaneous presence of human/sentient and nonhuman/sentient, verb-derived and noun-derived formations or -ee words with non-volitional versus clearly agentive and volitional character. Rather, there are several syntactic and semantic sources of the high variability of this pattern, such as, for instance, ambiguity through changing and parallel rules in the development of -ee verb-derivation (direct/indirect object, subject) or ambiguity through underlying noun-derivations from -er words, i.e., if -ee words can also be formed as correlative nouns from -er words, the correlation may sometimes be extended to include non-human, immaterial, etc. words. This was seen as particularly relevant with regard to technical descriptions/tools. Furthermore, language contact was illustrated to be a source of ambiguity, for instance, through loanwords and parallel French participle formations, as well as through partial overlapping in meaning with other formations with -ee, especially in language contact situations. As a model for the description of this variability of patterns of -ee formation, a gradable framework of more prototypical/central and less prototypical/marginal features was used. As an explanation for the simultaneous presence of the patterns, a diachronic approach was used to supplement, question and, at times, rectify the prevalent synchronic perspective. A detailed description of the syntactic and semantic changes over the course of six centuries explained how and why there is such heterogeneity of patterns. By compiling and analysing a comprehensive list of ee words from the fourteenth to the nineteenth century, it was shown that the following parameters of change apply to this particular word-formation: (a) the word class of base of derivation (correlative noun, verb), (b) the type of passive object noun (indirect, direct), (c) the passive character (object, subject), (d) the



Conclusion 191

semantic field of occurrence (legal, general, ironic/jocular), (e) human reference (human, non-human, including animals & technical components), and (f) the specific use in particular varieties of English (U.S.A., Australia, Scotland). For the twentieth century (Chapter 4), the analysis of relatively recent formation raised the question of productivity of this word-formation pattern and the relationship between actual and possible words. As a solution to this problem and the problem of data collection and measuring productivity, a test of 1,000 potential new -ee words was designed and carried out in Chapter 5. This analysis resulted in a number of insights into the production of -ee words: firstly, the sheer size of the new data on -ee words and the ratio between potential words and actually realised words leads to cautiously suspect a higher rate of productivity for the word-formation pattern than previously assumed. Secondly, the analysis of the new data provides evidence for the dominant syntactic and semantic patterns in -ee word-formations. The role of collocations with -er words in the production of -ee words, for instance, leads to question the previously assumed unique role of the verbal base in -ee formation. It also shows the distribution of prototypical semantic and syntactic features in the various categories of frequency of occurrence in the data. Furthermore, a comparison with earlier patterns of -ee formation (Chapters 3 and 4) demonstrates both the continuities and processes of change in the new and recent -ee formations. With regard to the semantic domains in which the most frequently occurring new -ee words appear, we can note additional fields (e.g. professional-interpersonal relations, sex, military usage) to the already established domains of usage (e.g. legal, interpersonal, humorous). We can also note a growing diversity and ambiguity of meaning especially in those words classified as hapaxes. By classifying the new words according to their frequency of occurrence, we can also establish differences between more creative nonce words and more prototypical frequent words, offering conclusions on routes of establishment. The empirical data given here also provides a basis for future research on the career of present nonce words; it offers insights on the context-based production of nonce words and the role of “oral” or dialogic characteristics in ad-hoc creations, critically reviews the relationship between data sources, i.e., listed words (in dictionaries) and actual words (actual usage); lastly, it explores the potentials of using the Web as a corpus. All of these points go far beyond the immediate and exclusive application of the -ee formation pattern but can be seen as applicable in a broader sense to the study of word-formation patterns. In the last part – Chapter 6 – a look at -ee formation in varieties of English deals with the relationship between processes of globalisation and maintenance of local usage in both specific and more general ways. By yet another analysis of frequency patterns of -ee words on the Web in various regional environments, it was shown that there are preferences for certain formations in the British versus

192 Heterogeneity in Word-Formation Patterns

the American context. For Australian usage, a merging of the hypocoristic meaning of another suffix and the often non-volitional meaning of -ee illustrates the regional heterogeneity of the formation. It was also shown that several of the “new -ee words” have a specific origin and meaning in particular varieties of English, such as Australian, Indian, British, American and New Zealand Englishes. Influences on -ee words out of language contact situations have always marked the very development of this word-formation and continue to do so in an increasingly globalized communication world, without losing some of the specificities of patterns and meanings in various environments. This study is ultimately a plea for word-formation studies which take into account the possible variabilities of a word-formation pattern by using not only a synchronic but also a diachronic perspective, by looking at the formation patterns not in a regional and contextual void but considering heterogeneity in different varieties of English and, last but not least, by providing and analysing a solid material data base.

Works cited

I.

Dictionaries

Algeo, John (ed.). 1991. Fifty Years Among the New Words: A Dictionary of Neologisms, 1941– 1991. Cambridge: CUP. Barnhart, Robert K., Steinmetz, Sol & Barnhart, Clarence L. 1990. Third Barnhart Dictionary of New English. New York NY: Wilson. Green, Jonathan. 1992. Tuttle Dictionary of New Words since 1960. Nearly 2,700 Neologisms from acid to wysisyg. Boston MA: Tuttle. Lehnert, Martin. 1971. Rückläufiges Wörterbuch der englischen Gegenwartssprache. Leipzig: VEB Verlag Enzyklopädie. Mathews, Mitford M. 1951. Americanisms. A Dictionary of Selected Americanisms on Historical Principles. Chicago IL: The University of Chicago Press. Muthmann, Gustav. 1999. Reverse Dictionary. Based on Phonological and Morphological Principles. Berlin: Mouton de Gruyter. Oxford Dictionary of New Words. 1991. Compiled by Sara Tullock. Oxford: OUP. Oxford English Dictionary, 2nd edition online. . Pickett, Joseph P. (ed.). 2001. The American Heritage Dictionary of the English Language, 4th edn. New York NY: Random. Walker, J. 1936. The Rhyming Dictionary of the English Language. Revised and enlarged by Lawrence H. Dawson. New York NY: Dutton. Webster’s Encyclopedic Unabridged Dictionary of the English Language. 1989. New York NY: Gramercy. Wood, Clement. 1943. Wood’s Unabridged Rhyming Dictionary. Cleveland OH: The World.

II.

Secondary sources

Aarts, Jan (ed.). 1984. Corpus Linguistics: Recent Developments in the Use of Computer Corpora in English Language Research. Amsterdam: Rodopi. Aitchison, Jean. 1999. Language at war: We’ve altered the word refugee to deportee. There is a reason, The Guardian, April 9, 1999, <www.guardian.co.uk/world/1999/apr/09/balkans, retrieved March, 21, 2005>. Aitchison, Jean. 1981. Language Change: Progress or Decay? London: Fontana. Anderson, Stephen R. 1992. A-Morphous Morphology. Cambridge: CUP.

194 Heterogeneity in Word-Formation Patterns

Aronoff, Mark. 1983. Potential words, actual words, productivity and frequency. In Proceedings of the XIII International Congress of Linguists, S. Hattori & K. Inoue (eds), 163–171. Tokio. Aronoff, Mark. 1976. Word Formation in Generative Grammar. Cambridge MA: The MIT Press. Aronoff, Mark & Anshen, Frank. 1998. Morphology and the lexicon: Lexicalization and productivity. In The Handbook of Morphology, Andrew Spencer & Arnold M. Zwicky (eds), 237–247. Oxford: Blackwell. Asirvatham, Arul Prakash & Ravi, Kranthi Kumar. 2001. Web page classification based on document structure. <web.iiit.ac.in/~arul/paper.pdf> (September 11, 2009). Aston, Guy & Burnard, Lou. 1998. The BNC Handbook: Exploring the British National Corpus with SARA. Edinburgh: EUP. Baayen, Harald. 1993. On frequency, transparency, and productivity. In Yearbook of Morphology 1992, Gert E. Booij & Jaap van Marle (eds), 181–208. Dordrecht: Kluwer. Baayen, Harald & Lieber, Rochelle. 1991. Productivity and English derivation: A corpus-based study. Linguistics 29: 801–843. Baeskow, Heike. 2002. Abgeleitete Personenbezeichnungen im Deutschen und Englischen. Berlin: Walter de Gruyter. Bailey, Richard W. 1992. Images of English. A Cultural History of the Language. Ann Arbor MI: The University of Michigan Press. Baker, Sidney. 1945. The Australian Language. An Examination of the English Language and English Speech as Used in Australia, from Convict Days to the Present, with Special Reference to the Growth of Indigenous Idiom and its Use by Australian Writers. Sydney: Angus and Robertson. Barker, Chris. 1998. Episodic -ee in English: A thematic role constraint on new word formation. Language 74 (4): 695–727. Bauer, Laurie. 2001. Morphological Productivity [Cambridge Studies in Linguistics]. Cambridge: CUP. Bauer, Laurie. 1994. Watching English Change. An Introduction to the Study of Linguistic Change in Standard Englishes in the Twentieth Century. London: Longman. Bauer, Laurie. 1993. More -ee words. American Speech 68(2): 222–224. Bauer, Laurie. 1988. Introducing Linguistic Morphology. Edinburgh: EUP. Bauer, Laurie. 1987. -ee by gum! American Speech 62(4): 315–319. Bauer, Laurie. 1983. English Word-formation [Cambridge Textbooks in Linguistics]. Cambridge: CUP. Baugh, Albert D. & Cable, Thomas. 2002. A History of the English Language (5th edn). London: Routledge. Bengtsson, Elna. 1927. Passive Nouns with a Concrete Sense in English. Lund: Häkan Ohlsson. Biber, Douglas, Conrad, Susan & Reppen, Randi. 1998. Corpus Linguistics. Investigating Language Structure and Use. Cambridge: CUP. Bolinger, Dwight L. 1941. Among the new words. American Speech 16(4): 306–309. Booij, Geert & Lieber, Rochelle. 2004. On the paradigmatic nature of affixal semantics in English and Dutch. Linguistics 42(2): 327–357. Botha, Rudolph P. 1968. The Function of the Lexicon in Transformational Generative Grammar. The Hague: Mouton. Brandl, Annette. 2002. Webangebote und ihre Klassifikation – Typische Merkmale aus Expertenund Rezipientenperspektive. München: R. Fischer.



