AFRICAN LANGUAGES AND CULTURES SUPPLEMENT 1
PAPERS IN HONOUR OF R.C.BRAHAM (1890–1963)
PAPERS IN HONOUR OF R.C.ABRAHAM (1890–1963) Edited by
Philip J.Jaggar School of Oriental and African Studies Papers from the Symposium on R.C.Abraham (1890–1963) at the School of Oriental and African Studies, University of London September 14–15, 1990
African Languages and Cultures, Supplement 1 School of Oriental and African Studies, University of London 1992
Published by the School of Oriental and African Studies, University of London, Thornaugh Street, Russell Square, London WC1H 0XG This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © School of Oriental and African Studies 1992 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 0-203-98997-X Master e-book ISBN
ISBN 0 7286 0210 5 (Print Edition)
Contents
Preface
vii
Roy Clive Abraham: A biographical profile and list of writings Philip J.Jaggar
1
Afroasiatic Amharic R.C.Abraham’s work on Amharic David L.Appleyard On R.C.Abraham’s view of Amharic as a ‘tone language’ R.J.Hayward
4 12
Bole R.C.Abraham: the Bolewa and Bolanci John E.Lavers
24
Hausa Hausa orthography and Abraham’s transcription J.Carnochan
30
A note on Hausa literature and R.C. Abraham Graham Furniss
34
R.C.Abraham’s early insights into Hausa pre-datival verb forms Philip J.Jaggar
40
The previous reference marker in Hausa: R.C.Abraham’s insights and new analyses Paul Newman
55
Lexicographic method in R.C.Abraham’s Hausa Dictionary Roxana Ma Newman
64
Somali Reflections on R.C.Abraham’s Somali English Dictionary B.W.Andrzejewski
African Languages and Cultures, Supplement 1, 1992
80
vi
R.C.Abraham and Somali grammar: tone, derivational morphology and information structure John Ibrahim Saeed
89
Niger-Congo Idoma The contribution of R.C.Abraham to Idoma studies Shamsudeen O.O.Amali
101
Igbo R.C.Abraham and D.Alagoma: their contribution to Igbo studies Kay Williamson
105
Tiv R.C.Abraham’s books on Tiv D.W.Arnott
114
R.C.Abraham and The Tiv People Frances Harding
118
Linguistic studies on Tiv—before, by, and after R.C.Abraham Heinz Jockers
130
Yoruba R.C.Abraham: a link between two periods in the Yoruba written tradition J.Gbenga Fagbọrun
139
Abraham’s ‘magnetised tone’ in Yorùbá Benjamin Akíntúndé Oyètádé
149
General Fond recollections of Roy P.E.H.Hair Guidelines for contributors inside back cover
157
PREFACE
The year 1990 marked the centenary of the birth of Roy Clive Abraham, a household name to anyone working within the field of African linguistics. To honour the event, a Symposium on R.C.Abraham (1890– 1963) was held at the School of Oriental and African Studies, University of London, on September 14–15, 1990, and a group of distinguished scholars were invited to present papers assessing Abraham’s contribution to our knowledge and understanding of a range of African languages (and cultures) within the Afroasiatic and Niger-Congo families. He worked on about 12 different African languages over a period of more than 30 years, publishing grammars and dictionaries on seven (Amharic, Hausa, Idoma, Igbo, Somali, Tiv, Yoruba), and his wide-ranging and careful scholarship has served as a model and catalyst for many of the papers here (one can only guess at what he could have achieved had he been born into the age of wordprocessing and databases!). This celebratory volume is the product of that Symposium and contains 18 papers arranged alphabetically by language-family, according to the particular language and author (the papers by David Arnott, Paul Hair and Paul Newman are included although the authors were unable to attend the Symposium). Whilst Abraham’s prodigious output did not always attain the same consistently high overall quality— hardly surprising given the adverse circumstances in which he was sometimes forced to work—his language publications were without doubt the product of some original and perceptive linguistic research. A number of the papers in the collection—those by Fagbọrun [Yoruba], Hayward [Amharic], Jaggar [Hausa], P.Newman [Hausa], Oyètádé [Yoruba] and Saeed [Somali]—evaluate Abraham’s description and analysis of various language-specific phenomena, and reveal that in some areas he had imaginative and original insights which were to be the springboard for later treatments. The more general surveys by Amali [Idoma], Appleyard [Amharic], Arnott [Tiv], Jockers [Tiv] and Lavers [Bole] take a broader look at Abraham’s contribution to the study of various languages, against the backdrop of other relevant works. The remaining review-papers assess Abraham’s work in several areas and include: Andrzejewski and R.Newman on his Somali and Hausa lexicography respectively; Williamson on the techniques and methods he used for his Igbo research (together with Dagogo Alagoma); Carnochan on his orthographic representation of Hausa; Furniss on his translations of fragments of Hausa literature; and Harding on his ethnographic portrayal and perception of Tiv culture (colonial prejudices and all). The volume concludes with some of Hair’s personal reminiscences of Abraham. Some of the linguistically-oriented papers highlight similarities in Abraham’s approach to language description, and his methodology and presentation, e.g. his admirable thoroughness and meticulous attention to detail, features which, however, sometimes produced over-complicated and abstruse descriptions (certainly of interest to the serious linguist, but of limited pedagogical value). And in several papers, especially Hair’s, we get glimpses of an enthusiastic but at times abrasive and egocentric personality who was often at odds with the academic community. Above all, however, the collection of papers in this
viii
volume is vivid testimony to Abraham’s impact, especially in the field of African languages, and on behalf of all the contributors, this work is dedicated to the memory of an extraordinary scholar who advanced our knowledge and understanding so much. R.C.Abraham’s son Donald sent the following message to the Symposium: As someone with an intimate personal knowledge of the skill and single-minded perseverance that my father devoted to his African language studies over a period of many years, I extend, on his behalf, and for my family and myself, warmest greetings and a deep sense of appreciation to the gifted participants from far and wide, in the Centenary Symposium now opening at the Centre of African Studies, SOAS, under the guidance of Dr Philip Jaggar. I greatly regret my inability to be with and among you, and thank you one and all. Donald Abraham Columbia, South Carolina, USA September 13, 1990 As organiser of the Symposium, I would like to take this opportunity to thank the SOAS Research and Publications Committee, the A.G. Leventis Foundation, the British Academy, and the Ove Arup Partnership for their generous assistance in funding the Symposium. I would also like to thank Joyce Hutchinson who keyed in some of the papers, Anna Debska and Diana Matias for carefully proofreading the entire manuscript, and Michael Mann for technical assistance. Alphabetical list of contributors at the Symposium: Shamsudeen O.O.Amali, University of Jos Bogumil Andrzejewski, SOAS, University of London David Appleyard, SOAS, University of London Jack Carnochan, SOAS, University of London Gbenga Fagbọrun, University of York and Ife Graham Furniss, SOAS, University of London Frances Harding, SOAS, University of London Dick Hayward, SOAS, University of London Philip Jaggar, SOAS, University of London Heinz Jockers, Universität Hamburg John Lavers, Bayero University, Kano Roxana Ma Newman, University of Indiana Benjamin Akíntúndé Oyètádé, SOAS, University of London John Saeed, CSNL, Trinity College, Dublin Kay Williamson, University of Port Harcourt London, August 1992 Philip J.Jaggar
ROY CLIVE ABRAHAM: A BIOGRAPHICAL PROFILE AND LIST OF WRITINGS Philip J.Jaggar
Roy Clive Abraham was a major figure in the field of 20th-century African language scholarship. He was born on December 16, 1890 in Melbourne, Australia, and first educated at University College London Preparatory School, Clifton College, and Heidelberg College, Germany. He was a Brassey Italian Scholar at Balliol College, Oxford (1922–1924), where he obtained a first class degree in Oriental Languages (Arabic and Persian), and also offered Ethiopic which he had studied in Leipzig in 1920, but there was no examiner available at the time. In 1927 he received a Certificate in Anthropology from University College London, and in 1930 a Diploma in Arabic from the then School of Oriental Studies, University of London (his lowest mark was 73%!). During and after the First World War, Abraham served in the British Army (1914–1923) in Arabia, and also on the North West Frontier where he learned Hindustani. He was a member of the Nigerian Administrative Service between 1924–1944, during which time he was an Administrative Officer, Government Anthropologist, and was then seconded for language research for six years. As far as we know, his first independent research was on Bole (=Bolanci), a West Chadic (Afroasiatic) language spoken in northern Nigeria (see Lavers in this volume). He then assisted the Reverend G.P.Bargery, from whom he learned the principle of phonemic tone in West African languages, in the compilation of the latter’s monumental A Hausa-English Dictionary (1934). Abraham’s first published works were The Grammar of Tiv (1933) and The Principles of Hausa (1934), and in 1935 he published, in mimeograph form, The Principles of Idoma, the first detailed grammar of an eastern Kwa language. Between 1941–1942 he was Instructor in Hausa to the Royal West Africa Frontier Force, after which he served in Abyssinia, teaching Amharic and Somali. Later on in the Second World War, he served in Kenya, South Africa, France, Italy, and with the Military Mission to Russia where he learned Russian. He finished his service with the rank of Major. In 1945, Abraham received a Leverhulme Research Fellowship to study the modern Semitic languages of Ethiopia and Eritrea. Although in late 1946 he failed in his bid to be appointed as Bargery’s successor in Hausa at the School of Oriental and African Studies, he was subsequently appointed to the newly created Lectureship in Amharic, a position he held from 1948 until his retirement in 1951. During this same period he began research into Berber and Somali, and in 1949 he successfully supplicated for the DLitt of the University of Oxford on the basis of three of his published works on Tiv. Abraham’s Dictionary of the Hausa Language (with Mai Kano) appeared in 1949, and in 1951 The Principles of Somali (with Solomon Warsama) was published in mimeograph form. In 1952, he made his
African Languages and Cultures, Supplement 1 (1992):1–4
2
PHILIP J.JAGGAR
own way to Ibadan, Nigeria, where he began research on Yoruba, and his Dictionary of Modern Yoruba was published in 1958. His last years were dogged by ill-health, but his extraordinary determination and zeal meant that this did not deter him from working on Igbo, and a dictionary and grammar were close to completion when he died in June 1963 at the age of 73. A lengthy appreciation of Abraham is available in Armstrong (1964), an earlier bibliography in Hair (1965), and a detailed biographical sketch appears in Jaggar and Lavers (1992). REFERENCES Abraham, R.C. 1933. The Grammar of Tiv. Kaduna: Government Printer. ——. 1934. The Principles of Hausa. Kaduna: Government Printer. ——. 1935. The Principles of Idoma. London: Published by the author through Crown Agents for the Colonies. 1958. Dictionary of Modern Yoruba. London: University of London Press. Abraham, R.C., and Mai Kano. 1949. Dictionary of the Hausa Language. 1st ed. London: Crown Agents for the Colonies. Armstrong, Robert G. 1964. Roy Clive Abraham, 1890–1963. The Journal of West African Languages 1(1):49–53. Bargery, G.P. 1934. A Hausa-English Dictionary. London: Oxford University Press. Hair, P.E.H. 1965. A bibliography of R.C.Abraham—linguist and lexicographer. The Journal of West African Languages 2(1):63–66. Jaggar, Philip J., and John E.Lavers. 1992. Roy Clive Abraham (1890–1963). Encyclopedia of Language and Linguistics. Oxford: Pergamon Press. In press. Solomon Warsama, and R.C.Abraham. 1951. The Principles of Somali. Published by the second co-author. Writings of Roy Clive Abraham Amharic 1941. A modern grammar of spoken Amharic. Ms., Addis Ababa. 1942. The Principles of Amharic. London: Published by the author through Crown Agents for the Colonies. Hausa 1934. The Principles of Hausa. Kaduna: Government Printer. 1940a. An Introduction to Spoken Hausa and Hausa Reader for European Students. London: Crown Agents for the Colonies (published on behalf of the Government of Nigeria). 1940b. Phonetics and tones of Hausa. Ms., SOAS, University of London. 1941. A Modern Grammar of Spoken Hausa. London: Crown Agents for the Colonies (published on behalf of the Government of Nigeria). [Reprinted 1946, a revised version of his 1934 grammar.] 1949 (and Mai Kano). Dictionary of the Hausa Language. 1st ed. London: Crown Agents for the Colonies. [Reprinted 1962.] 1959a. Hausa Literature and the Hausa Sound System. London: University of London Press. 1959b. The Language of the Hausa People. London: University of London Press. [A reprint of his 1941 grammar.] 1962. Dictionary of the Hausa Language. 2nd ed. London: University of London Press. [A reprint of the 1949 edition, with the name of the second author omitted and a new ‘Preface’ added.] Idoma 1935. The Principles of Idoma. London: Published by the author through Crown Agents for the Colonies.
3
1951. The Idoma Language, Idoma Word-lists, Idoma Chrestomathy, Idoma Proverbs. Nigeria: Published by the author on behalf of the Idoma Native Administration. Igbo 1967. The Principles of Ibo (Archival edition of typescript). (Occasional Publication No. 4.) Ibadan: Institute of African Studies, University of Ibadan. Somali 1951. (and Solomon Warsama). The Principles of Somali. Published by the first author. 1964. Somali-English Dictionary. London: University of London Press. 1967. English-Somali Dictionary. London: University of London Press. Tiv 1933a. The Tiv People. Lagos: Government Printer. [Revised and reprinted 1940, Crown Agents for the Colonies, London.] 1933b. The Grammar of Tiv. Kaduna: Government Printer. 1940a. The Principles of Tiv. London: Crown Agents for the Colonies. 1940b. A Dictionary of the Tiv Language. London: Crown Agents for the Colonies. 1940c. A Tiv Reader for European Students. London: Crown Agents for the Colonies. 1940d. Tiv phonetics and tonal principles. Ms., SOAS, University of London. 1940e. The Bantu features of Tiv. Ms., SOAS, University of London. Yoruba 1958. Dictionary of Modern Yoruba. London: University of London Press. Miscellaneous 1952. Review of La Langue Berbère by A.Basset. Africa 22:255–258. 1954. Review of The Somali Language by C.R.V.Bell. Africa 24:181–182. 1958. Writing African dictionaries. West Africa, no. 2152 (July 12, 1958), p. 659. 1992. Handlist of the papers of Roy Clive Abraham (Ms. 193280). London: SOAS, University of London. [Includes manuscript material on Amharic, Berber, Bolenci [Bolanci], Fulani, Galla [Oromo], Igbo, Kabyle, Tiv and Yoruba.]
R.C.ABRAHAM’S WORK ON AMHARIC David L.Appleyard
1. Introduction and background R.C.Abraham is perhaps best known for his work on West African languages. Between 1941 and 1943, however, as a result of his secondment to the British occupying forces in Ethiopia, R.C.Abraham also prepared three works on Amharic. The first of these, ‘A modern grammar of spoken Amharic’, was circulated in an edition of one hundred copies in Addis Ababa by the Occupied Enemy Territory Administration, Abyssinia. This proved so popular that an enlarged and revised version, which also incorporated some 25 pages of conversational sentences, was produced by Abraham and published by the Crown Agents for the Colonies in 1942 under the title The Principles of Amharic. The latter are also advertised as publishing Abraham’s two other works on Amharic, his ‘English-Amharic vocabulary’ and his ‘Amharic texts for European students’, both in 1943. Of the above, I have only been able to see The Principles of Amharic, which, incidentally, is the only one of Abraham’s Amharic works to appear in subsequent lists of his publications, for example in his Somali-English Dictionary of 1964.1 The following discussion, therefore, is of necessity restricted to his grammatical description of the language. When Abraham turned to the study of Amharic, which he later taught for a while at the School of Oriental and African Studies, there were already available to students a number of grammars and textbooks of the language written in English, as well as in French, German, Italian, and even Latin. These ranged from the ‘practical’ format with exercises, phrases and dialogues, aimed at the traveller, missionary or colonial official, to the reference grammar, essentially based on the written language and directed towards a more academic and scholarly readership, typically in the field of Semitics. It is important to realise at this point that the position of Amharic vis-à-vis its linguistic study is somewhat different from that of the other languages on which Abraham worked. Ethiopia has its own, indigenous literate culture going back to the beginning of the Christian era, if not before, and whilst Amharic has been the principal language of literacy only since the middle of the last century,2 it is intrinsically bound up with and is, in some respects, a natural continuation of the older Ethiopic or Ge‘ez tradition. Amharic was, therefore, already a written language,
African Languages and Cultures, Supplement 1 (1992):5–14 1
It would seem that Abraham was a little premature in advertising his ‘Vocabulary’ and ‘Texts’, neither of which in fact appeared. 2 Attestations of Amharic do, however, predate the mid 19th century.
R.C.ABRAHAM’S WORK ON AMHARIC
5
and, more importantly, a language with its own, ancient script and culture of literacy, when Abraham embarked upon his study. As a living Semitic language, Amharic had also attracted the attention of European Semitists, and whilst the syntax of Amharic would be unfamiliar to a ‘classical’ Semitist, there is much in the morphology, particularly the verbal morphology of the language, that is recognisable and readily analysable within a traditional Semitic format. The first European scientific description of Amharic, Ludolf’s Grammatica linguae Amharicae, indeed, dates from 1698. The scholarly grammars and studies of Praetorius (1879), Guidi (1889) and Cohen (1936, 1939), to mention but the most illustrious, all rest upon both this literate and ‘Semitistic’ tradition. It therefore comes as something of a surprise that Abraham chose not only to abandon the Ethiopic script in his work with a rather dismissive, ‘as the Amharic alphabet is in some respects imperfect phonetically, so no use is made of it here’ (1942: 9), but also to ignore the terminology and apparatus of grammatical description already familiar from European scholarship. Rather oddly, in the ‘Preface’ Abraham only recommends to the student of Amharic his own works, while alluding to only one other, minor writer on Amharic, C.H.Walker, to whom he expresses his gratitude, while at the same time refuting his description of Amharic as a language the learning of which is ‘a formidable task which according to natives is beyond the capacity of normal Europeans’. Walker’s English-Amharic Dictionary (1928), which Abraham mentions, is, incidentally, so beset with errors and idiosyncrasies of arrangement as to render it of little practical use. It does remain difficult to understand, however, so that no mention is made by Abraham of other English works on Amharic, let alone the far more important grammars and dictionaries in French, German and Italian that were already available by the time his Principles appeared. It would seem highly unlikely, however, that he could have been unaware of these. Indeed, his use of the label ‘contingent’ for the aspectual stem that is elsewhere called the ‘imperfect’ probably derives from Armbruster who used the term in his Initia Amharica (1908); only Alone (1909), alongside his preferred ‘constructive present’, Dawkins (1960) and Bender and Hailu (1978), all writing in English, also used the term. Abraham divides his Principles into 43 chapters, including four appendices, two sets of lengthy addenda and corrigenda, and an index. Abraham’s decision to use only his own transcription of the Amharic, and that in a type-face that does not differ from the continuous English text, makes it difficult for the reader to isolate the Amharic from the surrounding sea of English. Of course, the Principles were published at a time that was not easy for book production in general, but the choice of the offset process from the author’s crowded and closely written typescript is far from ideal, particularly in what is after all a textbook for students struggling with a difficult and complex language. The book discusses the phonetics and phonology of the language in some detail over the first three chapters, following Abraham’s insistence that ‘nothing is more important for a real grasp than a thorough understanding of the Amharic phonetic system, including stress, tone and elision…’ (p. iv). Other, earlier grammars such as Armbruster (1908) and Cohen (1936) had also dealt with this aspect in detail, but relating it to the script, which Abraham of course ignores. The majority of the remaining 40 chapters are devoted to inflexional and derivational morphology interspersed with some discussion on basic syntax. Throughout there are copious examples of the usage of forms in context, with much cross-referencing. On the whole, the information is correct and, as would be expected from Abraham’s own inferences, describes a good, careful register of spoken Amharic. There is a slight mixture of dialect forms—it should be added, however, that Amharic does not display a particularly marked degree of dialect differentiation.3 This mixture is apparent, for example, in the variation between the k- and t- forms of the prepositions ke-/ te- and ĭske-/ĭste-(I use Abraham’s transcription throughout except where indicated), which Abraham sometimes uses in free variation, though perhaps with a slight preference for t-. This is exactly what would be expected of the educated speech of Addis Ababa where dialects from different parts of the Amharic-
6
DAVID L.APPLEYARD
speaking regions are to be heard but with a predominance of Shoan Amharic. His use of forms such as ‘it smoked’ instead of ĉese ( in the now usually accepted transcription), or ‘he wrote’ and ‘book’, in preference to and , are also indications of educated rather than rustic speech, whilst instances of the metathesis of initial rĭ (ĭrda and rĭda ‘help’), contractions like ske alongside ĭske ‘till’, wunet alongside ĭunet ‘truth’, demmwo alongside degmwo ‘again’, yĭgellall alongside yĭgedlall ‘he will alongside ‘when he knows’, and so on, are all features of the spoken language. kill’, In the discussion that follows I want to examine in greater detail two areas of his Principles where Abraham’s analysis and conclusions differ markedly from those of other scholars, including those who preceded him and those who have worked on Amharic since his day. The first of these concerns the question of stress and what Abraham calls ‘tone’ in Amharic. It has to be admitted that these questions, the subject of suprasegmentals in Amharic, have largely been ignored or glossed over in most descriptive grammars and detailed studies on the language. Abraham’s treatment almost stood alone at the time, at least in its pervading presence throughout the grammar, and is marked by its idiosyncratic and, with the hindsight of more recent studies on the subject, unusual conclusions. The second topic I wish to look at more closely has already been alluded to: Abraham’s analysis and classification of Amharic verb stems and.classes. This is the most complex area of Amharic morphology and one that has been subject to various detailed studies over the years. It is also the one area of Amharic structure that maintains a large part of the language’s Semitic inheritance intact, and recognisably so, and which therefore has ready made, as it were, a traditional mode of analysis and description. 2. Stress and tone in Amharic Abraham allots a mere two pages (Chapter 3, sections 15 & 16) to a discussion of this question in Amharic, yet the conclusions he draws there not only affect his transcription throughout the rest of the book, but also lead him to make numerous statements elsewhere about the inflexional morphology of the language, which may be at variance with the generally accepted view of Amharic structure as it is understood and taught today. As I have said above, stress and ‘tone’—or, to use a more neutral term—accent in Amharic, have been little studied until quite recently.4 There are a number of reasons for this. Of the descriptions of Amharic that preceded Abraham’s, Armbruster’s 1908 grammar devotes a little over 15 pages to the subject, mostly made up of examples of ‘variation of accent’, which in itself gives a clue to the problem. Armbruster also observed that the ‘accent in Amharic is in general much less marked than in English. Amharic speech, like French, runs with an almost even distribution of stress on each syllable’ (1908:36). This too is part of the explanation why the subject of accent is usually given only a minor place, if any, in grammars of the language. In his Traité de langue amharique (1936), Cohen said much the same as Armbruster, if somewhat more elegantly: ‘On n’y peut noter aucun accent d’intensité ou de hauteur notable à des places déterminées. Les études faites jusqu’à présent corroborent l’affirmation émise ici, soit explicitement, soit implicitement par la diversité des notations. Le discours a peu de relief, il est encore plus égal qu’en français’ (1936:63). Unlike Armbruster, and indeed Abraham, Cohen therefore avoids any form of accent notation in his transcribed material.
3 Amharic dialects have been little investigated (see, for instance, the summary in Bender et al. (1976:90–98)). Some variation which is ultimately attributable to differences of dialect also occurs in written Amharic, e.g. the kä-/täpreposition.
R.C.ABRAHAM’S WORK ON AMHARIC
7
Abraham tries to establish precise rules for accent or stress placement within Amharic and, like Armbruster, has to recognise both different rules for nouns and verbs, as well as a secondary stress in addition to a primary stress in some words ‘due (i) to double consonant [sic]…, or (ii) to final syllable having…high tone’ (p. 19). Rather confusingly, in words with only one stressed syllable he used the symbol ′ following the syllable to mark stress, whilst in words with two stresses ′ marks the secondary stress, while " is used for primary stress: deg’mwo ‘again’, but cĭg’gĭreñ”ña ‘unhappy’. Some of his stress markings seem at odds from what has been and can be observed elsewhere, at least for words in isolation—Abraham does not, however, speak of citation forms or words in isolation: bahĭr’ ‘sea’, amet’ ‘year’, gĭzyε’ ‘time’, all of which one would normally give light stress on the initial, not on the second syllable. Indeed, on the same page (19), he gives the phrase yeba’hĭr dar ‘seashore’, but ascribes this to a ‘rule’ that ‘when a disyllabic noun etc. is followed by a word stressed on its first syllable, the former’s stress shifts back’. This contains a hint to one of the problems that have made Amharic accent so difficult to analyse and quantify: lexical stress is further subject to patterns of phrase and sentence stress—‘Common to all Semitic Ethiopic languages is the “instability” of accent. In all of them the sentence accent predominates over such stress as might be apportioned to the individual word’ (Ullendorff 1955:197). It is also the case that a word in prepausal position may have different stress from the same item within a phrase: in other words lexical stress is not absolute, but is, as I have said, subject to phrase and sentence patterns. Some of this can of course be identified when one examines Abraham’s Amharic sample phrases and sentences throughout the book, but it appears that he failed to realise the importance of such ‘variants’, as Armbruster called them. An important area where Abraham’s view of accent in Amharic led him to make a glaring error was in his treatment of what he calls ‘tone’ (see also Hayward in this volume). I would venture to suggest that Abraham’s primary familiarity with various languages of West Africa led him into the error of seeing pitch functioning in Amharic in an admittedly different but significant way, such that he was compelled to make the statement that ‘Amharic is a tone-language’ (p. 20). It has to be conceded that what he goes on to say qualifies that statement in such a way that it can now be recognised that what he was describing would not be called a tone language in the sense we use today. He adds: ‘…tones not being employed to distinguish meaning as in West Africa, but to support the rhythm of the whole sentence or to indicate the relationship between the parts of the sentence’. Amharic is, in effect, an intonational or stress-accent language. It uses intonational features in association with different utterance-types, thereby assigning high pitch accordingly as a feature of the suprasegmental apparatus of the phrase or sentence. Amharic is decidedly not a tone language in the sense that variations of pitch have a lexical or grammatical function. Interestingly, the other North East African language that Abraham worked on, Somali, is somewhat more of a tone language in the latter sense (though the label ‘pitch accent language’ is probably more accurate), in that it uses ‘tone’ to distinguish certain grammatical categories. Despite the evidence to the contrary that he himself adduces, Abraham continues to think of Amharic as a tone language and adopts a simple form of tonal notation throughout his transcription. Though he seems to be thinking of ‘tone’ as operative over the sentence, supplying short sentences and phrases as he does to illustrate his analysis, he also observes, for example, that ‘trisyllabic nouns etc. have low tone on the syllable following the stressed one’ (p. 21). However, again there is overlooked the question of context, of whether a word appears as a citation in isolation, or as part of a phrase or sentence. Thus, in the example that ‘the road’ (underlining denotes low tone in his transcription), slightly higher pitch is he gives, meŋ’ associated with the weak stress on the first syllable because the word appears in isolation. In the sentence
4
Two recent excellent studies on this topic are Alemayehu Haile (1987a, b).
8
DAVID L.APPLEYARD
which he supplies as an example, ‘he brought the uncle’, the syllable with ‘high tone’ is also the one with ‘primary stress’, yet this is precisely because of the intonation pattern of this basic type of declarative sentence and is not necessarily a feature of the noun form aggwotun. The same pattern sometimes appears in similarly constructed sentences supplied by Abraham elsewhere, e.g. ‘he saw the horse’ (p. 25). Seemingly contrary examples, however, are also recorded by him without ‘he saw the horses’; ‘he saw the wicked child’; comment: (where ^ indicates a ‘rising tone’) ‘he saw a wicked child’; ‘he saw wicked children’. These variations of stress and pitch placement indicate, rather, that it is differing sentence intonation that is operative here and not a question of ‘tone’ at the level of the individual word. Abraham’s insistence on the validity of ‘tone’ as a relevant category in the grammar of Amharic further leads him to formulate rules governing the operation of ‘tone’ as a morphemic feature. When speaking of the formulation of subject and object within a sentence, for instance, he says that ‘both subject and object have inverted tones’ (p. 26), by which he means the switch of tonal patterns from what he has earlier described for nouns in isolation (though of course he does not speak in such terms). Thus, we find ‘dog’ on p. 21, but ‘a dog died’ and ’sa” ‘a lion killed a dog’ on p. 26. Or, when describing the construction of predicative sentences, he says, ‘the subject has inverted tones ‘his friend is a and the rest of the sentence has low tones’ (p. 27), e.g. ‘his friend’. soldier’, but It is self evident from the data, of course, that Abraham had a good ‘phonetic ear’, and in general one cannot doubt the validity of his transcription as regards stress placement and intonation in general. Some of ‘sea’, for instance, cited above—but then we do not his forms in isolation, however, seem odd— know the circumstances of his data collection.5 The important error, rather, is that he seemed to have missed the real role of these suprasegmental features in Amharic, and by his incorporation of stress and ‘tone’ in the description of the morphological and morphosyntactic apparatus of the language, he made his grammar unnecessarily complicated, to say the least, and, what is more important, thereby incorrect. 3. The analysis of the verb in Amharic The other part of Abraham’s description of Amharic that requires comment is his analysis of the verbal system. This is an important area of Amharic morphology, and not only because of its complexity. Like other Semitic languages, much of the process of derivational morphology in Amharic is centred on the verb: the ‘simplest’ or ‘basic’ form of a lexical root is typically expressed as a verb form (the 3rd masc. sing. of the Perfect, in traditional parlance, and of Abraham’s Past Tense). This is reflected, for example, in the organisation of dictionaries, including monolingual Amharic-Amharic dictionaries. The inflexion of the verb takes into account not only such paradigmatic categories as person of the subject, tense/aspect and mood etc., but also incorporates pronominal objects, differentiation of affirmative and negative forms, and, typical of various languages of the Horn, a formal distinction between main and subordinate verb forms. Before any of this can be tackled by the student of the language, however, s/he has to be aware of and appreciate the various stem-types of the verb, not only with reference to the segmental structure of the lexical root (typically expressed in Semitic grammar as a consonantal skeleton), but also as regards the operation and function of derived stems, expressing such categories as ‘passive’, ‘causative’, ‘reflexive’, 5
In his ‘Preface’ he acknowledges the help of only two people in Addis Ababa—the English Chaplain to the British Community, and an Ethiopian working for the British and Foreign Bible Society.
R.C.ABRAHAM’S WORK ON AMHARIC
9
‘reciprocal’, ‘reiterative’, etc. From a didactic point of view, therefore, it becomes imperative that a clear descriptive framework is set up from the outset. This is usually done, following the time-revered Semitic pattern, by setting up for each structural stem-type a matrix with the paradigmatic ‘principal parts’ on one axis and the commonest derived stems on the other axis, such as can be seen, for instance, in the appendices of Cohen’s Traité (1936). Abraham, however, does not adopt this system. Abraham incorporates structural stem-type and derived stem into an essentially linear system of analysis, further including as separate entities what in the traditional analysis are handled within the main framework, to produce an inventory of 131 verb-types with various additional sub-types. Again, if only for didactic reasons, this is undesirable and unnecessary. As an illustration of what I mean, Abraham’s Type 2 is ‘bridle’ which is differentiated in real terms from Type 1 fel’ exemplified by the verb leg’ ‘want’, only in that its medial radical is a labialised velar; differences between the two ‘types’ in the quality of the vowel following that radical are entirely automatic and are conditioned by the labiality of gw or not of 1. If the Ethiopic script is used, or a transcription system more closely reflective of the script, this becomes apparent: in phonemic terms, therefore, what Abraham records as the apparently different yĭleg’ and yĭfel’ (both Jussives) are and (I use the generally accepted ‘pick’ and transcription), which are, of course, structurally identical. So too, verbs like ‘cut’, which are really of the same stem-type and differ only in that the second has a labialised initial and Type 19 . Or again, his radical, are classified by Abraham separately: Type 17 ‘come to an end’ and includes verbs whose initial radical is a vowel Type 21 is exemplified by al’ (underlying *h in Bender and Hailu’s (1978) analysis, representing a ‘lost’ laryngeal). This group he ), yet the only difference in terms of their inflexion is that separates completely from Type 17 ( the latter has a ‘full’ consonant initial while the former has a zero-consonant, or underlying *h (or , or *′, or whatever symbol one chooses to represent the lost initial consonant). Again, the differences in the vowel : ) can be entirely understood with reference to the phonemic following that initial ( , system of Amharic and its representation in the Ethiopic script: after /*′/ and (surface) /h/ (i.e. and ) the underlying vowels /ä/ and /a/ are neutralised and are realised as a. While it would perhaps be -type as a sub-group because of the necessary elisions that occur when a vowellegitimate to see the ‘he picks, will pick’ but ‘he perishes, will perish’—it is final prefix is added— quite unnecessary to separate the two verb-types in the way that Abraham does. Also contributing to the excessive number of verb-types that Abraham sets up is his inclusion of derived stems in his list as distinct and equal categories, and not just the basic derivations. The prefix formatives a-, as-, and tä-, along with the combined patterns of tä-+infix -a- and aC-6+infix -a- and those patterns involving reduplication of a stem radical, are common to many verbs. In addition to these, Abraham also includes derived stems in astä-, an- and tän-, which are significantly rarer and of restricted occurrence. He also includes derived stems in ar-, au-, and ter-, which are not productive prefixes, if indeed they can be so ‘he loafed about’ is perhaps not derived from a root *d-l-d-l analysed at all: his example of audeled’ but represents a partial reduplication of the root w-d-1 with prefix a-! Thirdly, another cause for the expansion of Abraham’s list of verb-types is the way he handles the two inflexional sub-types that are elsewhere (including in Cohen 1936) conventionally called types A and B. This distinction is purely formal, reflecting the presence or otherwise of length or gemination on the ‘he picks’ (type A); (underlying) medial radical of the root in certain paradigmatic forms: ‘he wants’ (type B). The usual pattern is to see type A as ‘basic’ (it has more differentiation, distinguishing the Imperfect stem (Abraham’s Contingent) from the Jussive), and type B as somehow secondary, at least for didactic purposes. Abraham adopts the reverse view: fel’ is his Type 1, and his Type 17, and the latter and similar types
10
DAVID L.APPLEYARD
are described as ‘shrunken’, thus implicitly regarding the other pattern as somehow basic. I cannot see how any of this is helpful. 4. Conclusion: R.C. Abraham’s contribution to the study of Amharic Of course, the factual information that Abraham provides about Amharic is in every way quite correct and betrays his excellent gift as a student of languages, but his analysis of those data is unnecessarily complicated and confused, a fatal error in what is, after all, supposed to be a textbook for learners. There are throughout his Principles many other lesser instances of this kind of thing, which it would only be tedious to reiterate. Together with the layout of the book, in which topics of grammar are introduced in a jumbled way certain not to be helpful to the student—some of the complexities of adverbial subordinate clause formation (‘the use of the Contingent’) are introduced before discussion of the main-clause Present-Future tense form, for instance—this does not recommend Abraham’s Principles as an ideal textbook of Amharic. It has to be remembered that when Abraham wrote his book he was not working on a language that was poorly known or little described by Western scholars. There were already available to the prospective student several descriptions and textbooks of a wide range of quality in various European languages, including English. Abraham’s original contribution to this field is therefore difficult to quantify. The most outstanding feature of his Principles must be the sharpness of his phonetician’s ear which led him to record in minute and sometimes perplexing detail the sound of Amharic in all its sub-phonemic variation. It is surely an unfortunate error on his part, however, that he chose not to make use of the Ethiopic script, which, though indeed not perfect, does at least impose some kind of phonemic analysis. It is also unfortunate that his method of recording and analysing his material is confused and confusing and led him to make the now notorious statement that ‘Amharic is a tone language’, for which, more than anything else, he is now remembered in ethiopianist circles. The Principles of Amharic has now little more than curiosity value, at least to the learner of the language, and, in this respect, it is perhaps instructive to reflect that Cohen’s Traité de langue amharique, which appeared six years before and has since been reprinted, is, on the other hand, still the standard reference grammar for students of Amharic. REFERENCES Abraham, R.C. 1941. A modern grammar of spoken Amharic. Ms., Addis Ababa. ——. 1942. The Principles of Amharic. London: Published by the author through Crown Agents for the Colonies. ——. 1964. Somali-English Dictionary. London: University of London Press. Alemayehu Haile. 1987a. Lexical stress in Amharic. Journal of Ethiopian Studies 20:19–43. ——. 1987b. An autosegmental approach to Amharic intonation. PhD thesis, SOAS, University of London. Alone, J.P.H.M. 1909. Short Manual of the Amharic Language. London: Macmillan. Armbruster, C.H. 1908. Initia Amharica. An Introduction to Spoken Amharic. Cambridge: University Press. Bender, M.L., J.D.Bowen, R.L.Cooper & C.A.Ferguson (eds.). 1976. Language in Ethiopia. Cambridge: Cambridge University Press.
6
C here represents the lengthening or gemination of the following consonant, i.e. the initial radical of the root, e.g. ‘he helped to pick’ from basic . Abraham, incidentally, sees this type (his Type 33 al’ ) as involving an assimilation from .
R.C.ABRAHAM’S WORK ON AMHARIC
11
Bender, M.L., and Hailu Fulass. 1978. Amharic Verb Morphology. (Committee on Ethiopian Studies, Monograph No. 7, Language and Linguistics No. 1.) East Lansing: African Studies Center, Michigan State University. Cohen, M. 1936. Traité de langue amharique (Abyssinie). (Travaux et Mémoires de l’Institut d’Ethnologie XXIV.) Paris: Institut d’Ethnologie. ——. 1939. Nouvelles études d’éthiopien méridional. Paris: Librairie ancienne Honoré Champion. Dawkins, C.H. 1960. The Fundamentals of Amharic. Addis Ababa: Sudan Interior Mission. Guidi, I. 1889. Grammatica elementare della lingua amarica. Roma: Tipografia della Reale Accademia dei Lincei. Ludolf, H. 1698. Grammatica linguae Amharicae quae vernacula est Habessinorum in usum eorum que cum antiqua hac et praeclara natione Christiana conversari volent, edita. Francofurti ad Moenum: apud Johannem David Zunnerum. Praetorius, F. 1879. Die amharische Sprache. Halle: Verlag der Buchhandlung des Waisenhauses. Ullendorff, Edward. 1955. The Semitic Languages of Ethiopia: A Comparative Phonology. London: Taylor’s (Foreign) Press. Walker, C.H. 1928. English-Amharic Dictionary. London: The Sheldon Press.
ON R.C.ABRAHAM’S VIEW OF AMHARIC AS A ‘TONE LANGUAGE’* R.J.Hayward
1. Introduction In the course of a perusal of the first few pages of R.C.Abraham’s Principles of Amharic (hereafter referred to simply as Principles), we encounter an arresting claim to the effect that Amharic is a ‘tone language’. The bare statement of this claim is immediately qualified, as we shall see shortly, but coming as it does from the pen of a linguist with a considerable acquaintance of Hausa, Yoruba, and other well-recognised ‘tone languages’, one is obliged a priori to give it some serious thought, the more so since Abraham has proceeded to mark tone on every single word and utterance recorded in the book. Such a claim, and such painstaking efforts to transcribe pitch phenomena had never been made by any previous scholars of Amharic, though this, in itself, need hardly be seen as remarkable; very little had been said up to that time about such phenomena in the descriptions of other languages of The Horn either, and many of them are indisputably ‘tone languages’.1 In general, Abraham seems to have ignored previous work on Amharic when he wrote his Principles, and this has left him vulnerable to obvious criticisms. However, it is not unknown for an investigator tackling ab initio a previously well-studied field to achieve a genuine breakthrough; the inspired amateur, by-passing the established scholarly tradition, may sometimes gain new insights. Abraham’s claim about tone in Amharic might, therefore, be the good fruit of approaching Amharic in precisely the same way he would have approached some previously undescribed language. This, after all, is what linguists commonly do. 2. ‘Amharic is a tone language…’ The interesting statement to which I have been referring is as follows:
African Languages and Cultures, Supplement 1 (1992):15–28 *
I wish here to acknowledge my profound indebtedness to Alemayehu Haile, a former doctoral research studen t at SOAS, whose work on the intonation of Amharic (1987) has afforded me many insights into the subject. Alemayehu’s data base was far more wide ranging than the material considered here; moreover, the theoretical framework adopted in his thesis is different to that employed here.
ON R.C.ABRAHAM’S VIEW OF AMHARIC AS A ‘TONE LANGUAGE’
(1)
13
Amharic is a tone language as we shall be seeing, tones not being employed to distinguish meanings as in West Africa, but to support the rhythm of the whole sentence or to indicate the relationship between the parts of the sentence. High tone is left unmarked; low tone is indicated by a line below the vowel to which it refers. (Principles, p. 20)
Now it does seem to me that the qualification included here shields Abraham from the charge that he did not know what he was talking about. On the contrary, it demonstrates that he knew very well what was generally understood by the term ‘tone language’, and that he did not regard Amharic as a tone language in that sense. That is to say, in a sense that corresponds to the core phrase of Pike’s subsequent (and now well known) definition: 2 (2)
A tone language may be defined as a language having lexically significant, contrastive, but relative pitch on each syllable. (1948:3)
But Abraham was faced with a problem, for at that period there did not exist any clear-cut typological ideas concerning the use of pitch variation in language. The conceptual framework of phonemic phonology had indeed begun to be applied to the analysis of pitch phenomena some eighteen years before (cf. Beach 1924), but the term ‘tone’ had also frequently been pressed into service for describing the pitch phenomena of languages like English3, and a further twenty five years were to elapse before Abercrombie could say: (3)
The linguistic functions of speech are very varied, but of two fundamentally different kinds. In one case, the function of the speech melody patterns is to be part of the structure of sentences; in the other case, their function is to be part of the structure of words. In the former case, the patterns are called intonation, and in the latter case they are called tone. In every language the function of speech melody is predominantly either of one kind or the other, so that the languages of the world can be divided into two classes, intonation languages and tone languages. (1967:104)4
We may safely assume, therefore, that only quite rudimentary notions in the typology of the linguistic functions of pitch variation were in circulation at the time when R.C.Abraham was transcribing the utterances of his Amhara informants. Confronted with a choice between having to say that the role of pitch in Amharic was more like that in tone languages of the kind he was familiar with, or more like a non-tone language, as, for example, English, he opted for the former, and then proceeded to qualify his choice. To have said Amharic was an ‘intonation language’, even if that term had been available to him, could have suggested that he thought pitch in Amharic behaved as in English, and could have made readers of his book
1
Many of the Cushitic and Omotic languages of the region are typologically best described as having ‘tonal accent’, though the Nilo-Saharan languages of the west and south-west of Ethiopia and some of the Omotic languages employ pitch in the fully paradigmatic way characteristic of a truly tonal function.
2
A discussion of certain inadequacies of this definition and the proposal of a more satisfactory one are to be found in Welmers (1973:79ff.). 3 In many other ways, tonal analysis was very much in its infancy, and so complex is the range of variation in the relationship between a phonological tone and its phonetic exponence that at around the date of publication of Principles competent phoneticians were proposing quite excessive numbers of tones for various languages. For example, Armstrong in her Phonetic and Tonal Structure of KiKuyu (1940) speaks of seven tones in that language. 4 Abercrombie’s dichotomy is in error in implying that tone languages may not exhibit intonation, and is simplistic to the extent that it gives no recognition to the large number of languages having ‘tonal accent’ systems of varying types.
14
R.J.HAYWARD
feel they had no need to pay attention to it.5 So perhaps one possible reading of the statement in (1) is as an indication of the importance of pitch variation in Amharic; and if this were the case, it would be impossible to find fault with it. If the reader of Principles can at least salvage a sense of this from Abraham’s statements about ‘tone’, so that he or she is prepared to pay serious attention to pitch in Amharic, then something very important will have been communicated. 3. Abraham’s ‘inverted tones’ One thing that makes it seem as if Abraham really did think of Amharic as a tone language, sensu stricto, is his practice of assigning pitch patterns to words uttered in isolation, and then accounting for the changed pitch patterns in those words when found in syntactic constructions by means of a sort of ‘tone rule’: (4)
‘Inverted tones’ means low, high replacing high, low etc. [18] or one or more words with all low tones except final, high, stressed syllable [vide 15c.ii]. (p. 25)
‘[18]’ and the relevant partion of ‘[15c.ii]’ referred to in (4) are reproduced in (5) and (6) respectively:6 (5) (6)
Disyllabic nouns, adjectives and adverbs if standing alone have high followed by low tone: for trisyllabics, vide 16d. ‘horse’. ‘dog’. ‘cat’. (p. 21)7 The presence of two stresses may be due…[ii] to final syllable having [as e.g. in 24b] high tone, the preceding syllable being low tone and already containing one syllable for some reason stressed: cĭg ‘gĭreñ’ ‘unhappy’, ’ ‘he saw the road’. (p. 19)8
Now in spite of the difficulties experienced in getting to grips with Abraham’s tersity of style and his excessive use of cross-referencing, something extremely important is being said here; it lies at the heart of any discussion of pitch variation in Amharic. It concerns what we may call a problem of instability. The problem of instability, i.e. of variability in the pitch patterns of words according to context, has been remarked upon before by Western grammarians of the language,9 but (to the best of my knowledge) Abraham’s is the first attempt to treat it as a rule-governed phenomenon, and he deserves credit for this, even though his rules are patently inadequate. The point is this: if we listen to words uttered in isolation, they have one set of pitch properties; but if we listen to them used in constructions, they have other sets of pitch properties. (For the remainder of this paper, I shall employ the term melody to refer to the pitch properties associated with a word or utterance, in
5
Moreover, it is not the case that linguists have adhered to a uniform terminology in this area. Thus, while earlier writers sometimes talk about the ‘intonation’ of Chinese, more recent ones may speak about the ‘tones’ of English (cf. Halliday 1963). There is currently too, to my mind, a potential for confusion in contemporary writing where there is an overlap in the uses of ‘tone’ (properly a phonological term) and ‘pitch’ (properly a term in psycho-acoustics). 6 Abraham’s transcription for Amharic calls for some explanation. With regard to vowels, he departs from common practice in representing the high and mid central vowels (IPA and ) as ĭ and e respectively. The mid front vowel (somewhat more open than IPA [e]) he represents as ε. The transcription of other vowels should offer no problems. With regard to the consonants, we find the following novelties: and ĉ for the alveolar and palato-alveolar ejective stops (IPA [t’] and [t∫’]), and (as in Hausa usage) for the ejective velar stop. Probably the only other consonant symbol needing explanation is c for the plain voiceless palato-alveolar affricate.