Works cited 195

Bußmann, Hadumod. 1990. Lexikon der Sprachwissenschaft. Stuttgart: Alfred Kröner. Carlson, Greg N. 1991. Natural kinds and common nouns. In Semantik/Semantics, Arnim von Stechow & Dieter Wunderlich (eds), 370–398. Berlin: de Gruyter. Chomsky, Noam. 1970[1972]. Remarks on nominalization. Studies on Semantics in Generative Grammar, 11–61. The Hague: Mouton. Christie, Pauline. 2003. Language in Jamaica. Kingston: Arawak. Clark, Eve & Clark, Herbert. 1979. When nouns surface as verbs. Language 55: 767–811. Collins, Peter & Peters, Pam. 1988. The Australian corpus project. In Corpus Linguistics, Hard and Soft. Proceedings of the Eighth International Conference on English Language Research on Computerized Corpora, Merja Kytö, Ossi Ihalainen & Matt Rissanen (eds), 103–120. Amsterdam: Rodopi. Coseriu, Eugenio. 1967. Lexikalische Solidaritäten. Poetica 1: 293–303. Crystal, David. 1997. English as a Global Language. Cambridge: CUP. Crystal, David. 1991. A Dictionary of Linguistics and Phonetics, 3rd edn. Oxford: Blackwell. Dabke, Roswitha. 1976. Morphology of Australian English. München: Wilhelm Fink. Derwing, Bruce L. 1990. Morphology and the mental lexicon: Psycholinguistic evidente. In Contemporary Morphology [Trends in Linguistics Studies and Monographs 49], Wolfgang U. Dressler et al. (eds), 249–265. Berlin: Mouton de Gruyter. Derwing, Bruce L. & Skousen, Royal. 1989. Morphology in the mental lexicon: A new look at analogy. Yearbook of Morphology, 55–71. Dordrecht: Foris. Di Sciullo, Anne-Marie & Williams, Edwin. 1987. On the Definition of Word. Cambridge MA: The MIT Press. Elsen, Hilke. 2004. Neologismen: Formen und Funktionen neuer Wörter in verschiedenen Varietäten des Deutschen. Tübingen: Narr. Finkenstaedt, Thomas & Wolff, Dieter. 1973. Ordered Profusion. Studies in Dictionaries and the English Lexicon. Heidelberg: Winter. Firth, John Rupert. 1957. Modes of meaning. In Papers in Linguistics, 1934–1951, John Rupert Firth (ed.), 190–215. Oxford: OUP. Fischer, Roswitha. 1998. Lexical Change in Present-Day English: A Corpus-Based Study of the Motivation, Institutionalization and Productivity of Creative Neologisms. Tübingen: Narr. Fisher, John Hurt. 2001. British and American, continuity and divergence. In The Cambridge History of the English Language, Vol. IV: English in North America, John Algeo (ed.), 59–85. Cambridge: CUP. Ghani, Rayid, Jones, Rosie & Mladenic, Dunja. 2001. Mining the web to create minority language corpora. In Proceedings of the 10th International Conference on Information and Knowledge Management (CIKM), 2001, 279–286. Association for Computing Machinery. Görlach, Manfred. 1999. Regional and social variation. In The Cambridge History of the English Language, Vol. II: 1476–1776, Roger Lass (ed.), 459–538. Cambridge: CUP. Gold, David L. 1977. Queree: A correct formation? American Speech 52(1/2): 160. Greenbaum, Sidney. 1992. A new corpus of English: ICE. In Directions in Corpus Linguistics: Proceedings of Nobel Symposium 82, Stockholm 4–8 August 1991, Jan Svartvik (ed.), 171– 179. Berlin: Mouton de Gruyter. Gross, Steven. 2006. Natural Kind Terms. In Encyclopedia of Language and Linguistics Vol. 8, 2nd edn., Keith Brown (ed.), 492–496. Oxford: Elsevier. Gunn, Jeannie S. 1972. Change in Australian idiom. In Good Australian English and Good New Zealand English, George W. Turner (ed.), 47–63. Sidney: Reed Education.

196 Heterogeneity in Word-Formation Patterns

Halle, Morris. 1973. Prolegomena to a theory of word formation. Linguistic Inquiry 4: 3–16. Hohenhaus, Peter. 1996. Ad-hoc-Wortbildung: Terminologie, Typologie und Theorie kreativer Wortbildung im Englischen. Frankfurt: Peter Lang. Horn, Laurence R. 1980. Affixation and the unaccusative hypothesis. Proceedings of the Chicago Linguistics Society 16: 134–146. Horvath, Barbara M. 1985. Variation in Australian English. The Sociolects of Sydney. Cambridge: CUP. Hughes, Geoffrey. 2000. A History of English Words. Oxford: Blackwell. Hundt, Marianne, Biewer, Caroline & Nesselhauf, Nadja (eds). 2007. Corpus Linguistics and the Web. Amsterdam: Rodopi. Jackendoff, Ray. 1990. Semantic Structures. Cambridge MA: The MIT Press. Jackendoff, Ray. 1983. Semantics and Cognition. Cambridge MA: The MIT Press. Jackendoff, Ray. 1975. Morphological and semantic regularities in the lexicon. Language 51: 639–671. Johansson, Stig et al. 1986. The Tagged LOB Corpus: Users’ Manual, Bergen: Norwegian Computing Centre for the Humanities. Kastovsky, Dieter. 2006. Typological changes in derivational morphology. In The Handbook of the History of English, Ans van Kemenade & Bettelou Los (eds), 151–176. Oxford: Blackwell. Kastovsky, Dieter. 1986a. The problem of productivity in word formation. Linguistics 24 (3): 585–600. Kastovsky, Dieter. 1986b. Diachronic word-formation in a functional perspective. In Linguistics across Historical and Geographical Boundaries, Vol. 1: Linguistic Theory and Historical Linguistics, Dieter Kastovsky & Aleksander Swedek (eds), 409–421. Berlin: Mouton de Gruyter. Kastovsky, Dieter. 1978. Zum gegenwärtigen Stand der Wortbildungslehre des Englischen. Linguistik und Didaktik 36: 351–366. Katamba, Francis. 1994. English Words. London: Routledge. Kehoe, Andrew & Renouf, Antoinette. 2002. WebCorp: Applying the Web to linguistics and linguistics to the Web. Proceedings of the WWW 2002 conference. Honolulu HI. Kilgarriff, Adam & Grefenstette, Gregory. 2003. Introduction to the special issue on the web as corpus. Computational Linguistics 29: 333–347 (Special Issue on the Web as Corpus). Kiparsky, Paul. 1974. Remarks on analogical change. In Historical Linguistics II, John M. Anderson & Charles Jones (eds), 257–275. Amsterdam: Elsevier. Koziol, Herbert. 1937. Handbuch der englischen Wortbildungslehre. Heidelberg: Carl Winter. Kucera, Henry & Francis, W. Nelson. 1961. The Brown Corpus of Standard American English. Providence RI: Brown University. Labov, William. 1994. Principles of Linguistic Change, Vol. 1: Internal Factors. Oxford: Blackwell. Ladusaw, William A. & Dowty, David R. 1988. Toward a nongrammatical account of thematic roles in thematic relations. Syntax and Semantics, Vol. 21: Thematic Relations, 61–73. Lakoff, George. 1987. Women, Fire, and Dangerous Things. What Categories Reveal About the Mind. Chicago IL: University of Chicago Press. Lass, Roger. 1997. Historical Linguistics and Language Change. Cambridge: CUP. Lass, Roger. 1987. The Shape of English. Structure and History. London: J.M. Dent & Sons. Lees, Robert B. 1960. The Grammar of English Nominalizations. The Hague: Mouton.



Works cited 197

Leitner, Gerhard. 2004. Australia’s Many Voices. Australian English – The National Language. Berlin: Mouton de Gruyter. Lehrer, Adrienne. 1996. Why neologisms are important to study. Lexicology 2(1): 63–73. Levin, Beth & Rappaport Hovav, Malka. 1995. Unaccusativity at the Syntax-Lexical Semantics Interface. Cambridge MA: The MIT Press. Levin, Beth & Rappaport, Malka. 1988. Nonevent -er nominals: A probe into argument structure. Linguistics 26: 1067–1083. Levin, Beth & Rappaport, Malka. 1986. The formation of adjectival passives. Linguistic Inquiry 17: 623–662. Lieber, Rochelle. 2004. Morphology and Lexical Semantics. Cambridge: CUP. Lieber, Rochelle. 1992. Deconstructing Morphology. Chicago IL: University of Chicago Press. Lieber, Rochelle & Baayen, Harald. 1997. A semantic principle of auxiliary selection in Dutch. Natural Language and Linguistic Theory 15: 789–845. Lipka, Leonard. 2002. English Lexicology: Lexical Structure, Word Semantics & Word-formation. Tübingen: Narr. Lipka, Leonhard. 1975. Re-discovery procedures and the lexicon. Lingua 37(2): 197–224. Lüdeling, Anke, Evert, Stefan & Baroni, Marco. 2007. Using Web data for linguistic purposes. In Corpus Linguistics and the Web, Marianne Hundt, Caroline Biewer & Nadja Nesselhauf (eds), 7–24. Amsterdam: Rodopi. Lyons, John. 1977. Semantics. Cambridge: CUP. Lyons, John. 1995. Linguistic Semantics: An Introduction. Cambridge: CUP. Mair, Christian. 2004. Corpus linguistics and grammaticalisation theory: Statistics, frequencies, and beyond. In Corpus Approaches to Grammaticalization in English, Hans Lindquist & Christian Mair (eds), 121–150. Amsterdam: John Benjamins. Manning, Christopher & Schütze, Hinrich. 1999. Foundations of Statistical Natural Language Processing. Cambridge MA: The MIT Press. Marchand, Hans. 1969. The Categories and Types of Present-Day English Word-Formation. A Synchronic-Diachronic Approach. München: C.H. Beck. Mathews, M. M. 1945. The new element in American English (II). American Speech 20(2): 106–110. Matthews, Peter H. 1974. Morphology. Cambridge: CUP. McEnery, Tony & Wilson, Andrew. 1996. Corpus Linguistics. Edinburgh: EUP. Meyer, Charles F. 2002. English Corpus Linguistics: An Introduction. Cambridge: CUP. Motsch, Wolfgang. 1977. Ein Plädoyer für die Beschreibung von Wortbildungen auf der Grundlage des Lexikons. In Perspektiven der Wortbildungsforschung Herbert E. Brekle & Dieter Kastovsky (eds),180–202. Bonn: Bouvier Verlag Herbert Grundmann. Motsch, Wolfgang. 1981. Der kreative Aspekt in der Wortbildung. In Wortbildung, Leonhard Lipka & Hartmut Günther (eds), 94–118. Darmstadt: Wissenschaftliche Buchgesellschaft. Mühleisen, Susanne. 2007. Of lessees, retirees and beseeches: A corpus-based analysis of the productivity of a heterogeneous word-formation pattern in English. In Anglistentag 2006 Halle. Proceedings, Sabine Volk-Birke & Julia Lippert (eds), 413–422. Trier: WVT. Nevalainen, Terttu. 1999. Early Modern lexis and semantics. In The Cambridge History of the English Language, Vol. II: 1476–1776, Robert Lass (ed.), 332–458. Cambridge: CUP. Oesterreicher, Wulf. 1999. Ad-hoc-Formulierungen als Herausforderung für Lexikologie und Lexikographie. In Words, Lexemes, Concepts – Approaches to the Lexicon. Studies in Honour of Leonhard Lipka, Wolfgang Falkner & Hans-Jörg Schmid (eds), 69–84 Tübingen: Narr.