ON R.C.ABRAHAM’S VIEW OF AMHARIC AS A ‘TONE LANGUAGE’
15
accordance with the general practice of Autosegmental Phonology.)10 Thus, taking some of Abraham’s examples, and representing the Amharic in his own notation, we have examples such as the following: (7)
a.
Cf.
c.
b. ‘a dog died’ (p. 26)
’ ‘he wanted a dog’ (p. 26)
‘dog’=form uttered in isolation (p. 21)
Observe that this change of melody on the word for ‘dog’ has nothing to do with different surface case (with a lowfunctions (as it could have in certain other languages of the Horn); we have the word high melody) whether it is head of a noun phrase with subject or object function. Furthermore, when we construct a phrase consisting of a noun preceded by an attributive adjective, a numeral, or a demonstrative, we find the noun has a low-pitched melody throughout, e.g.: (8)
a.
b. ‘a big dog’ yĭh ‘this dog’
c.
(p. 26)
‘five dogs’
(p. 42)
(p. 51)
The same would be true in other noun phrase structures headed by wuš’ša; for example, if the modifying element were a genitive phrase or a relative clause,11 e.g.: (9)
a. b.
‘a man’s dog’ or ‘somebody’s dog’ ’heu ‘the dog which is barking’
Moreover, when we observe phrases such as these within sentences, we note further changes in the melodies of individual words,12 e.g.: (10)
a. ‘a big dog died’
(cf.
in (8a))
7 The reference to ‘16d’ reads: ‘Trisyllabic nouns etc. have low tone on the syllable following the stressed one, this stress depending on the position in first or second syllable of the double consonant: wurĭn”ĉa. meŋ’ ’ (p. 21). 8 I find Abraham’s references to ‘stress’ confusing. As a prosodic entity distinct from pitch/tone and gemination there are, as far as I am aware, no phonetic cues in the utterance for its identification. Even the possibility of attaching a more abstract definition to the term, to regard it, for example, as equivalent to ‘accent’ or ‘metrically strong syllable’, is unsatisfactory, and I suspect, far from Abraham’s intention. Ullendorff (1955:196) in discussing this general area, concludes: ‘I find that only in the verb does the accent generally possess any degree of stability: in the case of biliterals it falls on the penultimate, and in triliterals and quadriliterals on the antepenultimate.’ 9 Cf. the statement quoted in the preceding footnote and also to the works referred to there by Ullendorff (1955: Chapter III). 10 For a useful and comprehensive introduction to Autosegmental Phonology the reader is referred to Goldsmith (1990).
16
R.J.HAYWARD
b. ‘five dogs died’
(cf. am
in (8b))
c. ‘the dog which fell has died’ (cf.the relative verb (9b)) (p. 112)
’heu in
Finally, it should also be pointed out that there is nothing lexically idiosyncratic in this behaviour; any disyllabic noun would behave in the same way as the word selected for exemplification here (i.e. wuš’ša). Furthermore, as many later examples will show, the phenomenon is not restricted to nouns or noun phrases. When we look at Abraham’s statement about ‘tone inversion’ in (4), we may note that it is rather curiously stated and seems to contain two, not very obviously related, sub-rules; and that for only one of these is the name ‘tone inversion’ really appropriate. The first part says: H(igh)L(ow) becomes LH; and it is implied that this change takes place in some particular context. Now Abraham does not extract any generalisation about what this context is, though combing through Principles, one finds that in various phrases and sentences in the body of the grammar (not to mention the 630 ‘conversational sentences’ at the end of the book), there are a great many illustrations of what he means by ‘tone inversion’, though he himself does not usually point them out as such. For instance, we find an example of it in comparing the melody in the isolation form of the postposed quantifier hul’lu (which Abraham (p. 98) calls a ‘comprehensive’) with the melody it has in certain sentences: (11)
a. b.
‘all, the whole of, everything’ ‘the whole town was panic-stricken’
c. ‘all the people went away’
(p. 98)
We also find instances of it in what Abraham calls the ‘past participle’, which is a form better known to Ethiopianist linguists as the ‘gerund’ or, more suitably, the ‘converb’, e.g.: (12)
a. ‘he having thrown down’ b. ‘he gave up the work’, literally ‘having thrown down the work, he went away’ (p. 54)
But exactly the same thing happens in the ‘pluperfect’, which is a compound paradigm consisting of the ‘past participle/gerund/converb’ followed by an invariable form of the verb n-b-r, which as an independent verb means ‘was, existed’, though in the pluperfect it simply signals past tense, e.g.:
11
The examples here are my own as I could find no examples of these structures containing this particular word as head noun in Principles. There are, to be sure, many examples of both these structures with other head nouns, and they consistently illustrate the point being made; cf. Principles p. 50, section 79; p. 112, section 225. 12 The examples in (10a & b) are my own.
ON R.C.ABRAHAM’S VIEW OF AMHARIC AS A ‘TONE LANGUAGE’
(13)
a.
17
b. ‘he having worked’
‘he had worked’
(p. 104)
Limitations of space only prevent me from illustrating many more cases of this pattern in a wide range of syntactic structures; but I think a sufficient number of examples has been adduced for us to be able to make an initial generalisation about the context for a HL word to become LH; namely, when it is in penultimate position in the string. The accompanying generalisation (supported by all the examples so far, though not commented upon by Abraham) is that whatever its melody may be when uttered in isolation, the final item in the string is low-pitched. If we return to Abraham’s original statement of the ‘tone inversion’ rule, we find that in addition to the replacement of HL by LH on disyllabic words, there is another sub-rule. His wording is: ‘…one or more words with all low tones except final, high, stressed syllable’, by which, I take him to mean that a string of one or more words will come to have low tones on all syllables except the final one. The cross-reference to 15c.ii permits us to believe that Abraham has the same context in mind for this transformation as for the HL> LH change, for one of the examples adduced there exemplifies a change of HL becoming LLH in just the context we have already noted, namely when the word occurs penultimate in the string, e.g.: (14)
a.
b. ‘the road (accusative)’
” ‘he saw the road’
(p. 19)
More striking examples abound throughout the book, e.g.: (15)
a. work-my-acc. which-I-finish-with-it paper exists-to-me-neg.13 ‘I have not enough paper to finish my work’
(p. 238, sentence 474)
cold at-it-happens time all it-sickens-me ‘whenever it is cold, I am unwell’
(p. 239, sentence 503)
of-house-the-acc. lock that-you-repair-for-me I want ‘I want you to repair the lock for me’
(p. 230, sentence 304)
b.
c.
The generalisation made a little earlier (about high pitch being located on the last syllable of the penultimate word) is correct for the examples cited up till now, but things can be more complicated. In multiphrasal sentences, it is usual (though not obligatory) for each non-final phrase to exhibit this final high pitch. Once again, Abraham’s meticulously pitch-marked examples provide abundant illustrations of this, e.g.: (16)
a. lion dog it-killed ‘a lion killed a dog’ b. in-Amharic language how? it-passive-translates
18
R.J.HAYWARD
‘how is it translated into Amharic?’ (p. 239, sentence 509) c. from-office at-how-many? hour(s) you-get-away ‘at what time do you get away from the office?’ (p. 233, sentence 365)
d. I-having-sought-you past but at-home you-were-not-present ‘I called to see you but you were not in’
(p. 238, sentence 481)
at-you-came time hour-the it-having-passed past ‘it was already late when you came’
(p. 239, sentence 508)
e.
The same thing happens in non-final clauses as well, whether the verb form occurring at the end of the clause is subordinate or coordinate, e.g.: (17)
a. if-not-he-come I-be-angry-with-him ‘if he doesn’t came, I shall be angry with him’
(p. 102)
b. goods it-will-be-lost I-having-said I-having-feared wrapping-the-acc. I-having-tightened I-tied-it ‘I tied the parcel tightly, for fear something might get lost’ (p. 242, sentence 602)
It may well have been on account of the difficulty of encompassing the variety of contexts of its application that Abraham drew back from attempting to define in any rigorous way the circumstances under which his ‘tone inversion’ rule operated. (This, of course, is mere speculation on my part.) We could then simply be critical and make the somewhat trivial observation that the term ‘tone inversion’ is not really appropriate for the linguistic forms under consideration in the second sub-rule; ‘tone inversion’ being only suitable as a label when we are considering disyllabic words with an isolation melody HL. What is much more certain, however, is that Abraham intuitively felt the relatedness of these phenomena such that he grouped them together under a single label. And it should be emphasised that nobody else up to this time had given any explicit recognition to this. Something, however, that has emerged very clearly from our consideration of the longer (sentence-length) utterances is that we are dealing with some kind of intonation phenomenon. Furthermore, I believe that there is a relatively simple explanation for Abraham’s ‘tone inversion’ once we recognise that we are dealing with intonation. I am prepared to argue that Abraham himself recognised that pitch variation phenomena in Amharic were what we would call ‘intonational’, though, possibly for want of the correct term, he called it ‘tone’, like Beach (1924:84, 102) before him. His qualificatory statement that ‘…tones not being employed to distinguish meanings as in West Africa, but to support the rhythm of the whole sentence or indicate the relationship between the parts of the sentence’
13
The morpheme-by-morpheme gloss provided under the Amharic example sentences in (15) and subsequently are my addition, i.e. they do not appear in Principles.
ON R.C.ABRAHAM’S VIEW OF AMHARIC AS A ‘TONE LANGUAGE’
19
leaves me fully convinced that he knew he was talking about a syntagmatic function of pitch variation, rather than a paradigmatic one, and that, on this matter, the confusion is a superficial one of terminology only. With regard to having a clear conception of what intonation is, however, I think Abraham was confused; though in mitigation of that indictment, I would add that many others have been in the same condition. The root problem is that Abraham begins by looking at individual words and attempting to determine ‘basic’ (or in modern parlance ‘underlying’) accentual and melodic properties for them. He then progresses to phrases and sentences and finds that a great many of those word-based properties simply disappear, and are replaced by others, in accordance with the context; hence the need for massive applications of his ‘tone inversion’ rules. As far as pitch patterning is concerned, it is as if he thought that the sentence melody had to be built up out of a string of components each of which has to contribute to the whole, with allowances being made for the influence of neighbouring components. This, of course, is exactly what does happen in true tone languages of the type Abraham was familiar with, and it is his betrayal of this conceptualisation that makes it appear that he does not well appreciate the difference between tone and intonation. 4. An analysis If a pitch melody in Amharic (or in any other language) functions syntagmatically over a whole sentence, then its analysis must proceed from a consideration of the sentence. That is to say, we must start with an analysis of what is happening in the largest unit, and interpret properties of individual words, in context or out of context (i.e. in isolation), in the light of that. An intonation melody is, generally speaking, designed for a sentence-sized (more strictly clause-sized) utterance, so that when an utterance is less than a clause, we might expect to find one of two logically possible accommodations to that situation. One obvious strategy would be to drop part of the melody. The other possibility would be to cram all of the melody onto what is available. In order to be able to refer to the segmental parts of utterances with which a single intonation melody is associated, though which may not necessarily consist of a complete clause, it is convenient to make use of the term ‘word group’. We can illustrate this from English. According to one well-known analysis of intonation in English (O’Connor & Arnold 1961), a single word group: melody association typically divides into two portions, for which the designations ‘Head’ and ‘Nucleus’ are employed. The nuclear portion contains the stressed syllable of the last accented word, and the head portion begins with the stressed syllable of the first accented word (providing this is not also the nucleus) and ends with the syllable immediately preceding the nucleus. These points are illustrated in the English word groups in (18): (18)
a. b.
[‘Why have you] Hd. On [‘Wednesday I visited]
Nuc. 14
Hd.
Nuc.
As stated above, Heads always begin at the first stressed syllable in the word group to which an intonation melody is to be assigned. The Heads illustrated in (18) are melodic elements of the type termed by O’Connor and Arnold ‘High’ Heads. A High Head consists of a period of relatively high level pitch spread over the Head portion of the word group. The High Head is one of a small closed set of melodic elements that could occur preceding a nuclear melodic element of the type selected here. The nuclear element of the melody in this case involves relatively mid and lower level pitches on the two syllables of
20
R.J.HAYWARD
‘Oxford’ respectively. This type of nuclear element is called the ‘Low Fall’; it is one of several such melodic elements which could occur in the nucleus when preceded by a High Head. There is then a limited number of combinatorial possibilities between the melodic elements that can occur associated with the two portions of the word group, which, as I have said, are both present ideally. If, however, the single word ‘Oxford’ is uttered in isolation,15 it is not possible to accommodate both elements of the melody. Since there is no stressed syllable before the first one of ‘Oxford’, and this carries the Nuclear melodic element; there is simply no room for the Head melodic element, and the latter is dropped. Thus, when the word ‘Oxford’ is uttered in isolation, all that it carries is the Nuclear melodic element. Now this has nothing whatever to do with the fact that ‘Oxford’ is final when uttered in isolation, or that it happened to bear the Nuclear melodic element in (18). This becomes clear if we try to isolate any other word from (18). We find that it is simply not possible to utter them (naturally) with a melodic element appropriate for a Head, even though they may have had such a tune when internal to a clause-sized word group. What we do when we pronounce them in isolation is to pronounce them with Nuclear melodic elements. If, however, we replaced ‘Oxford’ in our sentence with, say, ‘Aberdeen’, which is a word with two stresses (a primary stress on ‘…deen’, and a secondary one on ‘A…’), we have a word which when uttered in isolation is able to accommodate both Head and Nuclear melodic elements. Similarly, if instead of ‘Wednesday’ in (18), which has initial stress and can therefore only be pronounced in isolation with a Nuclear melodic element, we were to select an adverbial such as ‘unexpectedly’, which has two stresses, then we would be able to ‘squeeze’ both melodic elements onto it. The important thing to note for English is that the presence of stress is crucial. Both Head and Nuclear melodic elements require a stressed syllable in order to become associated with a word group. If there is only one stressed syllable, the melodic element which is appropriate for association with a Head simply fails to associate, and is dropped (or in more modern parlance ‘receives no phonetic interpretation’). Let us return now to Amharic. I believe it is also appropriate to analyse Amharic intonation melodies as, in principle, consisting of two parts,16 and I shall, for convenience, use the the terms Head and Nucleus here too. The type of Head which we have been seeing many instances of in our survey of occurrences of Abraham’s ‘tone inversion’ has a period of low pitch spread over one or more syllables, followed by a higher pitch on the last syllable. (It should be added here that there are other types of Head melodic element in Amharic, but it is not germane to say much about them here, for the simple reason that Abraham includes very few examples of them.) The Nuclear melodic element in all the cases we have considered is simply a period of low pitch. (Amharic, however, does have a number of other Nuclear melodic elements, but Principles provides almost no examples of them. It should be added that this particular intonation, i.e. a LH melodic element on the Head and a L melodic element on the Nucleus, is extremely common and may well be the commonest in the language.) This analysis is exemplified in (19); (19a) illustrates this intonation melody in association with a sentence; (19b) illustrates it in association with an expanded noun phrase: (19)
a.
‘the dog barked’ Hd.
Nuc.
14 It will be observed that in the example in (18b) there is a stressless word to the left of the beginning of the Head. This portion is usually termed the Pre-head. It could be regarded as a proclitic element on the Head, and presents no difficulties for the analysis. 15 The same situation can arise in English if the focused/accentuated word, which necessarily carries the nuclear melodic element, comes at the beginning of the word group; again, there will be no possibility for the Head melodic element to be associated.
ON R.C.ABRAHAM’S VIEW OF AMHARIC AS A ‘TONE LANGUAGE’
b.
[
]
21
‘the dog that barked’
Hd.
Nuc.
Now the all-important question is: what happens in a one-word utterance? The answer is that Amharic, unlike English, always accommodates all of the intonation melody, both Head and Nuclear elements, as exemplified in (20): (20)
[fe]
‘horse’; [fel] [
] ‘he sought’; [se]
‘man, person’
This is why every Amharic word uttered in isolation contains High and Low pitched portions, and why no word in isolation is ever uttered on an even low pitch. This is also the basis for Abraham’s observation quoted in (5). We might stop then at an analysis as in (20), but this would, in fact, fail to account for two things (about neither of which Abraham ventures an opinion). The first is that the high pitch in the first syllable of a word uttered in isolation is not as high as the high pitch at the end of the Head in an expanded phrase or sentence. Secondly, this analysis fails to say anything about what happens to the low-pitched first part of the Head melodic element we are considering. Does this simply get dropped? There is a straightforward explanation for these things if we say that both the Low and High are, in fact, accommodated on the first syllable. The phonetic effect of this should be a rising pitch, and I believe it is possible to hear just this sometimes, though, more usually, a contour simplification occurs resulting in a mid level pitch; which is why the pitch of this syllable is lower than that of a syllable with High following one or more syllables associated with Low. This analysis for feres ‘horse’ and fellege ‘he wanted/sought’ is diagrammed in (21):17 (21)
a.
LH \/ [fe] Hd.
L | [res] Nuc.
b.
LH \/ [fel] Hd.
L /\ [lege] Nuc.
Two pieces of evidence support this analysis. Firstly, if we consider the isolation utterances of a foursyllable word like zenezene ‘pestle’, we observe a more optimal accommodation of the Head melodic element: (22)
LH || [zene] Hd.
L /\ [zene] Nuc.
16 To talk of a bipartite division is somewhat simplistic. It is clear, both in English and Amharic, that the Head melodic element may be repeated (cf. the examples in (16) and (17b). Coupled with the fact that a Head (in English) may be absent under certain conditions (though a Nucleus will never be absent), it would seem that there is a degree of independence between Head and Nuclear melodic elements. In both languages, Head melodic elements function as indicators of non-finality, and structuring discourse frequently calls for a series of ‘build-ups’ to a nuclear melody. In terms, however, of the number of distinct types of melodic element capable of being associated with a word group, this only ever consists of two.
22
R.J.HAYWARD
Secondly, there is another type of Head melodic element in Amharic, in which a High pitch comes first, rather than second.18 This also co-occurs with the all-Low Nuclear melodic element we have been considering. And as the analysis being suggested would predict, we find a clear phonetic (and phonological) contrast between disyllabic words uttered in isolation with these two Head melodic elements: (23)
a.
LH \/ [fe] Hd.
L | [res] Nuc.
b.
H | [fe] Hd.
L | [res] Nuc.
The difference (though perhaps one should say one difference) between English and Amharic intonation is that given a text which is insufficient, English will abandon the effort of associating all (i.e. both parts of) the melody with it, whereas Amharic will endeavour to get it all onto any text. This difference is, I believe, diagnostic of a deeper typological difference relating to stress.19 English is a language in which the association of intonation to (segmental) text is determined by stress. One could visualise an English text as having a series of pre-designated ‘anchor points’ for the association of particular elements of the intonation melody. Amharic, I would suggest, does not have word stress,20 and its ‘elasticity’ in accommodating clause-sized melodies on single words is on account of this. In support of the claim that Amharic does not have word stress, I would suggest that an examination be made of all the ‘stresses’ that Abraham marks in Principles— both primary stresses and secondary stresses (they have been reproduced in all of Abraham’s examples in this paper). The following questions should then be asked: (1) Is there any greater intensity or duration apparent in these stressed-marked syllables? (2) Is there any sign that vowels in unstressed syllables are undergoing any of the weakening or reduction processes typically found in languages with stress? (3) Do the syllables that Abraham marks as bearing stress show any consistent correlation with the location of any of the following: (a) high-pitched elements; (b) low-pitched elements; (c) the beginning of a pitch change; (d) the end of a pitch change? I submit that there is no consistent correlation. Thus, even if there is something in Amharic that we can call stress (or accent), it is quite independent of intonation. The strongest claim I could make would be that Amharic has no stress/accent, the weakest claim would be that Amharic does have stress/accent but that it plays no role in the intonation of multi-word word groups. I could go further and say that whereas in English the association of intonation with text is ‘stressdriven’, the association in Amharic is driven by the intonation itself. There are cases in Amharic of some intonation melodies the association of which actually triggers the creation of (meaningless) segmental material to carry the melody; but this would be beyond anything recorded in Principles. Yet strangely enough (though I would not care to press the interpretation too hard), my last suggestion comes rather close to Abraham’s statement that ‘Amharic is a tone-language…tones…being employed…to support the rhythm of the whole sentence…’.
17
The diagrams in (21–23) follow the usual conventions of Autosegmental Phonology in which tonal elements and segmental elements are diagrammed on separate ‘tiers’, and are linked by means of association lines which allow for oneto-many as well as many-to-one associations between the two sets of elements. 18 Perhaps this melodic element consists only of the high pitch; I am not sure. 19 20
More strictly, we should term this accent, rather than stress. Cf. footnote 8.
ON R.C.ABRAHAM’S VIEW OF AMHARIC AS A ‘TONE LANGUAGE’
23
REFERENCES Abercrombie, David. 1967. Elements of General Phonetics. Edinburgh: Edinburgh University Press. Abraham, R.C. 1942. The Principles of Amharic. London: Published by the author through Crown Agents for the Colonies. Alemayehu Haile. 1987. An autosegmental approach to Amharic intonation. PhD thesis, SOAS, University of London. Armstrong, Lilias E. 1940. The Phonetic and Tonal Structure of KiKuyu. London: Published for the International Institute of African Languages and Cultures. Beach, Douglas M. 1924. The science of tonetics and its application to Bantu languages. Bantu Studies, 2nd series II: 75–106. Goldsmith, John A. 1990. Autosegmental and Metrical Phonology. Oxford: Blackwell. Halliday, M.A.K. 1963. The tones of English. Archivum Linguisticum 15(i):1–28. O’Connor, J.D., and G.F.Arnold. 1961. Intonation of Colloquial English. London: Longmans. Pike, Kenneth L. 1948. Tone Languages: A Technique for Determining the Number and Type of Pitch Contrasts in a Language, with Studies in Tonemic Substitution and Fusion. Ann Arbor: University of Michigan Press. Ullendorff, Edward. 1955. The Semitic Languages of Ethiopia: A Comparative Phonology. London: Taylor’s (Foreign) Press. Welmers, Wm. E. 1973. African Language Structures. Berkeley & Los Angeles, California: University of California Press.