198 Heterogeneity in Word-Formation Patterns

Ong, Walter J. 1982. Orality and Literacy. The Technologizing of the Word. London: Routledge. Panten, Günter. 1959. Die Amerikanismen im ‘Manchester Guardian Weekly’ (1948–1954). Ein Beitrag zur Funktion der englischen Zeitung in den amerikanisch-britischen Sprachbeziehungen der Gegenwart. München: Max Hueber. Pennycook, Alistair. 2003. Beyond homogeny and heterogeny: English as a global and worldly language. In The Politics of English as a World Language. New Horizons in Postcolonial Cultural Studies, Christian Mair (ed.), 3–17. Amsterdam: Rodopi. Perlmutter, David. 1978. Impersonal passives and the unaccusative hypothesis. Proceedings of the Berkeley Linguistics Society 4: 157–189. Plag, Ingo. 2006. Productivity. In Encyclopedia of Language and Linguistics, Keith Brown (ed.), 21–128. Amsterdam: Elsevier. Plag, Ingo. 2004. Syntactic category information and the semantics of drivational morphological rules. Folia Linguistica 38(3–4): 193–225. Plag, Ingo. 2003. Word-Formation in English [Cambridge Textbooks in Linguistics]. Cambridge: CUP. Plag, Ingo. 1999. Morphological Productivity. Structural Constraints in English Derivation. Berlin: Mouton de Gruyter. Plag, Ingo. 1998. The polysemy of -ize derivatives. The role of semantics in word formation. In Yearbook of Morphology 1997, Geert Booij & Jaap van Marle (eds), 219–242. Dordrecht: Foris. Portero Muñoz, Carmen. 2003. Derived nominalizations in -ee: A Role and Reference Grammar based semantic analysis. English Language and Linguistics 7(1): 129–159. Pounder, Amanda. 2000. Processes and Paradigms in Word-Formation Morphology. Berlin: Mouton de Gruyter. Price, Jennie. 2003. The recording of vocabulary from the major varieties of English in the Oxford English Dictionary. In The Politics of English as a World Language [Cross/Cultures 65], Christian Mair (ed.), 119–127. Amsterdam: Rodopi. Quirk, Randolph, Greenbaum, Sidney, Leech, Geoffrey & Svartvik, Jan. 1972. A Grammar of Contemporary English. London: Longman. Raffelsiefen, Renate. 1999. Phonological constraints on English word-formation. In Yearbook of Morphology 1998, Geert Booij & Jaap van Marle (eds), 225–287. Dordrecht: Kluwer. Ramson, William Stanley. 1966. Australian English. A Historical Study of the Vocabulary 1788– 1898. Canberra: Australian National University Press. Rappaport Hovav, Malka & Levin, Beth. 1992. -Er nominals: Implications for the theory of argument structure. In Syntax and Semantics, Vol. 26: Syntax and the Lexicon, Tim Stowell & Eric Wehrli (eds), 127–53. San Diego CA: Academic Press. Rehm, Georg. 2004. Hypertextsorten-Klassifikation als Grundlage generischer Informations­ extraktion. In Automatische Text­analyse – Systeme und Methoden zur Annotation und Analyse natürlichsprach­licher Texte, Alexander Mehler & Henning Lobin (eds), 219–233. Wiesbaden: Verlag für Sozialwissenschaften. Rehm, Georg. 2002. Schriftliche Mündlichkeit im World Wide Web. In Kommunikationsform E-mail, Arne Ziegler & Christa Dürscheid (eds), 263–308. Tübingen: Stauffenburg. Resnik, Philip & Smith, Noah A. 2003. The web as a parallel corpus. Computational Linguistics 29: 349–380 (Special Issue on the Web as Corpus). Rissanen, Matti, et al. 1997. Introduction. In English in Transition. Corpus-based Studies in Linguistic Variation and Genre Styles, Matti Rissanen, Merja Kytö & Kirsi Heikkonen (eds), 1–15. Berlin: Mouton de Gruyter.



Works cited 199

Romaine, Suzanne. 1998. Introduction. In The Cambridge History of the English Language, Vol. IV: 1776–1997, Suzanne Romaine (ed.), 1–56. Cambridge: CUP. Schmid, Hans-Jörg. 2005. Englische Morphologie und Wortbildung. Berlin: Erich Schmidt. Schröder, Anne. 2008. On the Productivity of Verbal Prefixation in English. Habilitationsschrift, Martin-Luther-Universität Halle-Wittenberg. Schröder, Anne & Mühleisen, Susanne. In press. New ways of investigating morphological productivity. Arbeiten aus Anglistik und Amerikanistik (AAA) 35(1). Schultink, Henk. 1961. Produktiviteit als morfologisch fenomeen. Forum der Letteren 2: 110– 125. Sebba, Mark. 1997. Contact Languages. Pidgins and Creoles. Houndmills: Macmillan. Shastri, S. V. 1988. The Kolhapur Corpus of Indian English and work done on its basis so far. ICAME Journal 12: 15–26. Simpson, Jane. 2004. Hypocoristics in Australian English. In A Handbook of Varieties of English, Vol. 2: Morphology and Syntax, Bernd Kortmann et al. (eds), 643–656. Berlin: Mouton de Gruyter. Skousen, Royal. 1989. Analogical Modeling of Language. Dordrecht: Kluwer. Strang, Barbara M. 1970. A History of English. London: Methuen. Svartvik, Jan (ed.). 1990. The London-Lund Corpus of Spoken English. Lund: Lund University Press. Thiel, Gisela. 1973. Die semantischen Beziehungen der Substantivkomposita der deutschen Gegenwartssprache. Muttersprache 83: 377–404. Van Valin, Robert D. & Foley, William A. 1980. Role and reference grammar. In Syntax and Semantics, Vol. 13: Current Approaches to Syntax, Edith A. Moravscsik & Jessica R. Wirth (eds), 329–52. New York NY: Academic Press. Van Valin, Robert D. & LaPolla, Randy. 1997. Syntax: Structure, Meaning and Function. Cambridge: CUP. Webster, Noah. 1951[1789]. Dissertations on the English Language, facsimile. Gainesville FL: Scholars’ Facsimiles and Reprints. Wermser, R. 1976. Statistische Studien zur Entwicklung des Englischen Wortschatzes. Bern: Francke. Xu, J. L. 2000. Multilingual search on the World Wide Web. Proceedings of the Hawaii International Conference on System Science (HICSS-33). Maui HI. Zwicky, Arnold M. & Pullum, Geoffey K. 1987. Plain morphology and expressive morphology. Proceedings of the Thirteenth Annual Meeting of the Berkeley Linguistic Society, 330–340. Berkeley CA: University of California Press.

appendix 1

Documentation of established -ee words with their citation sources A comparison (in alphabetical order)

Sources: 1. Oxford English Dictionary. 2. Walker, J. 1936. The Rhyming Dictionary of the English Language. Revised and enlarged. New York: Dutton. 3. Wood, Clement. 1943. Wood’s Unabridged Rhyming Dictionary. Cleveland: The World Publ. Co. 4. Lehnert, Martin. 1971. Rückläufiges Wörterbuch der englischen Gegenwartssprache. Leipzig: VEB Verlag Enzyklopädie. 5. Various sources in American Speech (Bolinger 1941; Gold 1977; Mathews 1945); Kastovsky (1986b); Barnhart (1991). 6a. Bauer, Laurie. 1983. English Word-formation. Cambridge Textbooks in Linguistics. Cambridge: CUP (cites various sources). 6b. Bauer, Laurie. 1987. -ee by gum! American Speech 62 (4): 315–319. 6c. Bauer, Laurie. 1993. More -ee words. American Speech 68 (2): 222–224. 7. Barker, Chris. 1998. Episodic -ee in English: A thematic role constraint on new word formation. Language 74 (4): 1–33. 8. Muthmann, Gustav. 1999. Reverse Dictionary. Based on Phonological and Morphological Principles. Berlin: Mouton de Gruyter.

. Almost all of the words given in Bauer (1987) and Bauer (1993) are reiterated in a word list in Bauer (1994: 42–45). Bauer (1994) is therefore not given a separate column.

202 Heterogeneity in Word-Formation Patterns

abandonee abductee abortee absentee abusee acquiree actee adaptee addressee admittee adoptee advancee advertisee advisee advowee [affrontee] aggressee alienee allocatee allottee amputee amusee appellee appointee appraisee arrestee arrivee ascendee assessee assignee asylee attendee auditionee avowee bailee banteree baptizee bargainee [bargee] basaree beatee

1

2

3

4

X X X X X X

X

X

X

X

X

5

6a, b &c

X

X

X

X

X X

X

X

X X X X

X

X

X X X X X

X

X

X X X X X X

X X X

X X

X X

X

X

X X

X

X

X X X X X

X X X X X X

X X

X

X X X X X X X

8

X

X

X X X

7

X

X X

X

X X

X X X

X X X X X X X X X X X X X X X X X

X X X

X



Appendix 1 203

1 beateree befriendee benefactee best wishee billetee biographee bitee blackmailee blind datee bombee boree borrowee bowee boycottee bribee bumpee callee canee catapultee catchee causee challengee chargee charteree chasee chattee cheatee cheekee citee civilizee cliticee cloutee coachee cognizee cohabitee co-mortgagee committee commissionee communicatee communicee complainee

2

3

4

5

6a, b &c

7

X X

X X

X X X X

X

X

X

X

X X

X

X X X

X X X X X X X X X X

X X X

X X

X

X X

X

X X X

X X

X

X

X X

X X X X X X X X

X X X

X

X

X X

X X X X X X X

X

X

X X X

X

X

X X

8

204 Heterogeneity in Word-Formation Patterns

1 complimentee compromisee compromittee conferee conferree confessee confidee confirmee congratulatee conjuree conscriptee consecratee consignee constipatee consultee contactee contestee contractee controllee conusee convenee co-assignee co-mortgagee co-optee co-patentee cotrustee counsellee crack-upee crammee creditee cudgellee curee cursee custodee cuttee deadee debauchee debtee dedicatee defendee deferee

2

3

4

5

6a, b &c

7

8

X X X X X X X

X X

X

X

X

X X X X

X X

X

X

X X X X

X X X

X X

X X

X

X X

X X X

X

X X X X X X X X

X X X

X

X

X X

X

X X

X X X X X

X X X X

X X X X

? X

X X

X

X

X X

X

X X



Appendix 1 205

1 deferree delegatee deletee deportee depositee depurgee describee designee destinee detainee determinee devisee devotee dictatee dilutee dinee directee dischargee discontentee discontinuee dislocatee disponee disposee disseisee disseizee distinguee distrainee distributee divorcee donatee donee draftee dragee drainee drawee drivee dumpee editee educatee ejectee electee

2

3

4

5

6a, b &c

X X X X

X

X X X

X X

X X X X

X

X X

X X

X X

X X

X X

7 X X X X X X X X X X

8

X

X

X X X

X X X X X X X X X X

X

X

X

X X

X X X

X

X X X X X X

X

X X

X

X

X X

X

X X X X X X

X

X X

X X X

X

X

X X X X X X X

X X

X X X

X

X

206 Heterogeneity in Word-Formation Patterns

1 eliminatee eliminee embarkee employee endorsee engagee enlistee ennuyee enrollee entertainee erasee escapee espousee evacuee evictee examinee exchangee executionee expellee experimentee expiree explainee exploitee exposee extractee feedee feoffee festschriftee fiancee fillee financee firee flirtee floggee followee foolee forcee forgettee franchisee fuckee gaggee

2

3

4

5

6a, b &c

7

8

X

X X X X

?

X X X X X

X X

X

X X

X

X X X

X

X

X

X

X

X X X X X X

X

X

X

X

X X

X X

X

X

X

X X X X X X X X X

X

X X

X X X

X X

X X

X X X X X

X

X

X X (fiancé)

X

X X

X X X X X

X

X X

X X

X

X X

X

X

X X

X X X X X



Appendix 1 207

1 galee garnishee gazee geggee giftee gougee governee grantee guarantee guidee hackee handshakee hangee harpoonee haulee hazee head-bonkee hecklee hiree hittee hoaxee holdupee honoree huggee illuminee importee indemnitee indictee indoctrinee indorsee inductee infiltree inquisitee insuree interessee internee interrogatee interrogee interviewee introducee invadee

X X X X

2

3

4

5

6a, b &c

X X

X

7

8

X X X X X

X X X X

X X

X X

X X X

X X

X

X X X

X X X X X

X X X X X X X

X X X

X X X X X

X X

X X

X X

X X

X X

X

X X X

X X X X

X X

X X

X

X

X

X X

X X X X X

X

X X

208 Heterogeneity in Word-Formation Patterns

1 investee invitee jestee jiltee jokee kickee kidnapee kissee knockee laryngectomee laughee leakee lecturee legatee lessee letteree libelee libellee liberee licencee licensee likee loanee locatee lovee luggee lunchee magnetizee malefactee managee mancipee mandatee manipulatee manipulee markee meetee mentee mergee mesmeree mesmerizee missionee

X X X X X X

2

3

4

5

X X X

6a, b &c

7

X X

X X

X X

X

X X X

X

8

X X

X X X

X X

X X X X

X X

X

X

X

X

X X

X

X X

X X

X

X X X X X

X X

X X

X

X X

X X X X

X X

X X X

X X X X X X X X X

X

X

X X

X X X X X X

X X

X

X

X

X

X X X

X X



moneylendee mortgagee muggee murderee narratee noddee nominee objectee obligee offendee offeree operatee operee optionee ordinee pardonee parkee parolee party nominee passee patentee pawnee payee permittee persecutee petitionee philantropee photographee pickpocketee piedpipee pillagee pilotee placee plannee pleasee pledgee plottee pluckee politicee politico-politicee pollee

Appendix 1 209

1

2

3

4

X X X X X X X X

X

X

X

5

6a, b &c

7

8

X X

X

X X

X

X X X

X X X

X X X

X X

X

X

X X X

X X

X X X X

X

X X X X X

X

X X X X X X

X X X

X X

X

(X) check X X X X X

X

X X X

X X X

X

X X

X

X

X X X X X X

X

X X X X

X X

X X X

X

X

X

X

X

X

X X X X

X

X

210 Heterogeneity in Word-Formation Patterns

1 possessee prayee preachee preceptee predecree preferee presentee promisee promotee protegee protectee provokee publishee pumpee punchee puntee purgee queree questionee quizzee quotee raisee rapee rappee razee readee readjustee recognizee reconcilee recoveree recruitee reduncantee referee refugee rehabilitee rejectee relaxee releasee relessee remittee remuneratee

X X X X X X X X X (é) X X

2

3

4

5

6a, b &c

7

X

X

8

X

X X

X X

X X

X X

X X X

X

X X

X

X

X X X

X X

X X

X X X

X X X

X X

X X X

X X X

X X X

X

X X X

X X X X

X ?

X X

X X

X X X X

X

X

X

X

X

X

X

X X X

X X



Appendix 1 211

1 rentee reorderee representee rescuee rescussee reservee resignee restrictee retardee retiree retrainee return addressee returnee revisee revoltee roastee rubbee rushee sayee scendee secondee seducee selectee sendee separatee sequestree settee(?) shadowee sharkee shavee shootee signalee signallee signee sitee sittee slanderee slaughteree slittee smackee snatchee

2

3

4

5

6a, b &c

7

X

X X

8

X X X X X X X X X

X X X

X X

X X X

X

X X

X

X X X X X X X X X X

X X X

X

X X X X X X

X X X X ?

X

X

X X X X

X X X

X X X X

X

X X

X X X X X

X

X X

X X

X X X X X X

X X X

X

X X

212 Heterogeneity in Word-Formation Patterns

1 snubbee socializee sockee solicitee sornee squeezee stagee standee staree stickee sublessee submittie subornee supervisee surrenderee suspendee seatee staree sweatee tailee takee talkee tastee teasee telephonee tenderee testee throwee tipee tippee toastee toppee torturee trainee transferee transplantee transportee treatee

2

3

4

5

6a, b &c

X X

7

8

X X

X X X

X X

X

X X X X X X X X X

X

X X X

X

X

X

X

X X X X

X X X X X

X X

X X X X X

X X X

X X

X2

X

X

X

X X

X X

X X X X X X

2. In Bauer (1994: 45).

X

X

X X

X

X

X



trottee trustee tryoutee tutee twistee usee underlessee vaccinee vauchee vendee venerealee vestee visitee vivisectee vouchee vowee waitee wardee warrantee wishee withstandee writee yellee

Appendix 1 213

1

2

X X

X

3

X

6a, b &c X

7 X X X X X

8

X

X X X X

X

X

X X

X X X X X X X

5

X

X X X X X X

4

X

X X X X

X X X

X

X X X X X X X

X

X

X

X

X

X

appendix 2

Quantitative analysis of 1,000 potential -ee words (Web-search, February–June 2005)

NF = Not found HL = Hapax Legomenon (1 or max. 2 tokens) Rare = up to 20 tokens Established = 21 to 100 tokens Frequent = more than 100 tokens NQ = found but not quantifiable (e.g. because of competing French lexical items, names, brand names, etc.) No. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.

Lexical item

NF

abhorree abidee abjectee abolishee abrogatee absorbee acceptee acclaimee accommodatee accursee accusee acerbatee acknowledgee acquittee adaptee addictee adjustee admiree admonishee adoree adulteree aggroupee agitatee

X

HL

Rare

Est.

Frequ.

X X X X X X X X X X X X X X X X X X X X X X

NQ

216 Heterogeneity in Word-Formation Patterns

No. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65.

Lexical item

NF

agonizee alertee allegee amazee amendee analysee analyzee animatee annectee anointee answeree anticipatee apologee apologizee applaudee approvee arousee ashamee aspiree assassinee assaultee assuree astrologee attractee auctionee authorizee avengee avoidee awaitee awardee badgee baitee bangee barkee batee beggee bellowee bereavee beseechee besmearee besmirchee bewailee

X

HL

Rare

Est.

Frequ.

NQ X

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 217

No. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107.

Lexical item

NF

bewitchee blamee blessee blowee bluffee boastee bouncee boxee bruisee brushee burglaree buskeree butcheree calmee capitalizee caree caressee castee censoree checkee cheeree chuckee circumcisee claimee clappee claspee cleanee cleansee clubbee clutchee coercee collectee colonizee coloree combatee comfortee commandee commemoratee commentee compactee compensatee computee

X

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

218 Heterogeneity in Word-Formation Patterns

No. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149.

Lexical item

NF

concealee confinee confiscatee confrontee congratulee conqueree consentee consideree consolee consortee contaminatee contaminee contradictee contemnee contritee convertee convictee correctee correspondee corruptee coveree crapee creditee criticizee cuddlee cullee cussee damagee dancee (lap dancee) daree dashee datee dealee deceivee declaree decodee defacee defamee defloweree defyee degradee demeanee

X

HL

Rare

Est.

Frequ.

NQ X

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 219

No. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191.

Lexical item dementee demisee demolishee demoralizee demotee denigratee denominatee denominee denouncee dependee depravee deprecatee depressee deprivee deridee deservee desiree despairee despisee despoilee detectee detestee detractee devastatee devouree dickeree dietee dippee disappointee disapprovee disarmee disavowee disbandee disbelievee discardee disciplinee disclaimee disclosee discommodee discomposee disconnectee discouragee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

220 Heterogeneity in Word-Formation Patterns

No. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228. 229. 230. 231. 232. 233.

Lexical item

NF

discreditee discriminatee disgracee disgustee disheartenee dishonoree dishonouree disinheritee disintegratee dislikee dislodgee dismemberee dismissee disobligee disownee disparagee dispiritee displacee displeasee disploree dispossessee disprovee disputee disregardee disrobee disruptee dissentee dissociatee distortee distractee distressee distrustee disusee(e) ditchee dockee domesticatee dominatee doubtee dreadee dreamee drillee droppee

X

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 221

No. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259 260. 261. 262. 263. 264. 265. 266. 267. 268. 269. 270. 271. 272. 273. 274. 275.

Lexical item drownee druggee duckee eatee elevatee elopee eludee emaciatee emailee emancipatee embalmee embarassee empoweree enactee encouragee encroachee endearee enfeeblee enforcee engrossee enhancee enlightenee ennoblee enragee (loanword) enslavee entanglee enthralee enthrallee enticee entitlee entrustee envyee erodee eschewee escortee espionagee estrangee eulogizee evadee evaluatee exacerbatee excitee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

222 Heterogeneity in Word-Formation Patterns

No. 276. 277. 278. 279. 280. 281. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 296. 297. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 308. 309. 310. 311. 312. 313. 314. 315. 316. 317.

Lexical item excludee excommunicatee exculpatee excusee executee exercisee exhibitee exhumee exilee exoneree exorcizee expectee expelee explodee exportee expressee expropriatee expulsee extortee eyee (not usable) facee failee faintee falteree fancyee fantasizee favoree favouree fearee fencee fetchee fetteree finee flatteree floatee floutee flunkee flushee followee-up fondlee forbidee foretellee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

X

X X X X X X X X X X X X X



Appendix 2 223

No. 318. 319. 320. 321. 322. 323. 324. 325. 326. 327. 328. 329. 330. 331. 332. 333. 334. 335. 336. 337. 338. 339. 340. 341. 342. 343. 344. 345. 346. 347. 348. 349. 350. 351. 352. 353. 354. 355. 356. 357. 358. 359.

Lexical item forgee forgivee forsakee forwardee fosteree framee frequentee frownee frustratee fumblee fundee furtheree fussee gapee garottee gatheree gawpee gigglee givee glancee glaree gloatee glorifyee gogglee gossipee greetee grievee gripee grippee groinee gropee gullee hackee hagglee hailee hamperee harassee hatee hauntee healee hearee heedee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

224 Heterogeneity in Word-Formation Patterns

No. 360. 361. 362. 363. 364. 365. 366. 367. 368. 369. 370. 371. 372. 373. 374. 375. 376. 377. 378. 379. 380. 381. 382. 383. 384. 385. 386. 387. 388. 389. 390. 391. 392. 393. 394. 395. 396. 397. 398. 399. 400. 401.

Lexical item helpee hidee hijackee hissee hitchee hobnobee holleree honkee hookee hospitalizee housee hoveree huddlee huffee humiliatee huntee hurtee hushee hustlee identifyee ignoree ill wishee imaginee imitatee immunee immunizee impairee impeachee impersonatee imploree impostee inauguree incarnatee inclinee inconveniencee incriminatee indebtee indicatee indulgee infatuatee infectee infiltratee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 225

No. 402. 403. 404. 405. 406. 407. 408. 409. 410. 411. 412. 413. 414. 415. 416. 417. 418. 419. 420. 421. 422. 423. 424. 425. 426. 427. 428. 429. 430. 431. 432. 433. 434. 435. 436. 437. 438. 439. 440. 441. 442. 443.