R.C.ABRAHAM: THE BOLEWA AND BOLANCI* John E.Lavers
1. Introduction Captain Roy Clive Abraham entered the Administrative service of the Northern Provinces of Nigeria in 1924/5, when he was 35 years old.1 It was a surprisingly late age for a man to be employed as an ADO, especially in view of his qualifications. Abraham was a linguist, had an Honours degree in Arabic, and had studied ‘several European languages and Chinese’ (personal file [p.f.], SOAS). After University, he served in the Indian Army for some ten years during which time he acquired proficiency in half a dozen Asiatic languages, including Persian and Hindi. All this and he was to be a simple ADO—indeed in one source (note in personal file) he is referred to as a ‘Cadet’! How one would like to have greater detail on his earlier, preNigerian life. Given his skills in Arabic, it is perhaps surprising that he was not posted to Borno or Sokoto. Instead he was sent to Bauchi Province, which at the time also included Plateau and had its headquarters at Jos/ Naraguta. His early months of service, including the date of his first arrival, are all unknown since his personal file has been misplaced in the National Archives Kaduna. However, in 1926 he was sent on a tour of the Bolewa areas of the Province to study the history and other aspects of this group, but he also took advantage of his travels to undertake a study of the languages of the peoples he encountered. It cannot be doubted that Abraham had already acquired a working knowledge of one African language— Hausa. This was a requirement for all officers in the North, and without a basic knowledge of Hausa he certainly would not have been sent on tour. However, it is unlikely that he had embarked upon the deep investigations that would eventually lead to his later publications—certainly I know of no research materials dating from before this tour. Abraham made his tour, prepared a report, and in an unsolicited appendix included an extensive study of the Bolanci language.2 This was the first study of an African language by Abraham for which we have evidence. While the main report has been utilised by historians and others over African Languages and Cultures, Supplement 1 (1992):29–36 *
I would like to thank the School of Oriental and African Studies, University of London for kindly allowing me access to R.C. Abraham’s file which deals primarily with his application for the post of Lecturer in Hausa. Without the information contained in his file, my paper would have been very different. 1 Abraham lists both 1924 and 1925 in his personal file. 2 The modern spelling of the Hausa designation ‘Bolanci’ has been used in preference to the older variant ‘Bolenci’. Neither form is traditionally used by the people themselves.
R.C.ABRAHAM: THE BOLEWA AND BOLANCI
25
the years, the Bolanci study has lain unpublished, unused, and unknown even until very recent times. It is the purpose of this paper to attempt to redress this neglect and to bring the data to the attention of linguists unfamiliar with the riches of the National Archives Kaduna. I had anticipated that Russell Schuh would present a linguistic assessment of the Bolanci material and perhaps outline what had been done before the time of Abraham. Since he was unable to attend the Symposium, however, I have modified my paper to take into account some of these matters. 2. Abraham: the tour Abraham arrived in the Northern Provinces in 1925, though the precise date is unknown. By early 1926 he was at the headquarters of Bauchi Province, and on January 6, 1925 the Resident, Captain Lonsdale, wrote to ‘His Honour’ that ‘I intend seconding Capt. Abraham for research work regarding the Bolewa and preBolewa history in Gombe.’ It was not until March 11 that all the necessary administrative matters, approval etc., had been completed, and he was able to set out for Gombe. In the interim period he had had extensive consultations with Major Edgar and Mr Best. Edgar should need no introduction to the reader but Best, who is less well known, was an administrative officer with a considerable knowlege of the Bolewa of Gombe (N.A.K., S.N.P. 17/K1119, vol. I). Perhaps at this point we should pause and ask—who are the Bolewa? This is not the time or place for a detailed account, so let it suffice to note that they are a Chadic-speaking people divided into several groups. Those living in Fika, in close proximity to the Ngamo, Karekare and Ngizim in southwestern Borno, are the only community to have retained their independence as a consequence of the 19th-century jihads. The majority live to the south across the Gongola river in Gombe, where they have intermixed with the Tera. The two groups must have lived intermingled for a very long period of time as many Bolewa and Tera are equally at home in both languages. Indeed, what might be called the Fika ‘national anthem’—a song that refers to the wars against the Gombe Fulani in the early 19th century—is a mixture of Bolanci and Tera and is largely incomprehensible to younger generations. The Gombe Bolewa were divided into a number of small polities before they were subjugated by the Fulani, and the people of Pindiga, an important Jukun outpost, were also brought under Gombe at the same time. The Jukun are, of course, speakers of a Niger-Benue language, but notwithstanding the legends of origin of the Bolewa, the Tera and Jukun peoples are closely associated. Finally, we should note that in former times the Jukun and the Tera were non-Muslim while the Bolewa of Gombe were of mixed religions. The people of Fika had been part of Dar al-Islam since the end of the 17th century. Why was Abraham’s tour undertaken in the first place? There is no direct evidence, but this was the period in which Sir Hugh Clifford, Governor of Nigeria, was mounting an attack on the Lugardian policies of indirect rule in general, and against what was seen as victimisation of the non-Muslim areas in favour of the Muslim Emirates. This policy was in particular to be carried out far more thoroughly in the 1930s under Governor Cameron, at this point Clifford’s Civil Secretary, but no matter the North felt the pressure. We can safely assume that Abraham’s mission was a sop on the part of Kaduna to Clifford and his progressive policies. The tour began at old Gombe on March 21, thence to Kafarati or Geri Kom, capital of one of the prejihad Bolewa polities, then to Gadam, Bojude, Dukku, Old Gombe, along the northern bend of the Gongola to Nafada, then north across the river to Fika and back down the eastern part of the Emirate to Kupto and across to Pindiga.3 Abraham wrote: ‘Every village of the slightest importance was visited or knowledgeable
26
JOHN E.LAVERS
representatives interviewed’. He had also intended to visit Bui and Shillem in Borno, but ill health forced him to make an early return to Bauchi. We are fortunate that Abraham made several references to his working methods, e.g.: The method adopted was in every case, to let my informant tell his own story from his own knowledge without in any way suggesting to him any conclusion I might have formed; in this way I received some striking unsolicited testimony on several points about which I had formed theories in the earliest portions of my investigations. A very large amount of materials was collected in this manner and a very considerable amount of time has been necessary to sift the evidence and separate the grain from the chaff. It has been my aim to form a constructive history of the Bolewa as far as the somewhat meagre evidence available permits; I have therefore drawn such conclusions as seem to me the logical outcome of such evidence, but in order that these conclusions might be compared with the matter from which they are constructed, I have appended a verbatim transcript of all the relevant evidence. As we have seen, poor Abraham became ill and had to retire from the field; indeed he was invalided home six weeks before his scheduled leave, a decision which suggested a serious illness. He reported that he was ‘suffering from ill health during my first month at home, and in spite of every effort, was unable to get any real work done’. No matter, he continued under pressure to complete the report, and while he replied it would soon be ready, he was forced, in August, to approach the Colonial Office for help with the typing. They were unable to give him support, although Leo Amery, the Colonial Secretary, suggested he hire a private typist and apply for a refund of expenses. The report reached Kaduna on October 14, 1926, and was apparently so well received that attempts were made to get it published. There was correspondence with the editor of the Journal of the African Society in February 1926 and on several future occasions, but it came to nothing. It was in the next few years that Abraham seems to have been associated with Bargery in his labours to produce his great Hausa-English Dictionary (1934). Abraham did not forget his researches on the Bolewa and reworked the material, incorporating new data based upon further visits to the area, in particular to Pindiga. Already in the preliminary draft he had displayed an interest in the Jukun—this was the age of Elliot Smith, diasporas, Egypt, Divine Kingship and so on, all matters of academic fashion, not to say orthodoxy, at the time. C.K. Meek was preparing his monumental study of the Jukun that was to appear in 1931 as A Sudanese Kingdom. Not surprisingly, all these features were emphasised in Abraham’s revised text. At this time he was Acting Anthropological Officer—recognition of his special skills but hardly a position for a man now forty years of age. He was still en poste in Kano when, on October 12, 1930, he forwarded an ‘entirely new edition of my original Bolewa Report, as prepared for the press’. Somebody in the Secretariat in Kaduna suggested in a minute that it might be published in something called ‘J. of African Languages Institute’, or in ‘Africa’. Abraham himself sent a note that the ‘monograph is to appear in the “Journal of the Royal Anthropological Institute”, but not in its entirety; Professor Seligman has asked me to abridge [certain pages that are] historical and not ethnological’. He hoped all the excluded matter would be ‘published later in book-form as part of a larger scheme embodying other cognate matter’, but it was not to be, and his monograph remains in the Archives.
3Part
of Abraham’s 1926 itinerary through the Bolewa and related areas reads as follows: March 21: left Doma for Kafarati; March 22–25: at Kafarati, visited Gabukka; March 26: to Gadam; March 27: at Gadam; March 28: visited Tappi; March 29: Gadam to Jore; March 30: Jore to Bojude; March 31: at Bojude; April 1: Bojude to Dirri; April 2: Dirri to Zenge; April 3: at Zenge.
R.C.ABRAHAM: THE BOLEWA AND BOLANCI
27
The only fragment of his work that saw printer’s ink were a few paragraphs on the Bolewa and Jukun migrations in the introductory section. In his The Tiv People (1933), there is a reference to his ‘Report on Boli and Pindiga Jukun’, 1926. To be published shortly’ (p. 20, fn.). Vain hope. 3. Abraham and Bolanci I have thus far concentrated upon the main report and its unhappy publishing history, firstly because it illustrates a typical example of the problems that faced or often seemed to face Abraham, and secondly because it is the material in this report that has made historians in the past two decades become aware of him as an invaluable source for the pre-colonial history of the region. It was in the course of such work in 1965 that I first became aware of its value. I noted the extensive linguistic material at the end and took a week of valuable time to laboriously transcribe the data—there were then no xerox machines north of Ibadan at the time—in black and red. Some years later I showed it to David Arnott at SOAS and he informed the late Johannes Lukas of Hamburg University. Lukas borrowed my copy and made use of it in his studies, and Paul Newman and Russell Schuh have also consulted it. The study forms the first known piece of linguistic research by an indefatigable student of language, and it is my personal opinion that we should also mark the centenary of Abraham’s birth by arranging the publication of his Bolanci material together with a commentary by a qualified linguist. Abraham’s itinerary indicates that he stayed in Fika on April 23–26, 1926, and this would seem to be an amazingly short time for such a detailed report, especially if we remember that he also had to undertake his research for the Government. On the other hand, he had spent a considerably longer period among speakers of Gombe Bolanci. We are fortunate to have his own account of his field methods: I also took the opportunity of studying the languages spoken at Kafarati, Kalam, Fikka [sic], Pindiga and to a certain extent the Tera spoken in Bage which is a dialect of the Tera of Gwani but from which it differs in many particulars. These grammars which will be submitted as soon as they can be copied out and classified will, I trust, be found of interest and importance in studying the affinities and origins of the Haussas [sic] and Bolewa; each language is the work of three or four days only but each has been several times re-checked by speakers of the tongues in question; in each instance, the total number of words and phrases amounts, at a rough computation, to two thousand; a special study of the verbal system has been made involving a complete conjugation of 150 verbs; the numeral system, adjectives and nouns in masculine, feminine, singular and plural appear as well as the principal prepositions, adverbs and conjunctions; the whole is provided with phonetic transcription and accentuation of stressed syllables. Given the complex body of data suggested by the above, it is not surprising that when he approached the Colonial Office about the possibility of getting the report typed, he warned them that ‘the first part is straightforward English while the latter part consists of languages with many diacritical marks: I should prefer to type this portion myself.’
28
JOHN E.LAVERS
4. Bolewa studies before Abraham Abraham’s study was not unique. There had been a surprising amount of earlier material for which one would search in vain through his publications and unpublished writings. The earliest vocabulary I am aware of was collected from a freed Bolewa of Fika by Sigismund Koelle in Freetown and published in his Polyglotta Africana (1854). In passing it is interesting to note that there is no reference to the presence of Gombe Bolewa in Freetown, although there were several from Fika. Dr Overweg visited the town of Gombe in 1851 and collected a vocabulary (see Benton 1912 [1968]), subsequently inherited by Heinrich Barth (1857– 59 [1965]), who seems also to have collected material on his own account from the area. According to Cust (1883:260), ‘Barth left vocabularies of the Fika, Bedde and Keri-keri languages’ (see also Benton 1912:2). They later came into the possession of Gustav Nachtigal and from him they passed to his nephew Rudolf Prietze, though I have no idea of where they are now. Early in the Colonial period Captain (later Major) Merrick visited Fika and collected both historical and linguistic data. Some of the latter were utilised, for comparative purposes, in his Hausa Proverbs (1905). The anthropologist and administrator P.A.Talbot was passing through Fika in 1911, when he had an accident that confined him to his bed for a month, and he utilised the time to gather historical material and almost certainly data from local languages. None of this has been published however. P.A.Benton, Assistant Resident in Bornu Province, prepared a series of ‘Notes on Bolanci’ based primarily upon information from Moi Fika, Disa Dolai, Mallam Alhaj and Private Moman Bara. My contemporary informants have suggested that the latter was the main source of Benton’s data since numerous ‘Bara-isms’ are recognisable. The file remains in the Archives but Benton was allowed to publish the substance in his Notes on Some Languages of the Western Sudan (1912:1– 37). Whiteley, in his ‘Special report on Fika Emirate, Bornu Province’ (1919), makes a few remarks of little value and thereafter we have nothing until Abraham himself. C.K.Meek’s well known work Tribal Studies in Northern Nigeria, which contains a chapter on the Bolewa and their neighbours, together with a Bolanci vocabulary, was not published until 1931; it was however based upon his two or three months’ fieldwork in the area in 1926 (note the year). It would appear therefore, that their visits to Fika coincided, but neither Meek nor Abraham makes reference to the presence of the other. Meek is still remembered, and Abraham is known to the western educated as the (co)author, with Mai Kano, of the Dictionary of the Hausa Language (1949); but nobody is aware that he visited Fika or made a study of Bolanci. BIBLIOGRAPHY Abraham, R.C. 1933. The Tiv People. Lagos: Government Printer. ——. 1949 (and Mai Kano). Dictionary of the Hausa Language. London: Crown Agents for the Colonies. [Reprinted in 1962 as second edition, University of London Press, London, with the name of the second author omitted.] ——. 1959. The Language of the Hausa People. London: University of London Press. Armstrong, Robert G. 1964. Roy Clive Abraham, 1890–1963. The Journal of West African Languages 1(1):49–53. Bargery, G.P. 1934. A Hausa-English Dictionary. London: Oxford University Press. Barth, Heinrich. 1857–59 [1965]. Travels and Discoveries in North and Central Africa, 1849–1855, 3 vols. New York: Harper and Brothers. [Reprinted in 1965, Frank Cass, London.] Benton, P.Askell. 1912 [1968]. Notes on Some Languages of the Western Sudan. London: Oxford University Press. [Reprinted in 1968 as The Languages and Peoples of Bornu, Frank Cass, London.] Cust, R.N. 1883. A Sketch of the Modern Languages of Africa. London: Trübner & Co.
R.C.ABRAHAM: THE BOLEWA AND BOLANCI
29
Hair, P.E.H. 1965. A bibliography of R.C.Abraham—linguist and lexicographer. The Journal of West African Languages 2(1):63–66. Koelle, S.W. 1854. Polyglotta Africana. London: Church Missionary Society. Lange, D. 1987. A Sudanic Chronicle: The Borno Expeditions of Idris Alauma (1564–1576) According to the Account of Ahmad b. Furtu. Stuttgart: Franz Steiner. Lavers, John E. 1980. A note on the terms ‘Hausa’ and ‘Afuno’. Kano Studies 2(1):113–120. Macleod, O. 1912. Chiefs and Cities of Central Africa. London: William Blackwood. Meek, Charles K. 1931a. A Sudanese Kingdom. London: Kegan Paul, Trench, Trübner & Co. Ltd. ——. 1931b. Tribal Studies in Northern Nigeria, 2 vols. London: Kegan Paul, Trench, Trübner & Co. Ltd. Merrick, G. 1904–5. The Bolewa tribe. Journal of the Africa Society 4:417–426. ——. 1905. Hausa Proverbs. London: Kegan Paul, Trench, Trübner & Co. Ltd. Nachtigal, Gustav. 1971–1987. Sahara and Sudan, 4 vols. London: Christopher Hurst. [Translated by G.B.Allen, B.Fisher and H.Fisher.] Newman, P. 1969. Linguistic relationship, language shifting and historical inference. Afrika und Übersee 53:217–223. Temple, O. &. C.L. 1922. Notes on the Tribes, Provinces, Emirates and States of the Northern Provinces of Nigeria. Lagos: CMS. [Reprinted in 1965, Frank Cass, London.] UNPUBLISHED SOURCES Abraham, (Capt) R.C. 1926. Ethnographical notes on the Bolewa group, N.A.K., S.N.P. 17/8, K1119, 2 vols. ——. 1946. Personal file, School of Oriental and African Studies. Hewby W.P. 1903. Bornu Province, monthly report Apr-May, 1903, N.A.K., S.N.P. 15/1, 48A. Meek, C.K. 1931. The Bolewa of Fika, N.A.K., S.N.P. 17/2, K1119, vol. 2, 1931. Whiteley, W. 1919. Special report on Fika Emirate, Bornu Province, 1919. [Copy in my possession, J.E.L.] Yakubu b. Halilu. 1929. The history of Buba Yero of Gombe, trans. by R.C. Abraham in Historical Notes of Gombe Division, Bauchi Province (1929–30). N.A.K., S.N.P. 17/2, 1929. 12250.
HAUSA ORTHOGRAPHY AND ABRAHAM’S TRANSCRIPTION J.Carnochan
The fact that this Symposium is being held at all is, I think, a measure of the importance of Abraham’s work on African languages. Other scholars, like Bargery, have made their academic work the study of a single language, but Abraham has ranged over a number of languages in depth and in widely separated areas. His attitude to language was broadly catholic and his presentation encyclopaedic. In the ‘Preface’ to his Dictionary of Modern Yoruba (1958), he writes: This dictionary covers every aspect of Yoruba civilization: it therefore includes countless idioms, current phrases, proverbs and riddles: further, I have fully explained the historical, religious and ethnological facts which form the background of the vocabulary and without which, mere knowledge of lexicology would be meaningless. The needs of the student of nature have been catered for from the living organisms by detailed description of plants, trees, flowers, animals, insects and reptiles, these being illustrated by four hundred pictures. (p. iii) In the ‘Preface’ to his Dictionary of the Hausa Language (1962) he writes: The language is treated from the point of view of the Hausa speaker who uses it and clear explanations have been given of all customs and technical usages such as marriage, divorce, building, tailoring, cooking, etc. The tones have been shown in all examples throughout the Dictionary for this is a sine qua non of any tone language. (p. v) I note his attention to the Hausa speaker, rather than user or reader, and shall concentrate my observations on his transcription and its relation to the orthography. Abraham was a practical man when it came to matters of orthography, and he worked through a period when the way of writing Hausa was changing. In The Principles of Hausa (1934), he used capital B, D and K for what are now , and , and used lower case letters only elsewhere. This was rather in the nature of a transcription, but shows his freedom from the tyranny of symbols. He tried dots under the letters, and then the high commas in vogue during the years he worked with Bargery. Bargery’s A Hausa-English Dictionary (1934) obeys a simple principle. Each head word is given in the orthography of the day. The pronunciation is given in square brackets including vowel lengths and tones. If there is more than one pronunciation, these are given in turn and the meanings dealt with under sub-headings
African Languages and Cultures, Supplement 1 (1992):37–41
HAUSA ORTHOGRAPHY AND ABRAHAMS TRANSCRIPTION
31
I, II, etc. The examples are only given in spelling with no transcription. There is no doubt that Abraham’s dictionary owes a very great deal to Bargery’s (see also Roxana Ma Newman in this volume), but perhaps Abraham was more interested in how the language is spoken, or perhaps he was looking for ways of saving space. Whatever the reason, the fact is that the user often has to recover the spelling of an item for ‘donkey’. From this we know that both vowels himself from the transcription, as with the head word are long and that the tone-pattern is low-high; we can also easily deduce that the spelling is jaki. Sometimes it is not so easy. We should have to know a lot about the phonology of Hausa before we could deduce the from the head word ‘brother’, or the pronunciation where the high spelling of and (I prefer Bargery’s evaluation of the comma indicates a glottal stop. These two examples, -u- as short rather than long), are typical of those entries where there is only one pronunciation, and where Abraham gives the transcription only, without the orthography. In the case of sa’a the spelling form is given. As a native speaker of English, I do not know what the item ‘wind’ means until I am given its pronunciation; or I do not know how to pronounce it until I am given its meaning. Similarly, a native speaker of Hausa has ‘hour, time’. B. ‘P. of two possibilities when faced with sa’a. Abraham arranges these as: A. about one’s own age’. Here Abraham recognises a glottal stop when between vowels, but not between a above and ‘half-brother by same father’, closely following consonant and a vowel, as in and . He adds d.f. , and d.f. , , and Bargery and ignoring the spelling perhaps thought that a sufficient indication of the pronunciation, assuming that the user would already know and that -ŋ cannot in Hausa be immediately followed by a phonetic vowel. In such items as ‘that’, he gives two high commas, in recognition of the fact that the glottal stop is prolonged. within a single list, and similarly those Like Bargery, Abraham kept words beginning with b and beginning with d and , k and , and y and ’y (Newman and Newman (1977) were the first to make a separation). Like Bargery, he also confused or failed to sort out final long and short vowels on a low tone. I have written on this topic at length (Carnochan 1951, 1952), and Abraham acknowledged my findings in later editions (1962: v). The matter is of some importance in his ‘Table of Verbs’ at the front of the dictionary, where it is repeated from other publications, as the user is misled into thinking that the final low vowel of a verb is short before a pronoun object, when it is regularly long. He followed Bargery closely in his tone marking, and both failed to mark final long vowels on a low tone in hundreds of examples (the Newman and Newman dictionary now does this accurately). Abraham also makes some mistakes in final vowel length on a high tone. In A Modern Grammar of Spoken Hausa (1941), he gave on p. 38, Section 68, ‘after my departure’, and adds a note in brackets (short final under bāyā ‘back’, the example bāya -a in bāya!). Why the note, and why the exclamation mark? Bargery (p. 98) gives baya (bāya) (i.e. with a short final vowel), 1. (adv.) Back, backwards; and 2. (prep.) baya ga Audu ‘after Audu’s departure’. Later in his dictionary, Abraham gave baya A. (bāyā), m. the back, and B. (bāya) (1) adv. backwards, so he noted a difference in this case between the use of the item as a noun and an adverb. Similarly, the final vowels of safe, dare and rana, among others, are short when adverbs, but Abraham always marks them as long. For example, in da dare ne ya zo ‘he came at night’, the final vowel of dare is short, while in dare ya yi ‘night has come’, it is long. It is difficult and perhaps impossible to assess how much Abraham and Bargery owe to each other in the field of accurate phonetic observation. In his ‘Compiler’s Introduction’, Bargery has: 5. g and k before a front vowel i are palatalized, i.e. articulated nearer to the hard palate than the normal k and g; e.g. doki ‘a horse’, sounds like dokyi. (p. xxii)
32
J.CARNOCHAN
A short e is rarely heard in a closed syllable, it being usually replaced by a short a, or when following the letter k or g by ya, e.g. wannan mace, but macan nan; wannan mage, but magyan nan; wannan keke, but kekyan nan. (p. xxiv) Abraham certainly followed these observations closely in his publications, and we find: damō; bēkē, nā ga wajaŋ m. (2) wajan prep.