Lexical item influencee infringee ingratiatee initiatee injectee injuree inscribee insinuatee inspectee inspiree instigatee instructee insultee interactee interferee interpretee intervenee intimidatee investigatee invokee involvee isolatee jeeree jeopardizee jerkee jihadee keepee kiddee knockee knifee labelee labellee lamentee lashee laudatee launchee leechee leeree lendee liaisee lickee lieee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

226 Heterogeneity in Word-Formation Patterns

No. 444. 445. 446. 447. 448. 449. 450. 451. 452. 453. 454. 455. 456. 457. 458. 459. 460. 461. 462. 463. 464. 465. 466. 467. 468. 469. 470. 471. 472. 473. 474. 475. 476. 477. 478. 479. 480. 481. 482. 483. 484. 485.

Lexical item linkee listenee listee loathee lockee lodgee loiteree lookee lootee lubricatee luree lurkee lustee lullee lynchee mailee maimee maledictee manicuree maskee masteree meddlee mediatee medicatee meltee migratee milkee mimicee mimickee mincee misdirectee misguidee misinterpretee misnamee misplacee misquotee misrepresentee mistrustee mobbee mockee modernizee moralizee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 227

No. 486. 487. 488. 489. 490. 491. 492. 493. 494. 495. 496. 497. 498. 499. 500. 501. 502. 503. 504. 505. 506. 507. 508. 509. 510. 511. 512. 513. 514. 515. 516. 517. 518. 519. 520. 521. 522. 523. 524. 525. 526. 527.

Lexical item mournee mutilatee mystifyee naggee namee nativizee naturalizee navigatee neckee needee neglectee negotiatee notifyee nursee nurturee observee obsessee obstructee omittee onlookee opinionee opportunee opposee oppressee organizee orientee ostracizee outcastee outee outplacee outragee outsizee outwittee ovationee overbiddee overcompensatee overdressee overnightee overpoweree overpraisee overridee overrulee

NF

HL

Rare

Est.

Frequ.

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

NQ

228 Heterogeneity in Word-Formation Patterns

No. 528. 529. 530. 531. 532. 533. 534. 535. 536. 537. 538. 539. 540. 541. 542. 543. 544. 545. 546. 547. 548. 549. 550. 551. 552. 553. 554. 555. 556. 557. 558. 559. 560. 561. 562. 563. 564. 565. 566. 567. 568. 569.

Lexical item

NF

overseee overthrowee overwhelmee pacifyee packee paddlee paintee panderee panicee pastee peckee pedicuree peepee peltee penalizee perfectionee performee perfumee perishee perjuree perpetratee perplexee persiflagee persistee personatee persuadee perturbee phonee pickee piercee pilee pimpee pinchee pindownee pissee pitchee placatee plaitee plantee plasteree platformee playee

X

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 229

No. 570. 571. 572. 573. 574. 575. 576. 577. 578. 579. 580. 581. 582. 583. 584. 585. 586. 587. 588. 589. 590. 591. 592. 593. 594. 595. 596. 597. 598. 599. 600. 601. 602. 603. 604. 605. 606. 607. 608. 609. 610. 611.

Lexical item

NF

pleadee plightee plonkee pluggee plunderee plungee plushee poachee pointee poisonee pokee policee polarizee polishee pollinatee polutee pollutee ponderee posee positionee postee postponee postulatee pouncee poundee practisee praisee prankee precognizee preconceivee preconsultee predestinee predeterminee predictee predominatee prejudicee preparee prepossessee preselectee pressuree presumee preventee

X X

HL

Rare

Est.

Frequ.

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

NQ

230 Heterogeneity in Word-Formation Patterns

No. 612. 613. 614. 615. 616. 617. 618. 619. 620. 621. 622. 623. 624. 625. 626. 627. 628. 629. 630. 631. 632. 633. 634. 635. 636. 637. 638. 639. 640. 641. 642. 643. 644. 645. 646. 647. 648. 649. 650. 651. 652. 653.

Lexical item pretendee printee privilegee probee proclaimee procrastinatee procuree proddee prodee professee profferee profitee prognosee prohibitee projectee prolongee promptee pronouncee propagatee proposee proscribee prosecutee proselytee protestee providee publicizee puffee pummelee punishee punnee pursuee puzzlee quackee quailee qualifyee quarrellee quellee quenchee questee quibblee quittee rackee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 231

No. 654. 655. 656. 657. 658. 659. 660. 661. 662. 663. 664. 665. 666. 667. 668. 669. 670. 671. 672. 673. 674. 675. 676. 677. 678. 679. 680. 681. 682. 683. 684. 685. 686. 687. 688. 689. 690. 691. 692. 693. 694. 695.

Lexical item ragee raidee ratee ravee reaffirmee reassertee reassuree rebootee rebuffee rebukee reclaimee reclinee recoilee recommendee recompensee recomposee recordee recreatee recriminatee recruitee recuperatee redeemee redressee refinee reformee refusee regardee regeneratee rehearee rehearsee reincarnatee reinstatee rejoicee rejoinee rejuvenatee rekindlee relapsee relegatee relocatee remandee remarkee remarryee

NF

HL

Rare

Est.

Frequ.

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

NQ

232 Heterogeneity in Word-Formation Patterns

No. 696. 697. 698. 699. 700. 701. 702. 703. 704. 705. 706. 707. 708. 709. 710. 711. 712. 713. 714. 715. 716. 717. 718. 719. 720. 721. 722. 723. 724. 725. 726. 727. 728. 729. 730. 731. 732. 733. 734. 735. 736. 737.

Lexical item rememberee remindee remissee remonstratee remorsee removee renderee renouncee renunciatee repatriatee repayee repentee replacee reportee repossessee reprievee reprisee reproachee reprobatee reputee requestee requiree researchee resemblee resentee resistee respectee respondee restoree restrainee resurrectee retainee retaliatee retortee retracee retractee retreatee retrievee retrogressee reunitee revengee reveree

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 233

No. 738. 739. 740. 741. 742. 743. 744. 745. 746. 747. 748. 749. 750. 751. 752. 753. 754. 755. 756. 757. 758. 759. 760. 761. 762. 763. 764. 765. 766. 767. 768. 769. 770. 771. 772. 773. 774. 775. 776. 777. 778. 779.

Lexical item reversee revertee reviewee revilee revitalizee revivee revokee rewardee ridee ridiculee ringee rinsee riotee robbee ropee roughee rowee ruinee rulee rumoree rumouree sackee sailee salutee satisfyee saturatee savee scandalee scaree scoffee scoldee scoree scourgee scoutee scowlee scratchee screamee screechee screwee scrubbee scrutinee sculptee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

234 Heterogeneity in Word-Formation Patterns

No. 780. 781. 782. 783. 784. 785. 786. 787. 788. 789. 790. 791. 792. 793. 794. 795. 796. 797. 798. 799. 800. 801. 802. 803. 804. 805. 806. 807. 808. 809. 810. 811. 812. 813. 814. 815. 816. 817. 818. 819. 820. 821.

Lexical item searchee securee sedatee segregatee self-deprecatee self-sacrificee self-servee sequencee sermonee servee shaggee shakee shamee shelteree shieldee shittee shockee shortlistee shoutee showee shruggee singee sketchee slappee slayee sledgee slightee slippee slobbee slouchee smashee smilee smirkee smokee smotheree smudgee snarlee sneeree sneezee sniffee sniggeree snivellee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 235

No. 822. 823. 824. 825. 826. 827. 828. 829. 830. 831. 832. 833. 834. 835. 836. 837. 838. 839. 840. 841. 842. 843. 844. 845. 846. 847. 848. 849. 850. 851. 852. 853. 854. 855. 856. 857. 858. 859. 860. 861. 862. 863.

Lexical item snoree soakee sobbee soothee spankee sparee speakee spellee spendee spicee spinnee spiritualizee spitee spittee splashee splittee spoilee spongee spuree sputteree squashee squattee squealee squelchee stabbee stakee stalkee stampee startlee starvee stealee sterilizee stingee stinkee stonee stoppee strandee stranglee strappee stretchee strikee stringee

NF

HL

Rare

Est.

Frequ.

X X X X X X X X X X X X X X X X X X X (?) X X X X X X X X X X X X X X X X X X X X X X X

NQ

236 Heterogeneity in Word-Formation Patterns

No. 864. 865. 866. 867. 868. 869. 870. 871. 872. 873. 874. 875. 876. 877. 878. 879. 880. 881. 882. 883. 884. 885. 886. 887. 888. 889. 890. 891. 892. 893. 894. 895. 896. 897. 898. 899. 900. 901. 902. 903. 904. 905.

Lexical item strippee subjectee submittee subordinatee subordinee subscribee subsidizee substitutee subsumee subvertee successee suckee sufferee suffocatee suggestee supportee suppressee surfee surfeitee surmountee surpassee survivee suspectee swaggeree swallowee swankee swearee swipee taggee tamee tauntee teachee televisee temptee tenuree terrorizee testifyee textee thankee therapee therapeutee thrashee

NF

HL

Rare

Est.

Frequ.

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

NQ



Appendix 2 237

No. 906. 907. 908. 909. 910. 911. 912. 913. 914. 915. 916. 917. 918. 919. 920. 921. 922. 923. 924. 925. 926. 927. 928. 929. 930. 931. 932. 933. 934. 935. 936. 937. 938. 939. 940. 941. 942. 943. 944. 945. 946. 947.

Lexical item threatenee thrillee ticklee timee toleratee tormentee tossee touchee toutee tracee trackee tradee trailee traitee trampee tranquilizee transcribee transformee transgressee translatee transmittee treacheree tributee trickee trick-or-treatee triumphee turnee tutoree tweakee twistee tyrannee tyrannizee unauthorizee unbelievee uncaree uncomfortee unconfimree uncoveree underestimatee underminee underscoree understatee

NF

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X

238 Heterogeneity in Word-Formation Patterns

No. 948. 949. 950. 951. 952. 953. 954. 955. 956. 957. 958. 959. 960. 961. 962. 963. 964. 965. 966. 967. 968. 969. 970. 971. 972. 973. 974. 975. 976. 977. 978. 979. 980. 981. 982. 983. 984. 985. 986. 987. 988. 989.

Lexical item

NF

undeservee undressee uneducatee unforgivee unitee unobligee unsubscribee unsuspectee untamee upliftee usheree valedatee valedictee vampee vanishee vanquishee vengee vexee vindictee vitalizee vituperatee vociferatee vomitee votee wageree wailee wakee wankee warnee washee watchee wateree wavee weepee weighee welcomee well wishee wheedlee whippee whistlee wincee witnessee

X

HL

Rare

Est.

Frequ.

NQ

X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X



Appendix 2 239

No.

Lexical item

990. 991. 992. 993. 994. 995. 996. 997. 998. 999. 1000.

wonderee workee woundee wrappee wreckee wrestlee wrongee yammeree yearnee yieldee yodellee

NF

HL

Rare

Est.

Frequ.