; shānū; , waje B. (wajē) (1) wajēna ba ‘he is on the side of Audu, not on my side’.
This last example is interesting for three reasons: (1) Abraham recognises that the final nasal of wa jaŋ is velar before a word beginning with a vowel letter (in fact with a glottal stop); (2) he recognises that all vowels before the possessive pronoun suffixes -na/-ta ‘my’ are long; and (3) he does not recognise that the vowel of -na is long here, when it is non-final in the phrase; when final, it closes with a glottal stop. Before leaving the pronunciation and transcription of the vowel e, I must draw attention to the forms of (Hausa Literature and the Hausa Sound System, p. 25), and Amrem on wajeŋ the ‘Contents’ page of this same volume. The long vowel [ē] is a front half-close vowel, and the short vowel associated with it is a front half-open vowel [ε]. I believe that when Bargery and Abraham wrote it as ‘a’, they were confusing it with the English vowel in, say, ‘hang’, an open front vowel, rather than with the Hausa short [a], which is an open central vowel. In assessing Abraham’s contribution, it is evident that he modelled his transcription on Bargery’s, and took into account the phonetic details found in the ‘Compiler’s Introduction’. He made full use of these in the forms of his running texts. We are lucky that he put so much into transcription, contrary to Bargery. On the plus side, he did not follow Bargery with his unnecessary mid-tones (see also Roxana Ma Newman in this volume). As an example of his work, I have chosen p. 101 of his Hausa Literature and the Hausa Sound System (1959). The seven sections of running texts use the same transcription as his Dictionary of the Hausa Language (1962), while Section 3 (‘Idiomatic Sentences’) gives each example first in orthography and then in brackets in his transcription. The language data suitably illustrate six points of importance regarding Abraham’s transcription: A. Final nasal consonants, where the word is linked grammatically to the following word are written: ; . 1. m where the next word begins with a labial consonant, e.g.: cikim 2. n where the next word begins with an alveolar or palatal consonant, e.g.: in darē yā yī; gidan sarkī; kuŋ gan sun yi barcī. 3. ŋ where the next word begins with a velar consonant, a vowel letter (phonetically with a glottal stop), barcī; ; and on p. 103, im bā or h-, e.g.: sarkiŋ gidā; ap, tun mātā ba. Where the word is not linked grammatically to the following word, the orthographic ‘n’ is regularly naŋ ya transcribed as ‘ŋ’, e.g.: naŋ ; naŋ ta ji naŋ, . Bargery gave a detailed account of these phonetic facts in his ‘Compiler’s Introduction’, and it seems likely that Abraham just applied them. B. Treatment of orthographic ‘ai’. Bargery mentions two pronunciations [ai, ei], but hardly followed this and săi , but not with mai gidā nor with up in his entries. Abraham uses ai and ăi; I agree with făifăi. The distribution of [ai] and [ei] still needs to be established.
HAUSA ORTHOGRAPHY AND ABRAHAMS TRANSCRIPTION
33
C. Abraham marks question intonation on the last syllable of the example with a fall. This is an oversimplification and needs a more detailed approach, beyond the scope of a dictionary. D. He ignores the glottal stop, except within a word and between two vowels as in sa’a above. Considering the other details of his transcription, I think this a serious omission. E. Following Bargery, he usually transcribes orthographic ‘e’ in a final closed syllable as ‘a’, e.g.: ; wajē, wajam naŋ. As noted above, there are examples (e.g. p. 25) where the transcription is wajeŋ. and (line 2), F. Many final vowels on a low tone are long, but not marked so. This applies to and (line 3), (line 4), and (line 5). In the end, an accurate transcription depends upon careful listening and evaluation within a systematic framework. Among Abraham’s considerable attributes as a great dictionary maker, he did not, in my view, carry this ability at a very high level. REFERENCES Abraham, R.C. 1934. The Principles of Hausa. Kaduna: Government Printer. ——. 1941. A Modern Grammar of Spoken Hausa. London: Crown Agents for the Colonies (published on behalf of the Government of Nigeria). ——. 1958. Dictionary of Modern Yoruba. London: University of London Press. ——. 1959. Hausa Literature and the Hausa Sound System. London: University of London Press. 1962. Dictionary of the Hausa Language. 2nd ed. London: University of London Press. Bargery, G.P. 1934. A Hausa-English Dictionary. London: Oxford University Press. Camochan, Jack. 1951. A study of quantity in Hausa. BSOAS 13(4):1032–1044. ——. 1952. Glottalization in Hausa. In Transactions of the Philological Society, pp. 78–109. London. Newman, Paul, and Roxana Ma Newman. 1977. Modern Hausa-English Dictionary. Ibadan & Zaria: Oxford University Press (Nigeria).
A NOTE ON HAUSA LITERATURE AND R.C.ABRAHAM Graham Furniss
R.C.Abraham is known primarily as a linguist and lexicographer. On only one occasion, as far as I am aware, did he grace the title of one of his many books with the word ‘literature’, and that is in his 1959 volume Hausa Literature and the Hausa Sound System. At first sight, the book seems an odd combination of 125 pages of ‘literature’ and 61 pages of phonetic and phonological observations of some perspicacity. It soon becomes clear however, that neither the author nor, presumably, the publisher, can have considered the volume to be an entirely independent piece; rather it must have been viewed as containing two additional components in Abraham’s larger project comprising principally his Dictionary of the Hausa Language (1949, with Mai Kano), and his grammar The Language of the Hausa People (1959b). The ‘Preface’ relates only to the section on literature, omitting any reference to the second half of the book or the reasons for combining the two sections in this way. The intention behind the inclusion of a section entitled ‘literature’ is set out in the Preface and the aims are limited and pedagogic. Abraham eschews any discussion of the pieces of ‘literature’ included—they are there to ‘teach vocabulary’ and to ‘bed down’ grammatical principles in the mind of the foreigner learning Hausa. Other authors of the period presenting specimens of Hausa literature made the uncontentious claim that such material would allow an insight into the mores, values, social life, etc., of the Hausa people. Abraham makes no such claim in his Preface. On the contrary, his explicit claim is that by presenting ‘literature’ in the way he does, he is combining the ‘theory and practice’ of language through the cross-referencing of features in his translation with discussion in his grammar book. As he says of that time, ‘no other such book exists’. His deliberate purpose is then to provide ‘a very literal English translation’ of the Hausa on facing pages. Not only is there no literary or social comment, awkwardness in the English will on occasion be necessary in order to provide the learner with a clear understanding of how the Hausa is constructed. The texts are graded so that the easier texts come first, and Abraham is sure that the learner will gain in confidence as the experience of reading the texts confirms him in the accuracy of what he has already learnt, through working through earlier grammatical discussion. Finally, the odd word unfamiliar to the learner can be easily dealt with—‘all that remains is elucidation of words still unknown to him: here, my Dictionary of the Hausa Language will remove the last puzzles.’ Abraham was nothing if not confident! The few observations I have to make on R.C.A.’s Hausa Literature are therefore limited by the circumspect nature of his own presentation, and by my desire to respect the purposes he outlined for himself in undertaking the task of translating the pieces contained in the section.
African Languages and Cultures, Supplement 1 (1992):43–50
A NOTE ON HAUSA LITERATURE AND R.C.ABRAHAM
35
The first comment relates to the nature of the pieces selected for presentation. ‘Imaginative literature’ is represented by two sections: an extract first from the beginning of East and Tafida’s Jiki Magayi (1934), a 5page piece which is then followed by 18 pages from an early edition of Magana Jari Ce, credited to the Literature Bureau rather than Abubakar Imam (1938/9). Both pieces are, in view of the fact that they come toward the end of the section, viewed by Abraham as being more difficult to understand than the cultural/ historical pieces that precede them. The only exception is the placing, at the end, of a text relating to marriage and divorce, written in the Sokoto dialect and therefore presumably considered per se more difficult for the learner of Hausa. The earlier pieces are either selections from early Literature Bureau publications Labarin Kasa da Tarihi (Batten 1934), Ka Kara Karatu (anon. 1938), or selections from Edgar’s (1911/13) Litafi na Tatsuniyoyi na Hausa (Labarin Hausawa; Al’adar Hausawa), with the addition of a piece on the Shari’a law and another section giving examples of idiomatic sentences—essentially a brief supplement to material to be found in his Dictionary. The pieces from the Literature Bureau start with a geography text intended for use in Hausa schools, and enjoining the teacher to take his students out into the community to find out about the social and physical world about them. Set out for the Hausa reader of the 1930s is information on topography, climate, the seasons, crops, farming, attire, domestic animals, and the typical trades of Hausaland. Trade and the commercial activities of Europeans also feature here. The section ends with a brief discussion of the way in which pastoral Fulani, having no fixed abode, differ from the inhabitants of ‘Hausaland’, ‘Yorubaland’ and ‘Munshiland’. Whatever the educative intentions were of the Literature Bureau in Zaria, the impression given to a foreign reader is precisely that there is a discrete physical environment and social structure that one can ascribe to ‘the Hausa people’. In this regard the piece echoes many of the summary descriptions of ‘the Hausa’ that were current at the time and have been since. While it is beyond my brief to try and follow the construction by Europeans of a notion of Hausa ethnic identity, it is clear that such pieces as this enumerated to Hausa speakers, in the original, and to foreigners, in the translation, a series of characteristic features that were considered typical of ‘the Hausa’. The tacit assumption that language use was to a very large extent coterminous with ‘tribe’ obscured the possibility of looking at Hausa as a language that could have been recently acquired, that was spoken by very different groups historically, or indeed that Hausa had been the language of domination struggling to assimilate other languages. The only caveat generally presented in popular works related to the question of the way in which the Fulani had retained the use of Fulfulde in the countryside, whereas Hausa had become the language of the cities within the Hausa/Fulani empire. I have no case to put by offering an alternative view of the relation between language and history in northern Nigeria; rather my point is simply to draw attention to the assumptions behind the view, generally presented, that there was a largely homogeneous community of ‘Hausas’ living over large parts of what was then Northern Nigeria. The first extract makes the explicit statement that: ‘The Hausas differ in customs slightly, but are all united in tongue and state of life so much so that one might call them one stock’ (p. 20). It is unclear whether there is any significance in the fact that the English translation omits a brief passage ’ defines Hausaland as follows: ‘Inda Hausawa suke sosai, in which the author of ‘Labarin Kano, Sakkwato da rabin Zariya na arewa: tana da yawa’ (p. 17); but certainly the great variety of components that went into assessments of what constituted the Hausa bakwai, in addition to the banza bakwai, reinforces the impression that the composition of what constituted ‘Hausa’ had always been something of a moveable feast. R.C.Abraham’s ‘Literature’ is interesting as an assemblage of characteristics that in some way outlines a definition of ‘Hausa’, even though he clearly does not intend to present it as such. While the first extract presents physical characteristics of the people and their environment, the second, much briefer, section, ‘Ka Kara Karatu’, seems to wish to present cultural characteristics that both provide an insight into ‘Hausa
36
GRAHAM FURNISS
thinking’ and, in a very diffuse way, an indication of elements of morality. The section consists of 17 stock anecdotes that generally illustrate one or another proverb or saying. In one or two examples, the anecdotes are jokes based upon puns, the best known being the one about the Gwari man who tells the visiting malam not to eat his dog. As with many of the tatsuniyoyi in Edgar’s collection, the anecdote is a very much truncated and collapsed version of an oral narrative. The link to a particular saying is sometimes so close that the saying carries that anecdote as incidental baggage in discourse. There is a tendency in presenting proverbs to see them as representing a set of moral values that, again, typifies the society. Section 3, ‘Idiomatic Sentences’, is an odd addition to the other continuous texts; it is presented in the manner of a phrase-book (indeed he acknowledges that some examples are adapted from Bargery and Parsons’ (1924) Hausa Phrase Book), and gives the impression that it is a repository for some of the more discursive notes that were too long for the Dictionary. Sections 4 and 5 are taken from Edgar and present the clearest examples of the kind of writing that attempted to summarise ‘Hausa-ness’ by giving examples of greetings, customs relating to marriage, childbirth, betrothal, etc. One of the interesting characteristics of the examples given here is the fact that the picture presented of ‘Hausa’ is not all positive. Regardless of the significance of the negative images for either the Hausa author or the Hausa audience, it is clear that the process whereby Europeans constructed ubiquitous stereotypes of ‘the Hausa’ (or anyone else for that matter) involved precisely the same kind of mixture of positive and negative traits. An example of Abraham’s translation of one such remark is as follows: Similarly, if two Hausas dwell together and the head of their house or teacher is the same person: if that house-head of theirs consults one of them more than the other, then that one who is not consulted backbites his kinsman in secret, that he may be dismissed and himself put in his place. A Hausa takes his kinsman to where he’ll perish. God protect us from the temperament of a bad Hausa! Amen! (pp. 56–58, cross-referencing omitted) Section 5, ‘Notes on Muslim Law’, mostly concerning marriage-related issues in combination with the additional, Sokoto dialect final section on ‘Marriage and Divorce’, present a further typification of Hausa ways of doing things. The presentation reinforces for the foreign reader first the notion that there are intricate and complex aspects to Hausa social life, and second that these customs have a widespread, normative function that further adds to the definition of ‘Hausa’. The final two sections, being extracts from Jiki Magayi and Magana Jari Ce, are rather different in character. The first is a brief piece in which rivalry between the young lover, Abubakar, and the rich suitor, Malam Shaihu, for the hand of the lovely Zainab, is the theme. The language is straightforward and the extract serves to reinforce the immediately preceding discussion of Hausa marriage, thereby providing a degree of verification, through ‘imaginative literature’, of the generalisations that had previously been made on the basis of legal norms. The second longer extract of some 18 pages is very different. It provides for the first time in the book a sense of the existence of an oral literature—story-telling as an amusement and as a skill. Here there is the possibility that the reader may move from objective description to subjective appreciation and a rather different notion of ‘Hausa-ness’; but it is here that Abraham’s insistence on highly literal translation becomes particularly inappropriate. Abubakar Imam’s great skill was to be able to set down on paper wit and humour, bombast and many other styles of speech. The learner of Hausa, for whom Abraham is writing, might himself gain some inkling of such things through struggling with the Hausa; certainly the translation does not do the original justice. This leads me on to my second area of comment— Abraham’s translations.
A NOTE ON HAUSA LITERATURE AND R.C.ABRAHAM
37
Abraham says in his Preface that one of his prime aims is to teach vocabulary, and it is clear that he has in mind an individual student, probably a young ADO in Nigeria, working on his own with perhaps an ‘informant’, making simultaneous use of Abraham’s other texts. The intensely descriptive early pieces in the book provide a large amount of often technical vocabulary, and it is here that we can see Abraham’s aims being met. His translations read moderately well, while sticking closely to the construction of the original Hausa. Here is a typical example: , a yi masa jinka: kuwa Dakin hayi (watau kago) shi ne wanda ake ginawa kamar shan . Soro ya fi yawa wajem masu azziki cikin ba su da gini sai kara suke jerawa su shafe da Zazzau, ana yin tafarfara (watau shirayi) maimakon soro, ita ce garurukan arewa: a jinka mai kusurwa . Gidam mai arziki yana da zauruka na saukar , da barori, da dawaki da a jiye hatsi: daga nan, a kam bi ta wani zaure, a shiga tsakiyar gida filim barga inda ake mata suke, da wurin dafa abinci, da rumbuna, da rijiya, da wurin inda turaka da wanka, da shadda. Dukan gidan kewaye yake da katanga mai tsawo. (p. 29, tone and vowel length omitted) The thatched house, i.e. the kago, is what is built of circular shape and thatched: some, indeed, have no mudwork and all they do is to make a row of cornstalks and smear them with clay. The rectangular house is commoner in the case of the rich in the northern towns. In the Zaria Emirate, they make the taf arf ara, i.e. the shirayi instead of the soro, and it is a house of thatch with four corners. The compound of a wealthy man has lodges for the lodging of guests, servants’ houses, and an open space for a stable where horses are tied up and corn stored: from there one passes through another porch and enters the centre of the compound where the men’s quarters are and the women’s huts and the place for cooking food and some bins and a well and the bathing-place and the latrine. The entire compound is surrounded by a high wall: (cross-referencing omitted) This clearly provides the learner with a lot of vocabulary relating to households, and the translation matches could perhaps have done with a the original closely enough for most purposes (although shan footnote). The translation reads reasonably well. In the case also of the final ‘technical’ extract in Sokoto dialect, Abraham’s translation manages to read reasonably, while also retaining a closeness to the Hausa such that the learner is able, I would judge, to divine how the Hausa means what it does. For example: aure, babu Ba a aure sai da shugaba da sadaki da shaidu biyu, adillai. Im ba su shaida ba ga ga sadaki, zangu goma sha biyu da hamsin. Uba ya samu shi amre, sai sun shaida: abin da ak tai budurwa, ko ba ta sani ba: in ta balaga, sai ta yarda. Wanda ba uba ga budurwa, ko amrad da tai zawara aure, wasiyyi ko wani nai, ba su yi mata aure ba sai ta balaga, ta yarda. Uba ba shi wa ko wanin uba sai ta ce ‘Na yarda!’ (pp. 115–117, tone, vowel length and footnotes omitted) Marriage is not valid unless there is an intermediary and money gift to the bride and two lawful witnesses: if they have not testified at the wedding ceremony there’s no marriage till they’ve so testified. What is the minimum gift is 1,250 cowries. The father has the power to give his virgin daughter in marriage even though she does not know: but if she’s pubert, he cannot give her in marriage unless she agrees. A father does not give in marriage a daughter previously married, nor does one other than the father do so unless she says, ‘I agree’. (cross-referencing omitted) Where Abraham’s approach to translation comes a little unstuck is in his rendering of Abubakar Imam’s prose in the extracts from Magana Jari Ce. The extract from Jiki Magayi that precedes MJC demonstrates
38
GRAHAM FURNISS
the more direct narrative style of East and Tafida, producing in Abraham’s English version a certain quaintness, and an impression that such quaintness simply reflects the Hausa style of speech. For brevity, I include an extract of the English from Jiki Magayi without the Hausa version (the Hausa is on p. 75): She smiled and said, ‘Do be sensible! If it were possible that one could tread on his money from here to the place where the sun rises, nothing’d cause me to marry him.’ He said, ‘Come, come! If you see a dog smelling a shoe, he’ll lift it away: but whatever you do, so be it!’ They took leave of each other. Malam Shaihu said, ‘Let me show this boy that whoever crosses me will come off worse.’ Then he had called to him a certain Malam Sambo whom all the townspeople feared, even the chief. Shaihu dangled before him the promise of much money and told him of the matter between him and Abubakar. (cross-referencing omitted) Quaintness may simply be the price to be paid for Abraham’s sticking to his stated aim of producing ‘a very literal English translation’. With Magana Jari Ce however, it is not simply a matter of the occasional oddity. Not only does he lose the wit and humour and the bombast of the original, the English comes out occasionally mangled to such an extent that the learner can, I would imagine, only wonder at what might have been the implications of the style of the original. In the following extract I have underlined those sections of the English which seem to me to be particularly problematic: , ya fizge daga hannum maga takarda, Da sarki ya ji abin da takardan nan ke ciki, ya tashi da ya kyakketa: ya kama gemun wazirin Sinari, ya jefar da shi gefe guda. Sarakunansa suka shiga tsakani, suna ‘Hucewa mai duniya! Rashin hankali ne na yara’. Sarki ya amsa cewa ‘Ko Musa ya lalace? Ya auri ‘yar Sarkin Sinari? Me aka yi aka yi Sarkin Sinari balle ‘yatasa’. !’ Ya dubi mutanansa, ya ce ‘Ku yi ta dukansu sai sum bar Wazirin ya ruga, ya haye dokinsa, mutanansa suka dafi bayansa: ’yam birni suka bi su woho woho. , fadawa suka yi ta duban juna. Liman ya ce ‘Alla ya ba ka nasara! Abin nan da aka Fada du ta yi, a aika lafiya dai?’ (p. 87, tone and vowel length omitted) When the King got to know what this letter was about, he got up in a rage: he snatched the letter out of the scribe’s hand (and) tore it to pieces: he seized the beard of the Wazir of Sinari and hurled him to one side. His feudal lords interfered, they were (saying), ‘Cooling down, Lord of the World!: witlessness is for boys.’ The King replied, ‘Has Moses deteriorated? Should he marry the daughter of the Sinari chief? Of what account is he far less his daughter?’ He looked at his men (and) said, ‘Proceed to beat them till they’ve left my land!’ The Wazir went off like a flash: he mounted his horse, his men followed him (lit. pressed on his back). The townsmen followed them with booing. The whole palace was in an uproar (and) the courtiers began to look at each other. The priest said, ‘God give you victory! this thing which has been done, will one have done it well?’ (Interestingly, there is no cross-referencing in this passage) Furthermore, Abraham comes up against particular problems with Imam’s writing in seeking to translate literally. His rendering of Me aka yi aka yi Sarkin Sinari balle ’ya tasa? in the above passage as ‘Of what account is he far less his daughter?’ is in fact a very fair translation but could hardly be called literal. For his literal-minded learner he does not even attempt a footnote on it.