X X X X X X X X X X X

NQ

Name index

A Aarts  124 Adams  165 Aitchison  3, 45, 61, 75, 81, 88, 89, 168 Algeo  99, 185 Anderson  21, 22 Anshen  13, 14, 19, 91–93, 94, 100–102, 104, 105, 118 Aronoff  13, 14, 19, 27, 60, 91–94, 100–102, 104, 105, 117, 118 Asirvatham  152 Aston  122 B Baayen  7, 102, 104, 105, 117, 134 Baeskow  5, 7–10, 11, 19, 20, 22, 30, 44, 59 Baker  16, 176, 177 Barker  5, 7, 8, 10–12, 19, 20, 28, 29, 33, 34, 40–48, 50–53, 65, 81, 100, 108, 110, 112, 121, 201 Barnhart  114, 117, 201 Baroni  126, 129 Bauer  2–5, 7–13, 19–21, 24, 27, 28, 33–37, 39, 40, 41, 42, 65, 70, 74, 81, 82, 85, 86, 91, 94, 96, 97, 99, 100, 101, 108–110, 112–116, 121, 123, 143, 145, 159, 162, 201 Baugh  63 Bailey  165, 166 Bengtsson  36, 41, 65, 81 Biber  125, 130 Boake  157 Bolinger  113, 201 Booij  5, 7, 8, 10, 11, 12, 19, 20, 32, 44, 54, 55 Botha  97

Brandl  152 Burnhard  122 Bußmann  43 C Cable  53 Carlson  56 Chomsky  91, 92 Chopin  157, 158 Clark  51 Collins  122 Coseriu  97, 136 Creswell  26 Crystal  94, 165 D, E Dabke  176, 177, 178 Derwing  97 Di Sciullo  93, 118 Elsen  14, 97 Evert  126, 129 F Finkenstaedt  88 Firth  136 Fischer  9, 14, 82, 89, 97, 115, 116 Fisher  81 Foley  53 Franklin  170 G Gärtner  52 Ghani  129 Görlach  63, 64 Gold  26, 27, 189, 201 Gramsci  89 Green  114, 117 Greenbaum  122 Grefenstette  11, 14, 123, 125, 126, 127, 130

Gross  56 Gunn  178 H, J Halle  92 Hohenhaus  14, 92, 97, 99, 115 Horn  41 Horvath  165, 176 Hundt  11, 123 Jackendoff  54, 92, 118 Johansson  123 Jones  129 K Kastovsky  13, 63, 67, 68, 71, 97, 106, 115, 201 Katamba  2, 36, 45 Kehoe  129 Kilgariff  11, 14, 123, 125, 126, 127, 130 Kiparsky  97 Koziol  48, 62, 63 Kumar Ravi  152 L Labov  65 LaPolla  53 Lass  9, 61 Leisi  115 Leitner  176 Lees  91, 92 Lehmann  61 Lehnert  65, 112, 121, 201 Lehrer  14, 69, 97, 98, 116 Levin  43, 51 Lieber  5–8, 10–12, 14, 19, 20, 22, 32, 44, 54, 55, 59, 96, 117 Lipka  115 Lüdeling  126, 129 Lyons  14, 36, 56

242 Heterogeneity in Word-Formation Patterns

M Mair  65 Manning  125 Mathews  16, 169, 170, 171, 201 Matthews  94, 101 Marchand  16, 20, 59, 62, 63, 66, 70, 74, 78, 169 McEnery  124, 125 Meyer  124 Mladenic  129 Motsch  92 Murray  86 Mühleisen  127, 163 Muthmann  65, 112, 201

Peters  122 Plag  2, 13, 22, 52, 54, 94, 97, 98, 100, 101, 103, 105, 117, 118, 189 Portero Muñoz  7, 8, 9, 12, 19, 20, 27, 37, 44, 47, 53, 54, 100, 101, 117, 118 Pounder  100, 102, 103, 105, 106 Price  165, 166, 167 Pullum  96, 99

N, O Nash  178 Nevalainen  62, 63, 65, 67, 71, 76, 105 Oesterreicher  97, 151 Ong  151

Q, R Quirk  20, 36, 47, 58 Raffelsiefen  22 Ramson  16, 176 Rappaport Hovav  43, 51 Rehm  151, 152 Renouf  129 Resnik  11, 14, 123, 127 Rissanen  88 Romaine  9, 79 Roy  146

P Panten  171, 172 Pennycook  185 Perlmutter  43

S Safire  96 Schmid  100 Schröder  12, 97, 118, 163

Schütze  125 Schultink  14 Sebba  170 Simpson  178, 179 Skousen  97 Smith  11, 14, 123, 127 Strang  9 T, V, W Thiel  116 Van Valin  53 Walker  112, 201 Webster  79, 165 Wermser  62 Williams  93, 118 Wilson  124, 125 Wolff  88 Wood  201 Wycherly  157 X, Z Xu  127 Zappa  157 Zwicky  96, 99

Subject index

A Absolutive  41, 43 Actual words  8, 14, 19, 21, 25, 60, 69, 90, 92, 93, 94, 104, 105, 106, 117, 118, 123, 127, 142, 143, 163, 164, 189, 190, 191 Ad-hoc creations  10, 14, 17, 90, 91, 115, 116, 122, 126, 139, 142, 149, 151, 152, 158, 164, 177, 184, 189 Agentivity  5, 6, 27, 31, 32, 35, 40, 42, 43, 47–49, 51–53, 58, 59, 62, 67, 68, 70, 73–75, 77, 86, 87, 102, 109, 112, 114, 145, 157, 159, 161, 162, 167, 190 Ambiguity  5, 6, 11, 12, 15, 16, 17, 19, 30, 31, 37, 39, 40, 48, 49, 55, 58, 59, 73, 86, 88, 90, 103, 108, 109, 112, 126, 132, 142, 159, 161–163, 170, 178, 179, 190, 191 American English  2, 11, 39, 77, 79, 81, 83–85, 89, 110, 111, 113, 123, 164, 166, 168–171, 173, 175, 177, 182, 186, 191 Americanism  16, 79, 170–174, 186 American Speech (journal)  65, 96, 112, 169, 201 Analogy  13, 69, 74, 92, 98, 162, 173, 175 Analogical coining  6, 8, 12, 39, 59, 72, 97, 103, 164 Arawak  80 Australian aboriginal language  80, 170, 177 Australian English  11, 16, 49, 58, 83, 85, 122, 157, 166–170, 175–183, 192 Australianism  182

B Backformation  31 Bantu language  80 Borrowings  13, 17, 22, 32, 62, 63, 64, 66, 67, 68, 69, 71, 72, 74, 75, 78, 79, 80, 81, 86, 87, 89, 106, 113, 150, 158, 176, 186, 190 British English  16, 18, 58, 59, 79, 85, 114, 122, 123, 168, 169, 172–175, 180, 181–183, 186 Britishism  183 C Collocation  15, 17, 31, 122, 131, 136–143, 152, 156, 161, 163, 191 Compound  22, 29, 32, 98, 113, 115, 116, 133, 161, 167 Co-indexation of affixes  32, 54, 55 Co-occurrence of wordformation patterns  2, 8, 15, 31, 33, 136, 141 Corpus analysis; corpora  10, 11, 25, 31, 32, 65, 90, 104, 108, 118, 122, 123, 124, 125, 126, 130, 131, 175 Correlative nouns  30, 31, 32, 33, 48, 58, 68, 69, 74, 75, 89, 106, 190 D Diachronic development  8, 9, 12, 13, 14, 15, 16, 17, 19, 33, 39, 59, 60, 61–90, 95, 100, 105, 106, 108, 132, 142, 165, 169, 190, 192 Dictionaries as data sources  8, 10, 20, 21, 65, 82, 85, 86, 91, 104, 105, 106, 107, 114, 115, 116, 117, 121, 122, 123, 136, 140, 141, 164, 167, 176, 191

E Early Modern Period  62, 63, 67, 71 Eighteenth century  76–81, 86, 87, 105, 106, 165, 170 Episodic linking  7, 12, 50–52, 100, 145, 147, 148 Ergativity  12, 41, 43, 44 -er suffix  1, 2, 5, 6–8, 15, 19, 20, 31–33, 35, 40, 44, 46, 47, 48, 51, 54, 55, 57, 58, 59, 62, 63, 68, 75, 86, 99, 100, 102, 114, 128, 136–143, 145, 152, 156, 162, 163, 190, 191 F Fifteenth century  13, 30, 62–68, 88, 107, 167 Fossilization  36 Fourteenth century  13, 64, 87, 89, 107 French influence  59, 62–72, 74, 75, 78–81, 86, 87, 89, 106, 113, 127, 131, 134, 158, 176, 186, 190 G, H Globalization (of English)  16, 18, 165, 185, 186 Hapax legomena  15, 17, 81, 104, 131, 134, 135, 141, 149, 150–152, 155–164, 184, 191 Haplology restriction  22 Hindi  79 Humorous formations  34, 53, 74, 78, 81, 85, 96, 99, 113, 114, 158, 163, 191 Hypocoristics  16, 49, 50, 74, 87, 132, 149, 161, 169, 170, 175, 177–179, 187, 192 Hypostatization  115

244 Heterogeneity in Word-Formation Patterns

I, J Indian English  87, 122, 170, 179, 180, 182, 183, 192 Irish English  86, 175 Jamaican English  168, 186 K, L Language change  9, 17, 60, 61, 64, 66, 69, 75, 79, 81, 88, 90, 93, 98, 106, 118, 169, 184, 185 Language contact  8, 11, 13, 17, 59, 64, 93, 169, 186, 190, 192 Latin  63, 64, 68, 87, 101, 165 Law French  13, 66, 70, 106 Legal domain  13, 34, 37, 42, 49, 53, 57, 61, 63, 64, 65, 66, 67, 68, 72, 74, 78, 81, 85, 89, 114, 121, 142, 147, 151, 152, 155–158, 163, 176, 179, 180, 182, 183, 191 Lexical Conceptual Structure  54 Lexicalization  36 M Middle English Period  62, 63, 64 Minimalist word-formation theory  7 Morphological creativity  14, 60, 81, 90–93, 95–99, 101, 115, 190 Morphology and the lexicon  14, 60, 91, 92, 93, 118, 189 N Native speaker, concept of  11, 16, 81, 95, 131, 155, 174, 185, 186 Nativization  21, 67, 69 New English Dictionary  86, 166, 167 Neologisms  1, 8, 9, 11, 12, 14, 15, 17, 20, 21, 59, 65, 66, 69, 81, 82, 89–91, 95–99, 104, 105, 108, 114–118, 121, 123, 126, 127, 130, 135, 136, 155, 162, 164, 167, 184, 186, 187 New Zealand English  11, 164, 175, 179, 180, 184, 192