A NOTE ON HAUSA LITERATURE AND R.C.ABRAHAM
39
Abraham manages well enough when faced with narrative, though his problems become apparent when he is handling dialogue, particularly when there are rhetorical overtones behind the interchange. For his ‘learner’, however, a feel for different registers of speech would have been so important, had Abraham felt himself able to try to render them through the English. A final example of such a problematic passage is: fiffike, ya ce ‘Alla ya ba ka nasara! Kada ka yi fushi bisa Da dai tsuntsun nan ya ga haka, sai ya , a ganina, ya karya mini daraja ne. Kamata ai, bai ga ubangijina! Ni ma, a jaka gudan nan da ya ba.’ kamata a sallama ni jaka guda tsuntsu, ya ce ‘Kai kuwa, tsuntsun nan!: menene Sai sarki ya ce wa bayi su tsaya. Ya dubi dalilinka na wannan cika baki haka?’ da na ke yi: ba Sai aku ya sunkwiyad da kai, ya ce ‘Akwai kuwa, Alla ya ja zamaninka, ni ke yi ba, ba ko don sabo da bakin nan da Alla ya ba ni ba, don sabo da kyan jikina amma sabo da baiwa wadda Ubangijimmu ya yi mini ta wajan iya duba.’ (p. 89, tone and vowel length omitted) When the bird saw thus, it flapped its wing (and) said, ‘God give you victory! Do not be angry with my master! I, indeed, at that one bag (i.e. a hundred pounds) which he said, in my opinion, he has derogated from my status. One like me, indeed, it is not fitting that I be surrendered for one bag.’ Then the King said to the slaves that they wait (a bit): he looked at the little bird, saying, ‘You Birdie!: what is the reason for this presumptuousness thus?’ Then the parrot bent its head saying, ‘There is as a matter of fact, God lengthen your days! a vaunting that I do: it’s not only because of my beauty of body that I do it nor even (ko) because of this eloquence which God has given me, but owing to the gift which the Lord gave me in reference to the being able to do divination.’ (cross-referencing omitted) As a pedagogic tool therefore, Abraham’s Hausa Literature has its limitations, yet it is important to remember that in the context of its time it was one of the few available examples of side-by-side translation available to the learner of Hausa, as Abraham says in his Preface. Combining modest aims with a marked lack of false modesty, he sets out another component in an ambitious scheme to encompass all the explanatory and pedagogic needs. of his ever-present learner of Hausa. REFERENCES Abraham, R.C. 1949 (and Mai Kano). Dictionary of the Hausa Language. 1st ed. London: Crown Agents for the Colonies. ——. 1959a. Hausa Literature and the Hausa Sound System. London: University of London Press. ——. 1959b. The Language of the Hausa People. London: University of London Press. Anon. 1938. Ka Kara Karatu. Zaria: SIM Bookshop. Bargery, G.P., and A.C.Parsons. 1924. A Hausa Phrase Book. Oxford: Oxford University Press. Batten, T.R. 1934. Labarin Kasa da Tarihi. Lagos: CMS Bookshop. East, R.M., and John Tafida. 1934. Jiki Magayi. Zaria: Literature Bureau. Edgar, F. 1911/13. Litafi na Tatsuniyoyi na Hausa. Belfast: Erskine Mayne. Imam, Abubakar. 1938/9. Magana Jari Ce. Zaria: Literature Bureau.
R.C.ABRAHAM’S EARLY INSIGHTS INTO HAUSA PRE-DATIVAL VERB FORMS* TIVAL VERB Philip J.Jaggar
1. Introduction R.C.Abraham’s The Principles of Hausa, published by the Government Printer (Kaduna) in 1934, represented a considerable improvement on all previous Hausa grammars and was the first comprehensive grammar to mark phonemic tone and vowel length (one of the major advances was the reduction of Bargery’s (1934) six-tone system to the correct three-tone system). As a model of erudition and thoroughness, it was designed as a reference work for language students, and although too densely-packed and at times abstruse to be recommended as a practical pedagogical aid for language-learners,1 it is nevertheless a goldmine of fascinating information, especially relating to verbal behaviour, and so instantly repays close scrutiny by the serious researcher.2 In this paper, rather than embarking upon a wide-ranging review of Principles, I have decided that it would be more fruitful to selectively evaluate what Abraham had to say about a still poorly-understood area of Hausa grammar—the morphosemantics of a category of verbs as they occur before indirect objects (the so-called ‘D-forms’). Inspection of his Principles reveals that Abraham had some sharp insights into this problem, some of which anticipated later analyses. He also provided us with the key to understanding a previously unexplained fact—why one particular subset of verbs (‘Grade 4’ final -ee) is frequently used before indirect objects (§4.2).
African Languages and Cultures, Supplement 1 (1992):51–66 *
I would like to thank Malami Buba, Graham Furniss, Yakubu Mukhtar, and Paul Newman for valuable comments on earlier drafts of this paper. I am also grateful to colleagues at the Seminar für Afrikanische Sprachen und Kulturen, Universität Hamburg, Germany, for kindly supplying me with a photocopy of Abraham’s Principles. 1 See Appleyard, Saeed and Williamson (this volume) for similar criticism of Abraham’s Amharic, Somali and Igbo grammars respectively. 2 The greater part of the book (approximately 170 pages out of a total 230) is devoted to a descriptive profile of Hausa verbal morphology and semantics. Abraham had at the time been working on his classification of verbs for inclusion in Bargery’s (1934) A Hausa-English Dictionary, and as indicated in the ‘Preface’, Principles was in fact meant to be used in conjunction with Bargery’s Dictionary (see also Roxana Ma Newman in this volume). Principles was subsequently revised and published in more accessible form (1941, 1959).
R.C.ABRAHAMS EARLY INSIGHTS INTO HAUSA PRE-DATIVAL VERB FORMS
41
2. The pre-indirect object D-forms within the Hausa ‘Verb Grade’ system Parsons (1960, 1962, 1971/72) divides Hausa verbs into seven morphologically distinct ‘Grades’, each with its own final vowel (or -VC in the case of Grade 5) and tone pattern. According to his system, verbs operating ‘basic’ Grades 1–3 are semantically ‘neutral’ for the most part, and those occurring in the ‘derived’ Grades 4–7 add their specific semantics to the core meaning of the verb. Verbs have the following four syntacticallydetermined forms: a B-form, used before a direct object pronoun; a C-form, used before a direct object noun; a D-form, used before an indirect object; an A-form, used in all other contexts (see Table 1). Table 1: The Hausa Grade system (disyllabic and trisyllabic verbs)3 Grade 1 Grade 2
A-form -aa -aa
HL(H) LH(L)
Grade 3
-a
LH(L)
Grade 4 Grade 5
-ee
HL(H) HH(H)
Grade 6 Grade 7
-oo -u
HH(H) (L)LH
B-form -aa -ee
HL(H) (L)LH
C-form -a -i
HL(L) (L)LH
D-form -aa -aa -aa
-ee -shee -oo
HL(H) HH(H)
-e
HL(L) HH(H)
-ee
HH(H)
-oo
HH(H)
-oo
HL(H) HL(H) HH(H) HL(H) HH(H) HL(H) HH(H) HH(H) HH
Grade 1=basic and Applicative verbs; Grade 2=basic transitive and Partitive; Grade 3=basic intransitive; Grade 4=Separative and Totality (?) (see §4.2); Grade 5=Efferential-Causative; Grade 6=Ventive; Grade 7= Affected-Subject. With Grades 1, 4, 5, and 6, the pre-datival verb forms (=D-forms) are straightforward: the form occurring before the indirect object markers wà [+NP] and ma- [+pronoun] is identical with the isolation/citation (A-) form. Examples (using disyllabic verbs only to simplify the presentation): Grade (1)
Gr. 1
(2)
Gr. 4
3
Citation A-form
D-form
taa dafàa she.PERF cook.1 ‘she cooked (it)’ yaa saacèe he.PERF steal.4 ‘he stole (it)’
taa dafàa masà she.PERF cook.1 IO.PRO food ‘she cooked the food for him’ yaa saacèe wà Audù he.PERF steal.4 IOM Audu money ‘he stole the money from Audu’
àbinci
The following transcription system has been used for all the Hausa examples: L(ow) tone=à(a), Falling=â(a), H(igh) is unmarked; aa=long vowel; r= flap, =roll/trill. Abbreviations: 1, 2 etc.=Verb Grade number; APP= Applicative; Gr.=Grade; IMPERS=impersonal; intr=intransitive; IO=indirect object; IOM=indirect object marker; PERF=perfective; NP=noun phrase; PREP =preposition; PRO=pronoun; SUBJV=subjunctive; tr=transitive;=indicates syntactic and semantic equivalence. Underlying verbs on which derivative Grade forms are based are listed only once.
42
PHILIP J.JAGGAR
Grade (3)
Gr. 5
(4)
Gr. 6
Citation A-form
D-form
yaa he.PERF deliver.5 ‘he delivered (it)’ mun kaawoo we.PERF bring.6 ‘we brought (it)’
yaa manà (dà) he.PERF deliver.5 IO.PRO (PREP) message ‘he delivered the message to us’ mun kaawoo wà maalàmîn aikìi we.PERF bring.6 IOM teacher ‘we brought the work to the teacher’
work
With regard to Grades 2, 3, and 7 (and some ‘irregular’) verbs, however, the D-form is not the same as the citation form, but entails the (near-) obligatory use of a special verbal extension,4 either -aa with H-L with H-H tones ( optionally assimilates and geminates with the following consonant),5 tones, and/or e.g.: Grade 2: (5)
(6)
b.
(7)
a.
A-form:
b.
D-form: (H-H )
a.
A-form:
D-form: (H-H ) =(H-L -aa)
a.
A-form:
yaa he.PERF choose.2 wà yaa he.PERF choose.PRO IOM his wife ‘he chose a wrapper for his wife’ naa nèemaa I.PERF look for.2
naa I.PERF look for.PRO IO.PRO work naa neemàa I.PERF look for.l/APP IO.PRO work ‘I looked for some work for him’ taa she.PERF tell.2
‘he chose (it)’ zanèe wrapper ‘I looked for (it)’
masà
aikìi
masà
aikìi6
‘she told (it)’
4 A lexically restricted subset of Grade 2 verbs allows the final -i C-form before an indirect object, i.e. instead of a derivational suffix, e.g. yaa dàami/bùgi wà Audù yaaròo (he.PERF bother/beat IOM Audu boy) ‘he bothered/beat Audu’s boy’ (see Munkaila (1990:152–156), and Jaggar and Munkaila (1992) for details). Tuller (1990a, b) argues convincingly that the general incompatibility of Grades 3 and 7 verbs with wà-marked indirect objects derives from their ‘unaccusative’ argument structure. Caron (1987) and Gouffé (1988) attempt to account for the restriction in terms of ‘diathèse’ or voice distinctions, proposing that Grades 2, 3, and 7 express a middle voice or ‘diathèse interne’; see, however, Swets (1989:75ff.) and Tuller (1990a:118ff., 1990b:21ff.) for criticism of this semantic characterisation, especially of the claim that Grade 2 conveys a middle value. 5 In Jaggar and Munkaila (1992), we suggest that the pre-datival extension might derive historically from a frozen pleonastic 3psf (IO) pronoun *ta, hence the morphological gloss PRO in the examples.
R.C.ABRAHAMS EARLY INSIGHTS INTO HAUSA PRE-DATIVAL VERB FORMS
b.
Grade 3: (8) a.
(9)
D-form: (H-L -aa)
taa she.PERF tell.1/APP IO.PRO news ‘she told the news to me’
A-form:
ciiwòo yaa illness it.PERF appear.3 ‘the illness appeared’ ciiwòo yaa illness it.PERF appear.PRO IO.PRO ‘the illness attacked her’ ‘I agree’ naa I.PERF agree.3 makà naa I.PERF agree.1/APP IO.PRO you.SUBJV come ‘I agree that you should come’
b.
D-form: (H-H )
a.
A-form:
b.
D-form: (H-L -aa)
Grade 7: (10) a. b.
A-form: D-form: (H-H )
mèe ya àuku? what it.PERF happen.7 mèe ya what it.PERF happen.PRO IOM Audu ‘what happened to Audu?’
The final
43
minì
matà
kà
zoo
‘what happened?’ wà
Audù?
3. and -aa D-forms of Grades 2 and 3 verbs
Over the past twenty years, several detailed investigations of the morphosemantics of Grades 2 and 3 and -aa D-forms have appeared (§3.1). Abraham (1934), however, was the first to record some of the more common final (and some final -aa) variants, and to note the differing interpretations (§3.2). 3.1. Recent analyses Parsons (1971/72:71–72), in what remains the most detailed treatment, observed that datival constructions frequently co-occur with what he called ‘verbs of the applicative class’, i.e. final -aa verbs,7 e.g.: (11)
ta shaafàa wà she.PERF rub. 1/APPIOM her face powder ‘she powdered her face’
foodàa (
6 I am treating the two (6b) variants and neemàa as being equivalent in meaning. Munkaila (1990:177ff.) and Jaggar and Munkaila (1992), however, report tense-aspect restrictions on use of the final D-forms, and with some verbs, use of one or the other extension correlates with a Benefactive as distinct from Malefactive interpretation (see §3. 1).
44
(12)
PHILIP J.JAGGAR
ya masà he.PERF tie.1/APP IO.PRO rope ‘he fastened a rope round it’
igiyàa (
‘tie’)
Parsons went on to assert that the different D-forms were basically ‘synonymous’, i.e. there were essentially no meaning-differences between final , -aa (and -ee) D-forms. However, he did observe that the final D-forms of some Grade 3 motion-verbs ‘tend to acquire an aggressive or pejorative meaning’ (p. 85, fn. 78), e.g.: (13)
kù fitam minì gàriinaa! (
and that with the final D-forms of some Grade 2 verbs, e.g. (
dà kà bugam masà rather you.SUBJV strike.PRO IO.PRO dog ‘rather than strike his dog’
kàree (
Newman (1973, 1977:281ff., 1982) was the next to take a close look at these pre-datival forms, and claimed D-form derived historically from a ‘Destinative’ extension *-in, used ‘to indicate that the that the final action of a verb was destined for, done for the benefit of, or otherwise affected or pertained to someone. It was probably most commonly used in sentences containing an indirect object’,8 e.g. (Newman 1977:292, examples taken from Parsons (1971/72:81): (15)
ya sookam minì he.PERF stab.PRO IO.PRO camel ‘he stabbed my camel’
(
Regarding the final -aa D-forms, Newman proposed that they in fact contained an ‘Applicative’ extension, a major function of which was ‘to effect the application of the action of a verb onto an indirect object’ (1973:339),9 e.g.: (16)
7
ya
sookàa
minì
‘he stuck a knife into me’
In earlier studies (1954:382; 1962:267ff.), Parsons used the labels ‘Projective-Applicative’ and ‘Transactional’ for these verbs, e.g. (Projective-Applicative) ta jeefàa wà kàrentà ‘she threw her dog a bone’, ta jeefà duutsèe cikin ruwaa ‘she threw a stone in the water’, (Transactional) nii nèe na kooyàa masà Hausa ‘I taught him Hausa’ (Parsons 1962). 8 Picking up on an earlier suggestion by Pilszczikowa (1969:20), Parsons (1971/72) had proposed that the pre-datival form was identical with the Grade 5 Efferential-Causative final verb (see Table 1). Newman (1973:339; 1977) was the first to
R.C.ABRAHAMS EARLY INSIGHTS INTO HAUSA PRE-DATIVAL VERB FORMS
45
he.PERF stab.1/APP IO.PRO knife
Newman’s analysis had the advantage of demonstrating that the meaning-differences between sentences such as (15) and (16) follow naturally from the obligatory use of two different derivational extensions predativally—his Destinative (15) vs Applicative -aa (16)—each with its own appointed semantics.10 Swets (1989:50ff.) pointed out that utilisation of the Applicative -aa extension often correlates with a Benefactive interpretation, as in (17a): (17)
a.
na amsàa mishì littaahìi (
Munkaila (1990:192ff.) subsequently reported that for individual speakers who have both extensions, with semantically ‘neutral’ verbs like ‘take’, ‘grab’, ‘steal’ etc., i.e. where the lexical semantics favour neither a Benefactive nor Malefactive reading on the IO argument, use of final -aa forces a Benefactive ‘for’ reading (18a, 19a), and final a Malefactive ‘from’ reading (18b, 19b):11 (18)
a.
yaa masà he.PERF take.1/APP IO.PRO stuff ‘he took the stuff for him’
b. a.
masà yaa he.PERF take.PRO IO.PRO stuff manà yaa he.PERF grab.l/APP IO.PRO money ‘he handed the money to us’
b.
yaa
kaayaa (
‘take’)
cf:
(19)
kaayaa
‘he took the stuff from him’ (
‘grab, take’)
cf: manà
call this identification into question, and Newman (1991) and Jaggar and Munkaila (1992) subsequently provided yet more evidence that the two extensions were indeed unrelated. Jaggar and Munkaila, however, were to demonstrate that Newman’s rival ‘Destinative’ hypothesis was also flawed, and proposed the alternative analysis outlined in fn. 5. 9 Grade 1 final -aa verbs thus consist of both basic verbs and derivative ‘Applicatives’. Newman (1973:339, fn. 49) attributes the cover-term ‘Applicative’ to Jungraithmayr (1969), who used it to describe a similar extension in the Ron group of languages; however, as noted above, Parsons (1954, 1962, 1971/72) had already in fact used the term for -aa verbs in Hausa. 10 In the same way that use of Grade 6 Ventive-centripetal verbs usually entails that the indirect object argument is (auto-) Benefactive, e.g. taa sayoo wà kântà zanèe ‘she bought a wrapper for herself (=Gr. 6 sayoo ‘buy (for s’one)’
46
PHILIP J.JAGGAR
he.PERF grab.PRO IO.PRO money ‘he grabbed the money from us’
3.2. Abraham on Grades 2 and 3 final
and -aa D-forms
Although Mischlich (1902:46–47) had in fact already captured the essence of the Applicative semantics of final -aa verbs some thirty years previously,12 Abraham (1934:78ff.) was nevertheless the first Hausaist to D-forms and to point to systematic meaningdifferences between document some of the more common the -aa and verbal extensions. Abraham observed that when the verb is ‘neutral’ with regard to the semantic D-form permits either a role of the IO argument, as is the case with the above ‘take’-verbs, the final Benefactive or Malefactive reading on the IO, e.g.:13 (20)
=
(21)
a.
yaa minì he.PERF take.PRO IO.PRO my money ‘he took my money for/from me’ minì b. yaa he.PERF take.PRO IO.PRO my money ‘he took my money for/from me’ minì dookìi yaa he.PERF steal.PRO IO.PRO horse ‘he stole a horse for/from me’
(
‘take’)
(
With regard to the Applicative -aa D-forms, Abraham (1934:80) made the following prescient observations: To indicate the carrying out of the action of a Mutable Verb [=Grade 2], plus the idea of “giving to a person”, “handing to a person for his use”, these verbs are often alternatively replaced by an Unchanging Verb in -A [=Applicative] from the same root’, e.g.: (22)
11
naa googàa masà I.PERF rub 1/APP IO.PRO oil
mâi (
Some speakers use only the final D-form on these verbs, e.g. , i.e. there is no morphological means of recovering the [± Benefactive] thematic role of the indirect object, in which case the decoder (hearer/reader) has to resort to pragmaticcontextual information to get the correct interpretation. 12 ‘Wird auf oder über einem Object eine Handlung vorgenommen, so wird der Schlussvocal des Verbs e in a verwandelt’, e.g. naa kaayaa ‘ich band eine Last’ vs naa fìtilàa à bisà kaayaa ‘ich band eine Laterne auf die Last’ (also noted by Furniss 1983:290). Cf. too Migeod (1914:104): ‘A [verbs]= motion from the speaker’, e.g. naa bugà gà duutsèe ‘I struck my foot against a stone’. And Taylor (1923:71) writes: ‘The A ending [is used] before prepositions and the indirect object, especially when ‘into’ is an inherent idea in the verb’, e.g. yaa masà ‘he spoke to him’, naa tuurà mùtûm à ruwa ‘I pushed the man into the water’. 13 Abraham’s marking of tone and vowel length in Principles was generally accurate, although he did mistakenly transcribe long vowels on Hi-Lo disyllabic Grade 1 -aa and Grade 4 -ee verb-final Lo tones as short in pre-pausal position, an error which stemmed from his belief that there was a (partial) correlation between word-final tone and vowel length. He did, however, correctly represent all verb-final vowels
R.C.ABRAHAMS EARLY INSIGHTS INTO HAUSA PRE-DATIVAL VERB FORMS
‘I rubbed oil on it’ matà naa I.PERF receive.1/APP IO.PRO her money ‘I received and handed her her money’ form in (20b)) (cf. Benefactive/Malefactive final makà (24) naa I.PERF choose.1/APP IO.PRO horse ‘I chose a horse and gave it to you’
47
(23)
dookìi (
‘choose’)
‘When it is desired to convey the idea that a liquid is spread over a surface, an Unchanging Verb in -A [=Applicative] followed by either the dative (or by the preposition a (locative)) is used’ (p. 86), e.g.: (25)
yaa shaafàa masà he.PERF wipe/plaster.1/APP IO.PRO white earth ‘he plastered (wiped) whitewash on it’ ‘Final -A is used to indicate “on” and “in”’ (p. 112), e.g.:
(26)
(27)
ruwaa yaa zuràaraa à rain it.PERF leak. 1/APP PREP hut ‘the rain has leaked into the hut’
naa googà mâi I.PERF rub. 1/APP polish PREP body of car ‘I rubbed polish on the car’
(
à
jìkin
mootàa
(Abraham 1959:19)
4. Use of the Grade 4 final -ee extension as a D-form before indirect objects Despite the fact that Newman’s (1973, 1977) ‘extension-switching’ framework permitted a natural and straightforward explanation of the differing semantics of (some of) the final -aa and D-forms, none of the recent studies summarised above have provided a convincing explanation of why the (Grade 4) final -ee extension is also regularly exploited in indirect object constructions. Close scrutiny of Abraham (1934), however, demonstrates that this gap in our analytical coverage results from the fact that we have been looking for the answer in the wrong place, and that Abraham was in fact showing us the way.
before indirect objects as long. In his 1941 and 1959 grammars, Abraham erroneously decided to mark them as short (see also Roxana Ma Newman in this volume).
48
PHILIP J.JAGGAR
4.1. Semantics of the Grade 4 final -ee verb The primary semantic function of the Grade 4 -ee extension has traditionally been described as adding the notion of ‘Totality’ or ‘Completeness’ to the core meaning of the underlying verb, i.e. a verbal action totally done or affecting a totality of objects (see Furniss 1983 and references therein),14 e.g.: (28)
yaa sayè kaayan he.PERF buy.4 stuff of food ‘he bought all (=up) the ingredients’
àbinci (
Both Swets (29, 1989:50ff.) and Munkaila (30, 1990:187–94) have also observed that when an IO argument follows a Grade 4 verb, it usually carries a Malefactive role, e.g. (17b) repeated as (29): (29)
(30)
naa amshèe mishì I.PERF take.4 IO.PRO book ‘I have taken (=stolen) his book’ manà yaa he.PERF grab.4 IO.PRO money ‘he grabbed the money from us’
littaahìi
(cf. 19a)
However, both Swets and Munkaila failed to provide a convincing answer to the obvious and important question of what triggered the switch to a Grade 4 -ee ‘Totality’ extension before an indirect object in the first place. It is easy to understand why an Applicative extension would be used before an indirect object (§3.1), but why would an extension whose primary meaning was supposed to express the notion of ‘Totality’ be manipulated to provide a Malefactive reading on a following indirect object? The reason for this analytical lacuna is simple—we have been searching for linkage in the wrong semantic domain. As long as we remained distracted by the conventional ‘Totality’ characterisation of the Grade 4 verb, there was no way of arriving at a plausible explanation of its Malefactive role. The correct explanation, however, lies in the fact that it is not the Grade 4 Totality semantics which supply and explain the Malefactive interpretation with indirect objects, but a different semantic sub-field of Grade 4 verbs— one which Abraham (1934) already had his finger on. 4.2. Abraham on the Separative ‘motion away from’ semantics of Grade 4 verbs The key to this puzzle is to be found in Abraham’s (1934) observation— repeated in his 1941 and 1959 reference grammars—that a number of Grade 4 verbs have what I have chosen to term a Separative meaning, entailing motion directed away from a specified reference-point.15 Hausa therefore switches to Grade 4 in pre-datival position not because of ‘their [Grade 4] on-going expansion throughout the Hausa verbal system’, as suggested by Newman (1973:339, fn. 50)—though this expansion has undoubtedly taken
14
There is also a subclass of ‘unaccusative’ Grade 4 verbs (=Furniss’ (1983) ‘reflexive-intransitives’), e.g. Gr. 4intr fashèe ‘break, be broken’ (
R.C.ABRAHAMS EARLY INSIGHTS INTO HAUSA PRE-DATIVAL VERB FORMS
49
place historically—but in order to utilise the intrinsic Separative semantics of many (kinetic) Grade 4 verbs. This directional component fits in nicely, moreover, with Newman’s (1983:415) claim that Hausa (and Chadic) verbal extensions combine a grammatical and directional meaning.16 When Grade 4 Separative verbs are used in datival constructions, the indirect object argument typically and understandably takes on a Deprivative-Malefactive role,17 and Abraham was clearly getting close to the correct answer as we can see from the following Separative-Deprivative paraphrases he suggests: The Grade 4 verb can often indicate simple ‘removal of the object itself’ (p. 83), i.e. without a following indirect object, e.g.: (31)
yaa askè geemùunaa (=Gr. 4 askèe ‘shave off’)18 he.PERF shave.4 my beard ‘he shaved off my beard’
15
The more I look at the Grade 4 verb, the more convinced I become that its core semantic property expresses a Separative, not Totality, value. A quick scan of the 70 or so Grade 4 verbs in Furniss (1983), for example, reveals that almost half are assigned English glosses which include the prepositions ‘away, apart, out, off, from, through etc.’. Some of these same verbs, moreover, e.g. zubèe Gr. 4intr ‘leak away’ (
50
PHILIP J.JAGGAR
(32)
naa I.PERF choose.4 them ‘I chose them and set them aside’
su (
‘choose’)
‘To denote removal, the Mutable Verbs [Grade 2] are often alternatively replaced by an Unchanging Verb in -E [=Grade 4] from the same root’ (p. 80), e.g.: (33)
yaa he.PERF take/steal.4 ‘he took/stole money from me’
/saacèe minì
(cf. 20a, 21) IO.PRO money
Final -ee verb plus indirect object indicates ‘depriving person by force or stratagem’=‘dative of the loser’ (pp. 82–83), e.g.: (34)
(35)
an masà rìigaa dà dookìi IMPERS.PERF take.4 IO.PRO gown and horse ‘a gown and a horse have been stolen from him (=taken from him by force)’ ruwaa yaa zuràaree à ràndaa water it.PERF leak.4 PREP water-pot ‘the water has leaked out of the water-pot’
(cf. 20b, 23)
(cf. 26)
The following contrastive verb-pairs from Abraham (36, pp. 84–85; 37, p. 168) nicely illustrate the semantic contrast between the Applicative -aa and Separative -ee forms, with or without an IO: (36)
a.
naa I.PERF haft. 1/APP my axe ‘I hafted my axe in its handle’
gàatariinaa à PREP handle
(
cf. Grade 4: gàatariinaa à b. naa I.PERF haft.4 my axe PREP handle ‘I unhafted my axe from its handle’ (37) a. taa girkà tukunyaa (=Gr. 1 girkàa ‘put on fire’) she.PERF put.1 pot
17
This is not to say that Grade 4 (centrifugal) verbs cannot co-occur with a Benefactive indirect object, e.g. yaa kashèe wà dà yawàa ‘he spent a lot of money on prostitutes’ (=Gr. 4tr kashèe ‘spend (money)’), taa masà bandeejì ‘she took the bandage off for him’ (=Gr. 4tr ‘take away/off, remove’
R.C.ABRAHAMS EARLY INSIGHTS INTO HAUSA PRE-DATIVAL VERB FORMS
51
‘she put the (cooking) pot on the fire’ cf. Grade 4: b. taa girkè furaa she.PERF put.4 millet gruel ‘she put away (to one side) the millet gruel’
‘Verbs denoting “impact”, when used in the -E form [=Grade 4], add the idea of “knocking over” to that of the impact’ (p. 85), e.g.: (38)
a.
cf. Grade 4: b.
naa bùgee I.PERF strike.2 him ‘I struck him’
shì (=Gr. 2 bùgaa ‘strike’)
naa bugèe I.PERF strike.4 him
shi
‘I struck him so that he fell over’
‘Final -E may indicate the idea of “recoiling from”, “eluding”’ (p. 111), e.g.: (39)
a.
cf. Grade 4: b. (40)
a.
cf. Grade 4: b.
yaa fìta (=Gr. 3 fìta ‘go out’) he.PERF go out.3 ‘he went out’ sìmintìi yaa cement yaa he.PERF die
ficèe it.PERF go out.4 mutù (=mutù vintr ‘die’)
‘the cement has lost its adhesiveness’
yaa he.PERF die.4
macèe manà IO.PRO
‘he died and left us’
‘he died’
(1959:18): the switch to Grade 4 -ee ‘often adds the notion of away’, e.g.: (41)
taa she.PERF wean.4 son Cf. too the following contrastive pair: (42) a. yaa googà he.PERF rub. 1/APP his shoes ‘he rubbed polish on his shoes’ cf. Grade 4: b. yaa googèe wà he.PERF rub.4 ‘he rubbed the dirt off his shoes’
‘she weaned her son’
tàakàlminsà dà
mâi PREP polish
tàakàlminsà dàtti IOM his shoes
dirt
52
PHILIP J.JAGGAR
Below is a (non-exhaustive) list of more Grade 4 verbs, culled at random from various dictionaries and grammars, which further illustrate the addition of a Separative (-Deprivative) meaning to the core semantics of the basic verb: (43)
Basic verb (where attested) v1tr ‘fasten, hook up’ bangàzaa v1tr ‘push against (door)’ dannàa v1tr ‘press down on gootàa v1tr ‘pass, exceed’ gusàa v1intr ‘move aside’ izàa v1tr ‘push’ v1tr ‘shake’ keetàa v1tr ‘cut, tear’ v1tr ‘wind on/around (turban)’ v1tr ‘shatter’ taaràa vltr ‘collect’ tuuràa v1tr ‘push’ àuraa v2tr ‘marry’ bànkaa v2tr ‘push’ v2tr ‘dip out, scoop up’ fìzgaa v2tr ‘snatch, grab’ gàbtaa v2tr ‘bite off kwàasaa v2tr ‘collect’ v2tr ‘take by force’ shàaraa v2tr ‘sweep’ v3intr ‘hide’ v3intr ‘drip’ tsiira v3intr ‘escape’ gudù vintr ‘run’ hau vtr ‘mount’ jaa vtr ‘pull/drag’
Final -ee Grade 4 [=Separative ‘away’] vtr & intr ‘unfasten, unhook, break apart/away’ bangà jee vtr ‘push away, break down (door)’ dannèe vtr ‘press/stamp on at a distance (snake)’ goocèe vintr ‘swerve aside, fall away’ gushèe vintr ‘pass by/away’ i jèe vtr ‘push to one side’ vtr ‘shake off’ keecèe vtr ‘tear off (and throw away)’ vtr ‘take off (turban), roll up and put aside’ vintr ‘fall apart’ taarèe vintr ‘collect (possessions) and move to new husband’s house’ tuurèe vtr ‘push aside, knock over’ aurèe vtr ‘marry (woman, and she has moved possessions to new husband’s house [=taarèe])’ bankèe vtr ‘push over’ vtr ‘remove, set aside’ fizgèe vtr ‘snatch away’ gabcèe vintr ‘move away/aside’ kwaashèe vtr ‘collect and remove’ vtr ‘take away by force’ shaarèe vtr ‘sweep away’ vtr & intr ‘hide away’ vintr ‘drip away’ tseerèe vintr ‘escape and run away’ gujèe vintr ‘run away from’ hayèe vtr ‘go beyond, cross over’ jânyee vtr ‘pull/drag away’ vintr ‘swerve, dodge aside, stray’ vtr ‘set aside’ kaucèe vintr ‘dodge, avoid, step aside’ saalèe vtr & intr ‘peel off, scrape away’ vintr ‘slip away’ waagèe vintr ‘fly open (door)’ waarèe vtr ‘separate, set aside’
R.C.ABRAHAMS EARLY INSIGHTS INTO HAUSA PRE-DATIVAL VERB FORMS
53
5. Summary In this paper we have seen how some of R.C.Abraham’s early work on verbal semantics and morphology anticipated important lines of research by later scholars. More specifically, his shrewd analytical insights have provided us with a basis for explaining some form-meaning correlations in an area of Hausa grammar we are only now really beginning to understand. REFERENCES Abraham, R.C. 1934. The Principles of Hausa. Kaduna: Government Printer. ——. 1941. A Modern Grammar of Spoken Hausa. London: Crown Agents for the Colonies (published on behalf of the Government of Nigeria). ——. 1959. The Language of the Hausa People. London: University of London Press. [A reprint of his (1941) grammar.] Bargery, G.P. 1934. A Hausa-English Dictionary. London: Oxford University Press. Caron, Bernard. 1987. Description d’un parler haoussa de l’Ader (République du Niger). Thèse de Doctorat d’Etat, Université de Paris VII. Furniss, Graham. 1983. The 4th Grade of the verb in Hausa. In Studies in Chadic and Afroasiatic Linguistics, ed. by Ekkehard Wolff and Hilke Meyer-Bahlburg, pp. 287–300. Hamburg: Helmut Buske. Gouffé, Claude. 1988. Fonction de la diathèse dans le verbe haoussa. In Studies in Hausa Language and Linguistics (In Honour of F.W.Parsons), ed. by Graham Furniss and Philip J.Jaggar, pp. 33–44. London: Kegan Paul International in association with the International African Institute. Jaggar, Philip J. 1988. Affected-subject verbs in Hausa: what are they and where do they come from? In Passive and Voice (Typological Studies in Language 16), ed. by Masayoshi Shibatani, pp. 298–330. Amsterdam & Philadelphia: Benjamins. Jaggar, Philip J., and Muhammed M.Munkaila. 1992. Evidence against the proposal that the Hausa pre-datival final verb=the “Grade 5” final /-s verb (and an alternative analysis). In Studia Chadica et Hamito-Semitica, ed. by Dymitr Ibriszimow. Frankfurt: University of Frankfurt. In press. Jungraithmayr, Herrmann. 1969. Der Applikative- oder Zielstamm im Ron. In Wort und Religion, ed. by H.J.Greschat and H.Jungraithmayr, pp. 89–93. Stuttgart: Evangelischer Missionsverlag. Migeod, Frederick William Hugh. 1914. A Grammar of the Hausa Language. London: Kegan Paul, Trench, Trübner & Co. Ltd. Mischlich, A. 1902. Lehrbuch der hausanischen Sprache (Hausa-Sprache). Archiv für das Studium deutscher Kolonialsprachen, Band I. Berlin: Georg Reimer. Munkaila, Muhammed M. 1990. Indirect object constructions in Hausa. PhD thesis, SOAS, University of London. Newman, Paul, 1973. Grades, vowel-tone classes and extensions in the Hausa verbal system. Studies in African Linguistics 4(3):297–346. ——. 1977. Chadic extensions and pre-dative verb forms in Hausa. Studies in African Linguistics 8(3): 275–297. ——. 1982. Grammatical restructuring in Hausa: indirect objects. Journal of African Languages and Linguistics 4(1): 59–73. ——. 1983. The efferential (alias ‘causative’) in Hausa. In Studies in Chadic and Afroasiatic Linguistics, ed. by Ekkehard Wolff and Hilke Meyer-Bahlburg, pp. 397–418. Hamburg: Helmut Buske. ——. 1991. Facts count: An empiricist looks at indirect objects in Hausa. Berkeley Linguistics Society 17. In press. Parsons, F.W. 1954. The ‘Mutable Verb’ in Hausa. In Proceedings of the 23rd International Congress of Orientalists, ed. by Denis Sinor, pp. 381–382. London: The Royal Asiatic Society. ——. 1960. The verbal system in Hausa. Afrika und Übersee 44:1–36.
54
PHILIP J.JAGGAR
——. 1962. Further observations on the ‘causative’ grade of the verb in Hausa. Journal of African Languages 1(3): 253–272. ——. 1971/72. Suppletion and neutralization in the verbal system of Hausa. Afrika und Übersee 55:49–97, 188–208. Pilszczikowa, Nina. 1969. The Changing Form (Grade 2) of the Verb in Hausa. Warsaw: Państwowe Wydawnictwo Naukowe. Swets, Francine. 1989. Grade 2 verbs with indirect objects in the Dogondoutchi dialect of Hausa. MA thesis, University of Leiden. Taylor, F.W. 1923. A Practical Hausa Grammar. Oxford: Clarendon Press. Tuller, Laurice. 1990a. L’hypothèse inaccusative et les verbes de “degré 3” et de “degré 7” en haoussa. Linguistique Africaine 5:95–126. ——. 1990b. Restricted argument structure in Hausa. Ms., University of Leiden.
THE PREVIOUS REFERENCE MARKER IN HAUSA: R.C.ABRAHAM’S INSIGHTS AND NEW ANALYSES* Paul Newman
1. Introduction at the end of nominals. As is generally the case Hausa has a definite determiner marked by a suffix with deictic elements, the exact semantic/pragmatic characterisation of Hausa’s determiner is hard to pin down (see Jaggar 1983:389ff.; 1985:144ff.); but it roughly indicates that the item to which it is attached has been previously referred to or is implied from the discourse context. The best source of examples illustrating its range of use is still the dictionary entry by the always remarkable R.C. Abraham (1962:692). A detailed analytical study of the Hausa determiner system, including valuable dialectal information, is provided by Gouffé (1971). Among Hausaists, the definite determiner is now commonly termed the ‘Previous Reference Marker’ [PRM], the designation that I will use here. For convenience, I shall simply with feminine singular nouns ending in the gloss it in translations as ‘the’. The form of the marker is with all other words, e.g.:1 vowel /a/, and (1)
‘the window’
The inherent tone of the determiner is Lo. When added to a word that ends in a Lo tone, as in the examples in (1), the tone of the PRM doesn’t appear overtly on the surface. When added to a word ending in a Hi tone, on the other hand, the Lo is attached to the preceding tone to produce a Fall, e.g.: (2)
wàndôn ‘the trousers’<wàndoo;
‘the hen’
African Languages and Cultures, Supplement 1 (1992):67–77 *
Research for this paper was carried out as part of a Hausa Reference Grammar Project supported by grants from the U.S. Department of Education (P0– 17A10037) and from the National Endowment for the Humanities (RT-21236). I am grateful to Russell Schuh and Philip Jaggar for comments and suggestions on an earlier draft. 1 The symbol / / represents the Hausa rolled R; the flap R is indicated by the unmarked /r/. In the examples, low tone is indicated by a grave accent /à/, falling tone by a circumflex /â/, and high tone is unmarked. With long vowels, indicated by double letters, tone is marked on the first vowel only.
56
PAUL NEWMAN
The PRM presents two general problems that need to be addressed, one analytical/synchronic, the other etymological/historical. 2. The analytical/synchronic problem Hausa has a genitive linker [L], which is segmentally identical to the PRM, i.e. it is -n or by the same conditions described above, e.g.: (3)
as determined
mootàa ‘the window of the car’; geemù-n-sà ‘his beard’; cookulà-n kuukù ‘the cook’s spoons’
Unlike the PRM, which has inherent Lo tone, the bound form of the L is toneless:3 (4)
wàndo-n Muusaa ‘Musa’s trousers’;
kàakaataa ‘my grandmother’s hen’
Because of the differences in their meaning and function, and because of the tonal distinction which shows up overtly with final Hi tone words, the PRM and the L are seldom confused with one another, in spite of their segmental identity. There are, however, exceptions to this statement, the most notable being constructions with a demonstrative. 2.1. Hausa demonstratives constitute a four-term system containing two oppositions: (a) NAN vs. CAN and (b) Hi tone vs. non-Hi (=Falling or Lo). Consider the following examples: (5)
shaahòn nân ‘this (near) hawk’, àkun nàn ‘this (near) parrot’; shaahòn nan ‘this hawk (in question)’, àkûn nan ‘this parrot (in question)’; shaahòn can ‘that hawk (there, visible)’, àkun can ‘that parrot (there, visible)’; shaahòn can ‘that hawk (there, distant or in question)’, àkûn can ‘that parrot (there, distant or in question)’4
The form of the demonstratives used with feminine nouns is exactly the same as that used with masculine nouns; the only difference is that the noun appears with instead of -n,5 e.g.: (6)
nân ‘this kettle’,
nàn ‘this hen’,
The question that needs to be asked is: what is the demonstrative?
can ‘that hut’,
can ‘that (distant) shed’
marker that occurs between the noun and the
2
In closed syllables, such as occur when the PRM is added, long vowels are automatically shortened in accordance with surface constraints against overheavy syllables (Newman 1972).
3
The Linker, but not the PRM, has corresponding long forms na(a)/ta(a) with inherent Hi tone. These allomorphs, which are used in specific morphosyntactic contexts, need not concern us in this paper. 4 The locative adverbs reflect the same four-term system except that the Lo tone variants do not appear, e.g. yanàa nân/ nan/cân/can ‘it is here/here (existent)/there/yonder’. 5 Actually, the assimilates to the initial /n/ of nan so that the grammatically required distinction between the masculine -n and the feminine , which is preserved in current Hausa orthography, is lost on the surface, i.e. / nàn/ =[kàazannàn]. This is correctly understood to be due to a phonological process rather than representing a grammatical irregularity.
THE PREVIOUS REFERENCE MARKER IN HAUSA: R.C.ABRAHAM’S INSIGHTS
57
2.1.1. Kraft & Kirk-Greene (1973:51) identify it as the PRM. This is undoubtedly on the basis of the forms with a Falling tone followed by the Hi tone demonstrative, e.g.: (7)
àkûn nan ‘this (known) parrot’, cf. àkûn ‘the parrot’ vs. àkun maalàm ‘the teacher’s parrot’; (distant) hen’, cf. ‘the hen’
can ‘that
They offer no explanation, however, why the presumed PRM added to a Hi tone noun sometimes appears with Hi tone rather than the expected Falling tone, e.g.: (8)
àkun (not *àkûn) nàn ‘this parrot’;
(not *
) càn ‘that hen’
Jaggar (1985:144ff.), who seems to accept the PRM analysis, similarly ignores the problem of the missing Falling tones. 2.1.2. Cowan & Schuh (1976:101, 298–99), Parsons (1963:207), and Tuller (1986:37) assert that the demonstrative is attached to the noun by means of the Linker. Their analysis works for the toneless forms, but, being the converse of the PRM analysis, fails to account for the cases such as in (7) where the noun displays a falling tone. Gouffé (1971:176–177), who also identifies the before the demonstratives as the Linker, explicitly acknowledges that the tones are problematical, but suggests that the variations are due to word internal morphotonemic factors which do not nullify the basic grammatical identification. 2.1.3. Galadanci (1969:60–61) also recognises the tonal problem. He concludes, however, that the tonal before the demonstrative evidence is too significant to be ignored and, as a result, argues that the cannot be identified with the Linker, which is always toneless, nor with the PRM, which always has Lo tone. For him, it is a third, distinct grammatical element whose use is limited to demonstrative constructions. is indeed the Linker, and that it serves to 2.2. The correct answer, I would suggest, is that the connect the noun to the following adverbial locative to form a cliticised demonstrative. In other words, an ‘that hen’ derives from a straightforward phrase ‘hen-of-there’. But if this is example such as so, why do final Hi tone nouns, such as in (7), become Falling and thus have the appearance of a noun with a PRM? It turns out that Abraham (1941:80–83) already had the correct explanation figured out some fifty years ago.6 Rephrased in present-day terms, his analysis is the following. The nan/can forms meaning ‘this/ that’ have inherent HL tone just like the homophonous locatives ‘here/there’. When added to a word ending in Lo tone (to which the toneless Linker is attached), the HL surfaces on the demonstrative as a Fall, e.g.: (9)
H
L
geemu-n nan here’ H L
HL → HL
H L | | geemù-n-nân H |
→
L |
HL \/ ‘this beard’, cf. yanàa nân ‘it’s HL \/ ‘that window’, cf. tanàa cân
‘it’s there’
When added to a word ending in Hi tone, on the other hand, the Hi component of the HL demonstrative is tonally absorbed into the preceding tone. The result is that the demonstrative does not surface with a Falling
58
PAUL NEWMAN
tone; rather its underlying HL pattern is manifested on the final two syllables composed of the noun plus the demonstrative, e.g.: (10)
L
H
HL
LH || àku-n -càn L H | |
→
aku-n can L
H
HL →
L | ‘that parrot’ L | ‘this hen’
The same tonal processes also account for the shape of the ‘long-form’ demonstratives made up of wa+the Linker+nan/can, e.g.: (11)
H
HL →
wa-n-nan H
HL →
H | wannàn H | waccàn
L | ‘this (one)’ L | ‘that (one)’
By contrast, the homophonous adverbials ‘here/there’, which do not cliticise to the preceding word, retain the HL falling tone regardless of the preceding tone, e.g.: (12)
yaa zoo nân (not *yaa zoo nàn) ‘he came here’ zaa tà tàfi can (not *zaa tà tàfi càn) ‘she will go there’
For Abraham, the other pair of nan/can forms meaning ‘this (existent)/that there (distant)’ has underlying LH tone. Given the fact that Hausa does not have surface rising tones—the corresponding adverbs nan/can ‘here (existent/visible)/there (distant)’ have Hi tone—this was an extremely perceptive insight by Abraham at the time. When added to words ending in a Lo tone, tonal absorption takes place, the result being Lo-Hi on the final two syllables. The pattern thus mirrors that illustrated in (10) for the HL demonstrative, e.g.:7 (13)
H
L
LH
geemu-n nan H L
LH
→
→
H L | | geemù-n-nan H L | |
H | ‘this (known) beard’ H | ‘that there window’
If the tonal situation in Hausa were entirely symmetrical, an underlying LH demonstrative added to a word ending in a Hi tone would be expected to attach to the demonstrative to produce a rising tone.