Nineteenth century  13, 34, 42, 60, 61, 64, 69, 72, 78, 80–88, 96, 106–109, 111, 113, 123, 166, 176, 178, 186, 190 Nonce words  10, 14, 15, 33, 61, 77, 81–84, 96, 97, 110, 114–116, 121, 139, 141, 149, 154, 158, 159, 161, 162, 164, 177, 184, 189, 191 Non-native speaker influences  11, 16, 131, 155, 174, 185, 186 No Phrase Constraint  114 O Old English  62 Old French  66, 67, 87 Orthographic influences  12, 21, 25, 79, 129, 149 Oxford English Dictionary  2, 3, 5, 6, 10, 14, 20, 25, 27, 28, 30, 32–38, 50, 59, 65, 66, 67, 68–75, 77, 78, 80, 81, 83–86, 88, 91, 101–103, 106–108, 110, 112, 113, 115–117, 121, 127, 128, 132, 138, 139, 141, 142, 146, 157, 165–167, 171, 173, 174, 201 P Paradigm pressure  2 Passivity in -ee words  2, 5, 7, 12, 20, 25, 36, 41, 42, 48, 58, 63, 68, 69, 70, 71, 72, 74, 78, 85, 89, 101, 102, 108, 113, 190 Passivizability hypothesis  41, 42 Personhood in -ee words  2, 3, 4, 5, 7, 20, 30, 45, 46, 48, 69, 73, 75, 77, 79, 80, 85, 109, 111, 140 Phonological constraints  7, 11, 12, 16, 18, 20–27, 30, 36, 91, 95, 99, 102, 117, 129, 149 Polysemy  1, 5, 6, 8, 14, 17, 39, 54, 55, 58, 60, 88, 97, 98 Possible words  8, 10, 14, 15, 19, 91–94, 105, 117–119, 123, 129, 162, 163, 189, 191 Productivity  2, 7, 9, 10, 11, 13, 14, 16, 20, 21, 31, 40, 60, 90, 91–119, 121, 162, 163, 189, 191

and frequency  95, 100, 103, 104, 105 diachronic  14, 16, 61, 62, 63, 65, 76, 81, 89, 100, 105, 106, 107, 108 gradual  101 measurement of  8, 17, 35, 62, 65, 104, 105, 134 qualitative  101, 102, 103 quantitative  103, 104, 105, 134 synchronic  13, 14, 35, 95, 100, 108–114 Prototypes  8, 12, 17, 33, 48, 50, 53, 55–59, 74, 142, 145, 147, 162, 163, 164, 190, 191 R Relational grammar  43 Religious domain  30, 72, 146, 156 Reverse dictionaries  8, 10, 201 S Scots English  84, 85, 86, 166 Semantic contraints  7, 11, 13, 14, 18, 20, 21, 44–54, 108 Sentience  7, 12, 15, 45, 46, 47, 48, 52, 55, 57, 58, 100, 102, 111, 142, 145, 190 Seventeenth century  13, 71–78, 105–107, 157, 158 Sixteenth century  54, 58, 60, 69–72, 74, 86, 97, 107 Spanish  80 Spelling  66, 79, 87, 129, 131, 132, 140, 158, 165, 166, 178 Syntactic constraints  27–43, 44 T Text types  11, 15, 17, 89, 123, 126, 131, 149–156, 177, 184, 187, 190 Thematic role  7, 20, 44, 48, 52, 53 Truncation  22–24 Twentieth century  13, 14, 34, 50, 60, 61, 65, 71, 85, 88, 90, 91, 96, 99, 102, 106–115, 121, 157, 167, 174, 189, 191



U, V Unaccusative Hypothesis  41, 43 Unitary Base Hypothesis  27, 103 Urdu  79 Verbal base of -ee words  2, 4, 7, 13, 20, 27, 30, 32, 33, 52, 54, 55, 57–59, 89, 127, 135, 136, 142, 149, 163, 190, 191

Subject index 245

Volition, lack of  16, 40, 42, 45, 48, 49, 50, 53, 55, 57, 58, 74, 100, 102, 132, 142, 145, 146, 147, 157, 161, 170, 179, 190, 192 W Web-corpus  50, 65, 105, 119, 123, 124, 125, 126, 127, 129, 130, 164

Word-formation rules  1, 5, 12, 14, 19, 26, 36, 58, 59, 60, 91–98, 101, 115, 117, 118, 189, 190 World Wide Web  8, 10, 11, 14, 30, 37, 38, 82, 104, 119, 123, 125–130, 149, 151, 155, 162, 168, 187, 189

Studies in Language Companion Series A complete list of titles in this series can be found on the publishers’ website, www.benjamins.com 118 Mühleisen, Susanne: Heterogeneity in Word–Formation Patterns. A corpus-based analysis of suffixation with -ee and its productivity in English. 2010. xiii, 245 pp. 117 Spevak, Olga: Constituent Order in Classical Latin Prose. xv, 315 pp. + index. Expected February 2010 116 Nordström, Jackie: Modality and Subordinators. 2010. xvii, 341 pp. 115 Hasko, Victoria and Renee Perelmutter (eds.): New Approaches to Slavic Verbs of Motion. x, 381 pp. + index. Expected February 2010 114 Roby, David Brian: Aspect and the Categorization of States. The case of ser and estar in Spanish. 2009. xiii, 191 pp. 113 Comrie, Bernard, Ray Fabri, Elizabeth Hume, Manwel Mifsud, Thomas Stolz and Martine Vanhove (eds.): Introducing Maltese Linguistics. Selected papers from the 1st International Conference on Maltese Linguistics, Bremen, 18–20 October, 2007. 2009. xi, 422 pp. 112 Dufter, Andreas and Daniel Jacob (eds.): Focus and Background in Romance Languages. 2009. vii, 362 pp. 111 Polguère, Alain and Igor A. Mel’čuk (eds.): Dependency in Linguistic Description. 2009. xxii, 281 pp. 110 Dimmendaal, Gerrit J. (ed.): Coding Participant Marking. Construction types in twelve African languages. 2009. xvi, 389 pp. 109 Narrog, Heiko: Modality in Japanese. The layered structure of the clause and hierarchies of functional categories. 2009. xxii, 277 pp. 108 Barðdal, Jóhanna and Shobhana L. Chelliah (eds.): The Role of Semantic, Pragmatic, and Discourse Factors in the Development of Case. 2009. xx, 432 pp. 107 Butler, Christopher S. and Javier Martín Arista (eds.): Deconstructing Constructions. 2009. xx, 306 pp. 106 Vanhove, Martine (ed.): From Polysemy to Semantic Change. Towards a typology of lexical semantic associations. 2008. xiii, 404 pp. 105 Van Valin, Jr., Robert D. (ed.): Investigations of the Syntax–Semantics–Pragmatics Interface. 2008. xxiv, 484 pp. 104 Mushin, Ilana and Brett Baker (eds.): Discourse and Grammar in Australian Languages. 2008. x, 239 pp. 103 Josephson, Folke and Ingmar Söhrman (eds.): Interdependence of Diachronic and Synchronic Analyses. 2008. viii, 350 pp. 102 Goddard, Cliff (ed.): Cross-Linguistic Semantics. 2008. xvi, 356 pp. 101 Stolz, Thomas, Sonja Kettler, Cornelia Stroh and Aina Urdze: Split Possession. An areallinguistic study of the alienability correlation and related phenomena in the languages of Europe. 2008. x, 546 pp. 100 Ameka, Felix K. and M.E. Kropp Dakubu (eds.): Aspect and Modality in Kwa Languages. 2008. ix, 335 pp. 99 Høeg Müller, Henrik and Alex Klinge (eds.): Essays on Nominal Determination. From morphology to discourse management. 2008. xviii, 369 pp. 98 Fabricius-Hansen, Cathrine and Wiebke Ramm (eds.): 'Subordination' versus 'Coordination' in Sentence and Text. A cross-linguistic perspective. 2008. vi, 359 pp. 97 Dollinger, Stefan: New-Dialect Formation in Canada. Evidence from the English modal auxiliaries. 2008. xxii, 355 pp. 96 Romeo, Nicoletta: Aspect in Burmese. Meaning and function. 2008. xv, 289 pp. 95 O’Connor, Loretta: Motion, Transfer and Transformation. The grammar of change in Lowland Chontal. 2007. xiv, 251 pp. 94 Miestamo, Matti, Kaius Sinnemäki and Fred Karlsson (eds.): Language Complexity. Typology, contact, change. 2008. xiv, 356 pp. 93 Schalley, Andrea C. and Drew Khlentzos (eds.): Mental States. Volume 2: Language and cognitive structure. 2007. x, 362 pp. 92 Schalley, Andrea C. and Drew Khlentzos (eds.): Mental States. Volume 1: Evolution, function, nature. 2007. xii, 304 pp.

91 Filipović, Luna: Talking about Motion. A crosslinguistic investigation of lexicalization patterns. 2007. x, 182 pp. 90 Muysken, Pieter (ed.): From Linguistic Areas to Areal Linguistics. 2008. vii, 293 pp. 89 Stark, Elisabeth, Elisabeth Leiss and Werner Abraham (eds.): Nominal Determination. Typology, context constraints, and historical emergence. 2007. viii, 370 pp. 88 Ramat, Paolo and Elisa Roma (eds.): Europe and the Mediterranean as Linguistic Areas. Convergencies from a historical and typological perspective. 2007. xxvi, 364 pp. 87 Verhoeven, Elisabeth: Experiential Constructions in Yucatec Maya. A typologically based analysis of a functional domain in a Mayan language. 2007. xiv, 380 pp. 86 Schwarz-Friesel, Monika, Manfred Consten and Mareile Knees (eds.): Anaphors in Text. Cognitive, formal and applied approaches to anaphoric reference. 2007. xvi, 282 pp. 85 Butler, Christopher S., Raquel Hidalgo Downing and Julia Lavid (eds.): Functional Perspectives on Grammar and Discourse. In honour of Angela Downing. 2007. xxx, 481 pp. 84 Wanner, Leo (ed.): Selected Lexical and Grammatical Issues in the Meaning–Text Theory. In honour of Igor Mel'čuk. 2007. xviii, 380 pp. 83 Hannay, Mike and Gerard J. Steen (eds.): Structural-Functional Studies in English Grammar. In honour of Lachlan Mackenzie. 2007. vi, 393 pp. 82 Ziegeler, Debra: Interfaces with English Aspect. Diachronic and empirical studies. 2006. xvi, 325 pp. 81 Peeters, Bert (ed.): Semantic Primes and Universal Grammar. Empirical evidence from the Romance languages. 2006. xvi, 374 pp. 80 Birner, Betty J. and Gregory Ward (eds.): Drawing the Boundaries of Meaning. Neo-Gricean studies in pragmatics and semantics in honor of Laurence R. Horn. 2006. xii, 350 pp. 79 Laffut, An: Three-Participant Constructions in English. A functional-cognitive approach to caused relations. 2006. ix, 268 pp. 78 Yamamoto, Mutsumi: Agency and Impersonality. Their Linguistic and Cultural Manifestations. 2006. x, 152 pp. 77 Kulikov, Leonid, Andrej Malchukov and Peter de Swart (eds.): Case, Valency and Transitivity. 2006. xx, 503 pp. 76 Nevalainen, Terttu, Juhani Klemola and Mikko Laitinen (eds.): Types of Variation. Diachronic, dialectal and typological interfaces. 2006. viii, 378 pp. 75 Hole, Daniel, André Meinunger and Werner Abraham (eds.): Datives and Other Cases. Between argument structure and event structure. 2006. viii, 385 pp. 74 Pietrandrea, Paola: Epistemic Modality. Functional properties and the Italian system. 2005. xii, 232 pp. 73 Xiao, Richard Zhonghua and Tony McEnery: Aspect in Mandarin Chinese. A corpus-based study. 2004. x, 305 pp. 72 Frajzyngier, Zygmunt, Adam Hodges and David S. Rood (eds.): Linguistic Diversity and Language Theories. 2005. xii, 432 pp. 71 Dahl, Östen: The Growth and Maintenance of Linguistic Complexity. 2004. x, 336 pp. 70 Lefebvre, Claire: Issues in the Study of Pidgin and Creole Languages. 2004. xvi, 358 pp. 69 Tanaka, Lidia: Gender, Language and Culture. A study of Japanese television interview discourse. 2004. xvii, 233 pp. 68 Moder, Carol Lynn and Aida Martinovic-Zic (eds.): Discourse Across Languages and Cultures. 2004. vi, 366 pp. 67 Luraghi, Silvia: On the Meaning of Prepositions and Cases. The expression of semantic roles in Ancient Greek. 2003. xii, 366 pp. 66 Nariyama, Shigeko: Ellipsis and Reference Tracking in Japanese. 2003. xvi, 400 pp. 65 Matsumoto, Kazuko: Intonation Units in Japanese Conversation. Syntactic, informational and functional structures. 2003. xviii, 215 pp. 64 Butler, Christopher S.: Structure and Function – A Guide to Three Major Structural-Functional Theories. Part 2: From clause to discourse and beyond. 2003. xiv, 579 pp. 63 Butler, Christopher S.: Structure and Function – A Guide to Three Major Structural-Functional Theories. Part 1: Approaches to the simplex clause. 2003. xx, 573 pp. 62 Field, Fredric: Linguistic Borrowing in Bilingual Contexts. With a foreword by Bernard Comrie. 2002. xviii, 255 pp.

61 Goddard, Cliff and Anna Wierzbicka (eds.): Meaning and Universal Grammar. Theory and empirical findings. Volume 2. 2002. xvi, 337 pp. 60 Goddard, Cliff and Anna Wierzbicka (eds.): Meaning and Universal Grammar. Theory and empirical findings. Volume 1. 2002. xvi, 337 pp. 59 Shi, Yuzhi: The Establishment of Modern Chinese Grammar. The formation of the resultative construction and its effects. 2002. xiv, 262 pp. 58 Maylor, B. Roger: Lexical Template Morphology. Change of state and the verbal prefixes in German. 2002. x, 273 pp. 57 Mel’čuk, Igor A.: Communicative Organization in Natural Language. The semantic-communicative structure of sentences. 2001. xii, 393 pp. 56 Faarlund, Jan Terje (ed.): Grammatical Relations in Change. 2001. viii, 326 pp. 55 Dahl, Östen and Maria Koptjevskaja-Tamm (eds.): Circum-Baltic Languages. Volume 2: Grammar and Typology. 2001. xx, 423 pp. 54 Dahl, Östen and Maria Koptjevskaja-Tamm (eds.): Circum-Baltic Languages. Volume 1: Past and Present. 2001. xx, 382 pp. 53 Fischer, Olga, Anette Rosenbach and Dieter Stein (eds.): Pathways of Change. Grammaticalization in English. 2000. x, 391 pp. 52 Torres Cacoullos, Rena: Grammaticization, Synchronic Variation, and Language Contact. A study of Spanish progressive -ndo constructions. 2000. xvi, 255 pp. 51 Ziegeler, Debra: Hypothetical Modality. Grammaticalisation in an L2 dialect. 2000. xx, 290 pp. 50 Abraham, Werner and Leonid Kulikov (eds.): Tense-Aspect, Transitivity and Causativity. Essays in honour of Vladimir Nedjalkov. 1999. xxxiv, 359 pp. 49 Bhat, D.N.S.: The Prominence of Tense, Aspect and Mood. 1999. xii, 198 pp. 48 Manney, Linda Joyce: Middle Voice in Modern Greek. Meaning and function of an inflectional category. 2000. xiii, 262 pp. 47 Brinton, Laurel J. and Minoji Akimoto (eds.): Collocational and Idiomatic Aspects of Composite Predicates in the History of English. 1999. xiv, 283 pp. 46 Yamamoto, Mutsumi: Animacy and Reference. A cognitive approach to corpus linguistics. 1999. xviii, 278 pp. 45 Collins, Peter and David Lee (eds.): The Clause in English. In honour of Rodney Huddleston. 1999. xv, 342 pp. 44 Hannay, Mike and A. Machtelt Bolkestein (eds.): Functional Grammar and Verbal Interaction. 1998. xii, 304 pp. 43 Olbertz, Hella, Kees Hengeveld and Jesús Sánchez García (eds.): The Structure of the Lexicon in Functional Grammar. 1998. xii, 312 pp. 42 Darnell, Michael, Edith A. Moravcsik, Michael Noonan, Frederick J. Newmeyer and Kathleen M. Wheatley (eds.): Functionalism and Formalism in Linguistics. Volume II: Case studies. 1999. vi, 407 pp. 41 Darnell, Michael, Edith A. Moravcsik, Michael Noonan, Frederick J. Newmeyer and Kathleen M. Wheatley (eds.): Functionalism and Formalism in Linguistics. Volume I: General papers. 1999. vi, 486 pp. 40 Birner, Betty J. and Gregory Ward: Information Status and Noncanonical Word Order in English. 1998. xiv, 314 pp. 39 Wanner, Leo (ed.): Recent Trends in Meaning–Text Theory. 1997. xx, 202 pp. 38 Hacking, Jane F.: Coding the Hypothetical. A comparative typology of Russian and Macedonian conditionals. 1998. vi, 156 pp. 37 Harvey, Mark and Nicholas Reid (eds.): Nominal Classification in Aboriginal Australia. 1997. x, 296 pp. 36 Kamio, Akio (ed.): Directions in Functional Linguistics. 1997. xiii, 259 pp. 35 Matsumoto, Yoshiko: Noun-Modifying Constructions in Japanese. A frame semantic approach. 1997. viii, 204 pp. 34 Hatav, Galia: The Semantics of Aspect and Modality. Evidence from English and Biblical Hebrew. 1997. x, 224 pp. 33 Velázquez-Castillo, Maura: The Grammar of Possession. Inalienability, incorporation and possessor ascension in Guaraní. 1996. xvi, 274 pp. 32 Frajzyngier, Zygmunt: Grammaticalization of the Complex Sentence. A case study in Chadic. 1996. xviii, 501 pp.

31 Wanner, Leo (ed.): Lexical Functions in Lexicography and Natural Language Processing. 1996. xx, 355 pp. 30 Huffman, Alan: The Categories of Grammar. French lui and le. 1997. xiv, 379 pp. 29 Engberg-Pedersen, Elisabeth, Michael Fortescue, Peter Harder, Lars Heltoft and Lisbeth Falster Jakobsen (eds.): Content, Expression and Structure. Studies in Danish functional grammar. 1996. xvi, 510 pp. 28 Herman, József (ed.): Linguistic Studies on Latin. Selected papers from the 6th International Colloquium on Latin Linguistics (Budapest, 23–27 March 1991). 1994. ix, 421 pp. 27 Abraham, Werner, T. Givón and Sandra A. Thompson (eds.): Discourse, Grammar and Typology. Papers in honor of John W.M. Verhaar. 1995. xx, 352 pp. 26 Lima, Susan D., Roberta Corrigan and Gregory K. Iverson: The Reality of Linguistic Rules. 1994. xxiii, 480 pp. 25 Goddard, Cliff and Anna Wierzbicka (eds.): Semantic and Lexical Universals. Theory and empirical findings. 1994. viii, 510 pp. 24 Bhat, D.N.S.: The Adjectival Category. Criteria for differentiation and identification. 1994. xii, 295 pp. 23 Comrie, Bernard and Maria Polinsky (eds.): Causatives and Transitivity. 1993. x, 399 pp. 22 McGregor, William B.: A Functional Grammar of Gooniyandi. 1990. xx, 618 pp. 21 Coleman, Robert (ed.): New Studies in Latin Linguistics. Proceedings of the 4th International Colloquium on Latin Linguistics, Cambridge, April 1987. 1990. x, 480 pp. 20 Verhaar, John W.M. S.J. (ed.): Melanesian Pidgin and Tok Pisin. Proceedings of the First International Conference on Pidgins and Creoles in Melanesia. 1990. xiv, 409 pp. 19 Blust, Robert A.: Austronesian Root Theory. An essay on the limits of morphology. 1988. xi, 190 pp. 18 Wierzbicka, Anna: The Semantics of Grammar. 1988. vii, 581 pp. 17 Calboli, Gualtiero (ed.): Subordination and Other Topics in Latin. Proceedings of the Third Colloquium on Latin Linguistics, Bologna, 1–5 April 1985. 1989. xxix, 691 pp. 16 Conte, Maria-Elisabeth, János Sánder Petöfi and Emel Sözer (eds.): Text and Discourse Connectedness. Proceedings of the Conference on Connexity and Coherence, Urbino, July 16–21, 1984. 1989. xxiv, 584 pp. 15 Justice, David: The Semantics of Form in Arabic. In the mirror of European languages. 1987. iv, 417 pp. 14 Benson, Morton, Evelyn Benson and Robert F. Ilson: Lexicographic Description of English. 1986. xiii, 275 pp. 13 Reesink, Ger: Structures and their Functions in Usan. 1987. xviii, 369 pp. 12 Pinkster, Harm (ed.): Latin Linguistics and Linguistic Theory. Proceedings of the 1st International Colloquium on Latin Linguistics, Amsterdam, April 1981. 1983. xviii, 307 pp. 11 Panhuis, Dirk G.J.: The Communicative Perspective in the Sentence. A study of Latin word order. 1982. viii, 172 pp. 10 Dressler, Wolfgang U., Willi Mayerthaler, Oswald Panagl and Wolfgang Ullrich Wurzel: Leitmotifs in Natural Morphology. 1988. ix, 168 pp. 9 Lang, Ewald and John Pheby: The Semantics of Coordination. (English transl. by John Pheby from the German orig. ed. 'Semantik der koordinativen Verknüpfung', Berlin, 1977). 1984. 300 pp. 8 Barth, E.M. and J.L. Martens (eds.): Argumentation: Approaches to Theory Formation. Containing the Contributions to the Groningen Conference on the Theory of Argumentation, October 1978. 1982. xviii, 333 pp. 7 Parret, Herman, Marina Sbisà and Jef Verschueren (eds.): Possibilities and Limitations of Pragmatics. Proceedings of the Conference on Pragmatics, Urbino, July 8–14, 1979. 1981. x, 854 pp. 6 Vago, Robert M. (ed.): Issues in Vowel Harmony. Proceedings of the CUNY Linguistics Conference on Vowel Harmony, May 14, 1977. 1980. xx, 340 pp. 5 Haiman, John: Hua: A Papuan Language of the Eastern Highlands of New Guinea. 1980. iv, 550 pp. 4 Lloyd, Albert L.: Anatomy of the Verb. The Gothic Verb as a Model for a Unified Theory of Aspect, Actional Types, and Verbal Velocity. (Part I: Theory; Part II: Application). 1979. x, 351 pp. 3 Malkiel, Yakov: From Particular to General Linguistics. Selected Essays 1965–1978. With an introduction by the author, an index rerum and an index nominum. 1983. xxii, 659 pp. 2 Anwar, Mohamed Sami: BE and Equational Sentences in Egyptian Colloquial Arabic. 1979. vi, 128 pp. 1 Abraham, Werner (ed.): Valence, Semantic Case, and Grammatical Relations. Workshop studies prepared for the 12th International Congress of Linguists, Vienna, August 29th to September 3rd, 1977. 1978. xiv, 729 pp.