6
The same material is found in the more easily accessible 1959 edition of the work (Abraham 1959:53–55).
THE PREVIOUS REFERENCE MARKER IN HAUSA: R.C.ABRAHAM’S INSIGHTS
59
However, since Hausa does not have rising tones, the initial Lo of the LH demonstrative is attached to the preceding syllable (producing a surface Falling tone), in order to preserve the underlying LH pattern, e.g.: (14)
L H aku-n can L H
LH →
àkû-n -can ‘that there parrot’
LH →
‘this (known) hen’
With the long form demonstratives, the initial Lo of the LH demonstrative either attaches to the preceding syllable or fully occupies it and overrides the initial Hi tone, e.g.: (15)
H LH wa-n -can H LH
= →
wâncan
wàncan ‘that (one) there’ =
→
wânnan
wànnan ‘this (one) known’
One should mention that the beauty in Abraham’s analysis is that it not only accounts in a natural way for the variation in the form of the noun preceding the demonstrative, but it also accounts for the Lo vs. Falling allomorphs of the demonstratives themselves. 2.3. The ability of natural, phonologically shallow tone rules to create morphological confusion can be illustrated with the use of the PRM in relative clause constructions. In Hausa, relative clauses follow the head noun and are introduced by a particle dà. The head noun usually, but not invariably, occurs with the PRM, which, when added to a final Hi tone noun produces a Fall,8 e.g.: (16)
àkû-n dà na sàyaa ‘the parrot that I bought’ dà ta kaamàa ‘the hen that she caught’
In certain Hausa dialects, for example that of the Maradi area (and probably considerably beyond), the tonal sequence Falling+Lo in relative clauses typically simplifies to Hi Lo (cf. Gouffé 1981:28; PN notes),9 e.g.: (17)
àku-n dà (<àkû-n dà) na sàyaa ‘the parrot that I bought’ kàaza-d dà (
The result is that the PRM in the relative construction looks just like the Linker. Not only is the linguist likely to be misled by this, but over time native speakers themselves can be expected to reinterpret the erstwhile PRM as a Linker and thereby reformulate their grammars.
7
An alternative analysis would be to say that the LH remains as is on the demonstrative and that it surfaces as Hi due to the operation of a late LH to Hi simplification rule. My intuition is that the analysis presented here is the right one; but both analyses equally generate the correct surface forms.
60
PAUL NEWMAN
3. The etymological/historical problem Viewed historically, where does the PRM come from? 3.1. Hypothesis 1. Abraham (1941:82) views the PRM as a reduced form of the Noun-Linker LHdemonstrative construction with the demonstrative omitted. In this analysis, the Lo tone on the PRM is accidental, so to speak, and derivative, e.g.: (18)
shaahò-n ‘the hawk’<*shaahò-n nan/can ‘this/that known hawk’ ‘the hen’<* nan/can ‘this/that known hen’ maatâ-n ‘the women’<*maatâ-n nan/can ‘these/those known women’
Abraham implies that the PRM could equally represent a reduction from L+ nan or from L+can, the semantic difference between the two forms being neutralized. Jaggar (1983:176, n. 13), who supports the idea of the PRM having come from a reduced demonstrative construction, asserts on universal linguistic grounds that the source construction most likely would have contained the distal demonstrative, i.e. the PRM can/ and / can/. Jaggar does not, however, explicitly equate the would have come from / marker with the Linker; rather he calls it a ‘gender/number-sensitive suffix’ (p. 145). and derive directly from Lo tone deictic markers, 3.2. Hypothesis 2. Schuh (1983) proposes that *nà and *tà respectively. (The change of t> in syllable final position in Hausa is regular.) The Lo tone associated with the PRM represents the preservation of an original tone after the loss of its host vowel. ), Etymologically, *nà and *tà are presumed to be cognate with the Hi tone Linker na/ta (=toneless but the relationship is historically much deeper than that suggested by Abraham’s derivation.10 In other words, for Schuh *NA/*TA represents a single pair of deictics, whose different functions, namely genitive linker vs. previous reference marker, are signalled by tonal distinctions. 4. The problem with Abraham’s derivation is that it treats the Noun+L+nan/can as historically basic and the Noun+PRM as derivative, whereas there is good reason to believe that this is the reverse of the actual case. The older demonstrative in Hausa was probably a form related to the -ga (with underlying LH tone!) that is used in Western dialects, e.g. zoobè-n-ga ‘this ring’, màatâg-ga ‘this wife’. The present-day Standard Hausa demonstratives that employ the locative adverbs nan/can are almost certainly more recent paraphrastic constructions which replaced the earlier demonstrative forms. Schuh’s analysis thus strikes me as being on the right track; but it too is wanting in some respects. I would like, therefore, to propose two additional hypotheses which I deem worthy of consideration and evaluation. and should come from forms originally containing an underlying 4.1. Hypothesis 3. The idea that vowel with Lo tone seems reasonable. There is nothing, however, which forces us to accept the idea that the original vowel had to have been /a/. Schuh specifies the PRM forms as *nà/*tà because he conceives of them as being etymologically related to the na/ta Linker; but in so doing he minimises the tonal differences. 8
The statement by McIntyre & Meyer-Bahlburg (1991: ix) that final Hi tone nouns plus always remain Hi before the relative marker dà is incorrect. 9 With the long relative forms wândà ‘he who’ wâddà ‘she who’ (<wa+PRM+dà), the tonal simplification to wandà/ waddà is also becoming the norm in Standard Kano Hausa. 10
Abraham sees the PRM as deriving from the Linker by surface ellipsis. Schuh, on the other hand, views the derivation as having gone in the opposite direction, i.e. he considers the Linker to be a secondary development by semantic bleaching and grammatical reinterpretation from an original determiner/deictic element.
THE PREVIOUS REFERENCE MARKER IN HAUSA: R.C.ABRAHAM’S INSIGHTS
61
I would suggest that a better analysis exists which provides a natural explanation for the distinct Lo tone on and derive not from the same source as the Linker, but rather from the PRM. This alternative is that ,11 e.g.: distinct definite determiners of the form *nì and (19)
bàkâ-n ‘the bow’<*bàkaa-nì;
‘the hen’<
The postulated determiners relate to widespread third person singular clitic pronouns in Chadic. The use of these forms as determiners can be illustrated by examples from Kanakuru (Newman 1974:86), where their occurrence is lexically restricted, and from Sura (Jungraithmayr 1963:24), where only the ni form is used since the language does not preserve grammatical gender: (20)
a. b.
máamí-nì ‘the man’ (=máamí-ì) [Kanakuru] támnó-rò (where rò<*tò) ‘the woman’ (=támnó-ì) [Kanakuru] gùrùm-ni ‘the man’, màt-ni ‘the woman’, lú-ni ‘the house’ [Sura]
One should note that in neither Kanakuru nor Sura is a form such as ni used as a Linker. from *nà/*tà or from are similar in that 4.2. Hypothesis 4. The historical derivations of and are monomorphemic and that the Lo tone is or was an integral part of the they presume that and morpheme. Another possibility, which as far as I am aware has never been suggested, is that represent the fusion of a toneless Linker plus a floating Lo tone /`/, where the floating tone is the true càn ‘that hen’ are composed of PRM determiner. In the same way that gida-n nàn ‘this home’ and ‘the hen’ would be composed of NounNoun-Linker+demonstrative, so gidâ-n ‘the home’ and Linker+PRM, where the PRM is /`/. An interesting consequence of this analysis is that the PRM, like the demonstratives, turns out to be unmarked for gender. Whereas Hausaists have come to view the PRM as one of the more important exponents of gender —and synchronically one undoubtedly would want to continue to think of it in those terms—it is a very common, albeit not ubiquitous, characteristic of Chadic languages for the determiner (=definite article=referential marker) to be invariant in form and not gender-sensitive. The proposal, then, is that the PRM was not attached directly to the noun, but rather was connected to it by means of a Linker. The following illustrates the presumed *Noun-L+PRM pattern: (21)
*ràagoo-n+`→ràagôn ‘the ram’;
+`→
‘the hen’
If the above analysis is correct, the question naturally follows, where did the floating tone come from? The most likely answer is that it is a vestige of the Lo tone suffixal determiner *-ì, which Schuh (1983:58) reconstructs as a Chadic definite article on the basis of its wide distribution in the family. Consider the following examples from Kanakuru (Newman 1974:86) and Guruntum (Jaggar 1988:177): (22)
11
a. b.
tóró-ì ‘the farm’, máawò-ì ‘the stranger’, gám-íì ‘the ram’ [Kanakuru] bùushi- (y) ì ‘the ashes’, yiŋsù- (w) ì ‘the road’, - (w) ì ‘the boy’ [Guruntum]
The schwa utilised in the feminine form is a rough approximation. Some West Chadic languages (especially in the Bole group) have /to/, some have /ti/, and some have /ta/. The use of the abstract form is intended here simply to keep this item notationally distinct from the linker /ta/ which occurs in the na/ta pair.
62
PAUL NEWMAN
Actually, we do not have to turn to other Chadic languages to illustrate the -ì form of the determiner. This also exists in Hausa itself, although not in the standard dialect to which we often restrict our attention. In is the regular form of the PRM with feminine nouns Maradi and other north-west dialects, -ì instead of (Zaria 1982; PN notes), e.g.: (23)
kàazâ-i ‘the hen’, abdùgâ-i ‘the cotton’, cf. àkû-n ‘the parrot’
Interestingly, the dialects with -ì as the feminine PRM do not use this form as a Linker, thereby supporting the view that the Linker and the PRM are distinct morphemes. The feminine Linker in these dialects is -C (=gemination of the following consonant), which is simply a phonological manifestation of the same -t which underlies the Standard Hausa , e.g.: (24)
kàaza-t-tà ‘her hen’, not *kàaza-i-tà; abdùga-r riimii ‘kapok’ (‘cotton of silk-cotton tree’), not *abdùga-i riimii
When I first became aware of this final -ì form, I casually assumed that it represented a phonological manifestation of the feminine PRM marker like the other dialectal variants -t/ /-l etc. I now consider it much more likely that it represents a direct reflex of the Chadic *-ì suffix. Nevertheless, even if this is correct, one could argue that the presence of this morpheme in Hausa constitutes evidence against, rather than for, the interpretation of the PRM as having come historically from a Linker +-ì. The reason is that in Maradi dialect, as in Kanakuru and Guruntum, -ì is attached directly to the preceding noun without an intervening Linker. While some Chadic languages, e.g. Tera, do connect the suffixal determiner to the noun by means of a Linker, this does not seem to be the West Chadic pattern. This doesn’t mean that the analysis of bàkân ‘the bow’, e.g. as having come from three morphemes bàkaa ‘bow’+-n ‘Linker’+` (<*-ì) ‘determiner’ is ruled out, but it certainly weakens the case. 5. I have outlined four historical hypotheses regarding the etymology of the PRM in Hausa. In my opinion, none of the hypotheses is so unreasonable as to be rejected outright. Nevertheless, if I had to select one that I considered the most promising, I personally would choose Hypothesis 3, namely that the present suffixes go back to distinct determiners of the form *nì and . The purpose of this day communication, however, is not to argue for one analysis over another, but rather to clarify issues regarding the synchronic and etymological relationship between the Linker and the PRM, and to present new historical interpretations that heretofore had not been entertained. REFERENCES Abraham, R.C. 1941. A Modern Grammar of Spoken Hausa. London: Crown Agents for the Colonies (published on behalf of the Government of Nigeria). ——. 1959. The Language of the Hausa People. London: University of London Press. ——. 1962. Dictionary of the Hausa Language. 2nd ed. London: University of London Press. Cowan, J Ronayne, and Russell G.Schuh. 1976. Spoken Hausa. Ithaca: Spoken Language Services, Inc. Galadanci, Muhammad Kabir Mahmud. 1969. The simple nominal phrase in Hausa. PhD thesis, SOAS, University of London. Gouffé, Claude. 1971. Remarques sur le syntagme démonstratif en haoussa. In Afrikanische Sprachen und Kulturen— Ein Querschnitt [Festschrift J.Lukas], ed. by Veronika Six et al., pp. 170–179. Hamburg: Deutsches Institut für AfrikaForschung.
THE PREVIOUS REFERENCE MARKER IN HAUSA: R.C.ABRAHAM’S INSIGHTS
63
——. 1981. Comment recueillir et éditer les proverbes haoussa: quelques suggestions pratiques. In Itinérances II, pp. 115–135. Paris: Société des Africanistes. Jaggar, Philip J. 1983. Some dimensions of topic-NP continuity in Hausa narrative. In Topic Continuity in Discourse: A Quantitative Cross-Language Study (Typological Studies in Language, 3), ed. by Talmy Givón, pp. 365–424. Amsterdam & Philadelphia: Benjamins. ——. 1985. Factors governing the morphological coding of referents in Hausa narrative discourse. PhD dissertation, UCLA. ——. 1988. Guruntum (gùrdùŋ) (West Chadic-B): linguistic notes and wordlist. African Languages and Cultures 1(2): 169–189. Jungraithmayr, Herrmann. 1963. Die Sprache der Sura (Maghavul) in Nordnigerien. Afrika und Übersee 47:8–89, 204–220. Kraft, Charles H., and A.H.M.Kirk-Greene. 1973. Hausa. (Teach Yourself Books.) London: Hodder and Stoughton. McIntyre, Joseph, and Hilke Meyer-Bahlburg, assisted by Ahmed Tijani Lawal. 1991. Hausa in the Media: A Lexical Guide. Hamburg: Buske. Newman, Paul. 1972. Syllable weight as a phonological variable. Studies in African Linguistics 3(2):301–323. ——. 1974. The Kanakuru Language. (West African Language Monographs, 9.) Leeds: Institute of Modern English Language Studies, University of Leeds and West African Linguistic Society. Parsons, F.W. 1963. The operation of gender in Hausa: stabilizer, dependent nominals and qualifiers. African Language Studies 4:166–207. Schuh, Russell G. 1983. The evolution of determiners in Chadic. In Studies in Chadic and Afroasiatic Linguistics, ed. by Ekkehard Wolff and Hilke MeyerBahlburg, pp. 157–210. Hamburg: Buske. Tuller, Laurice A. 1986. Bijective relations in Universal Grammar and the syntax of Hausa. PhD dissertation, UCLA. Zaria, Ahmadu Bello. 1982. Issues in Hausa dialectology. PhD dissertation, Indiana University.
LEXICOGRAPHIC METHOD IN R.C.ABRAHAM’S HAUSA DICTIONARY* Roxana Ma Newman
1. Introduction All lexicographers, even the most individualistic, are beholden to their predecessors. R.C.Abraham’s comprehensive Dictionary of the Hausa Language, published in 1949 and considered by many to be his greatest work, was the culmination of a distinguished lexicographic tradition dating back to 1876.1 The comprehensiveness of his dictionary is due in no small part to the enormous debt he owed his immediate predecessor, G.P.Bargery, whose Hausa-English Dictionary, published in 1934, remains one of the finest achievements in African linguistics. In view of the undeniable greatness of Bargery’s dictionary, one has to wonder what drove Abraham to ‘challenge’ that work by producing, a mere 15 years later,2 an equally impressive dictionary, whose individual merits can be judged quite independently of Bargery (see reviews by Parsons 1949, 1950; Lukas 1949). Obviously there must have been some personal and professional rivalry between the two men, as Abraham had been an assistant to Bargery for two years.3 A glance at the prefaces to Abraham’s various Hausa publications over the years documents how his initial high regard for Bargery developed into an increasingly ambivalent one.4 By the time Abraham had his dictionary reprinted in 1962 in a second
African Languages and Cultures, Supplement 1 (1992):79–97 *
Research for this paper was carried out under a U.S. Department of Education grant no. G0085–40637. This support is gratefully acknowledged. 1 For a historical overview of Hausa lexicographic tradition, see Newman (1974). Newman & Newman (1991) contains a brief description of lexicographic studies on Hausa and other Chadic languages. 2 We should remember that Abraham produced this dictionary almost entirely without support from the government, compiled in his spare time while on war assignment to countries as distant as Egypt and Russia. As one reviewer has written, such single-minded dedication during such turbulent times was truly ‘an acte gratuit of the first order!’ (Armstrong 1964:50). 3 Bargery was quite straightforward about Abraham’s role:
I have to acknowledge much valuable help given by Capt. R.C.Abraham, of the Nigerian Administrative Service, who for two years was seconded as my assistant. In particular he contributed materially to the classification of the verbs and the noting of Arabic origins. (1934:vii)
LEXICOGRAPHIC METHOD IN R.C.ABRAHAMS ‘HAUSA DICTIONARY’
65
edition,5 he no longer saw it necessary to mention Bargery in his references to the current literature on Hausa. A much more crucial reason for producing a new dictionary, one which has been overlooked in assessing Abraham’s contributions to Hausa studies, has to do with the explicitly pedagogical aims of all his books. Although Abraham made highly original and lasting contributions to the structural analysis of a number of different African languages, the ultimate goals of all his research were practical. All his grammars, dictionaries, and readers were books which he earnestly believed would suffice to provide Europeans (and later Africans) with all they would need to learn the language in question and acquire good speaking skills. As Hair points out: ‘He never found time to discuss in learned papers points of linguistic analysis’ (1965: 64). Rather, his books were characterised by titles containing expressions like ‘spoken Hausa’, ‘modern Yoruba’, ‘for the European student’ etc., to stress the importance of learning the current spoken language, of using these languages as the people themselves spoke them. He berated his predecessors for writing Hausa grammars that ‘strike the eye and not…the ear’, so that students using those books ‘start with an
4 In Abraham’s ‘Preface’ to his Principles of Hausa (1934), published the same year as Bargery’s dictionary, he stressed that the two books were designed to be used side by side. His considerably revised grammar of a few years later—A Modern Grammar of Spoken Hausa—again contains a gracious acknowledgement to Bargery:
To Dr. G.P.Bargery I owe much, for the clarification in my material during the two years we were working together on the final revision of his monumental Hausa Dictionary. (1941:vii) However, in the ‘Preface’ to his own dictionary (1949, first edition) Abraham wrote in a completely new vein about this collaboration: My two years’ collaboration in Dr. Bargery’s Dictionary of 1934 consisted largely in introducing a coherent grammatical scheme into his work and reducing the millions of loose slips into the form of numbered paragraphs, but in spite of being entrusted with the checking of the whole material, my jurisdiction over rejection of matter was severely limited [emphasis mine]. For what was sound in that work, however, we must be for ever grateful to him. (1949:v) Abraham’s condescending tone, as well as his sense of frustration over Bargery’s inclusion of material that he, Abraham, would not have allowed, are all too clear. The most curious quotation reflecting Abraham’s deep rivalry with Bargery appears in the ‘Preface’ to his grammar, The Language of the Hausa People: …my Dictionary of the Hausa Language 1949…should be in the hands of every user of the present grammar. This is the most recent dictionary, as Dr. Bargery’s Hausa-English Dictionary which bears a later date on its title-page [presumably the 1957 third printing—RMN] is a photographic reproduction of our joint [NB! -RMN] Hausa-English Dictionary of 1934. (1959b: vii) A similar reference to Bargery’s dictionary as ‘our Hausa dictionary’ appears again in Appendix I (p. 181) of the same book. 5 The
‘second edition’ (Abraham 1962) is curious in several ways. It no longer includes the name of Malam Mai Kano, who was listed on the title page of the 1949 edition as co-compiler. Aside from that, it is a ‘second edition’ only in the sense of having a new preface, the body of the book being an exact reproduction of the 1949 edition. Finally, the 1962 edition erroneously states on the copyright page that the first edition was published in 1946 instead of 1949.
66
ROXANA MA NEWMAN
initial handicap which they never overcome and speak in a manner offensive to the Hausa ear’ (1959b: v). Similarly, he continually exhorted his audience to master tone in whatever language,6 finding ingenious ways to make the tonal/intonational systems seem less daunting (as in his apt description of Hausa downdrift as ‘a system of punctuation orally’ (1958)). Even the texts contained in his readers (e.g. 1959a) could be read aloud as ‘oral’ texts because, unlike most other collections of African language texts, Abraham’s were written entirely in phonemic transcription, even when standard orthographies (without diacritics) could have been used. Still, he warned students not to rely solely on such transcriptions: Visual ability…differs enormously and where one reader can from a transcription pronounce a word as accurately as the born speaker of the language, another seems to have little instinct in using the printed page unless he also hears the word said: you should have every word in this book said to you by an intelligent Hausa speaker…[emphasis mine]. Such collaboration is essential, for the book requires the help of the speaker and the speaker requires the help of the book… It is the function of this work to explain the ‘why’ when the Hausa tells you ‘that is how we express it, I cannot give you the reason’. (1941:1) Abraham’s objectives, then, were clearly pedagogical, and he consistently used what he perceived to be direct language teaching methods to assist his learners in expressing themselves well in the language.7 The pedagogical thrust of all his works was to promote the expressive/productive/active goals of language learning, what foreign language teachers today would call ‘communicative proficiency goals’. In assessing Abraham’s Hausa dictionary with this in mind, we shall see how these pedagogical goals in fact guided the form and content of his work, setting it clearly apart from Bargery’s dictionary, as he undoubtedly intended it to be. 2. Dictionary typology The pedagogical orientation of Abraham’s Hausa dictionary showed that he was well ahead of his time in his approach to the lexicography of African languages. Most bilingual lexicographers concerned with the role of dictionaries in language learning have come belatedly to the realisation that there is a theoretical contradiction in expecting a single unidirectional bilingual dictionary to satisfy equally the dual functions of expression (or encoding) and comprehension (or decoding). An English speaker learning Hausa optimally needs two dictionaries, an English-Hausa one to help him find the expressions he needs to communicate effectively, and a Hausa-English one to help him understand what he hears or reads. When the further needs of the Hausa speaker learning English are taken into consideration, the optimal number is doubled. This ideal configuration is shown in (1) (adapted from Kromann et al. (1984), who use the terms active/passive instead): (1)
6
expression/encoding
native→foreign language
He actually credits Bargery with first impressing upon him the tonal nature of Hausa (Abraham 1958). The fact that he evidently failed to realise that his works were often so linguistically compact and sophisticated as to be ‘forbiddingly difficult to the layman’ (Armstrong 1964:52) in no way detracts from his stated pedagogical aims. His grammars were actually reference works for students of the language, not textbooks as such, which explains the lack of exercises and translations.
7
LEXICOGRAPHIC METHOD IN R.C.ABRAHAMS ‘HAUSA DICTIONARY’
for English speakers for Hausa speakers comprehension/decoding for English speakers for Hausa speakers
67
English-Hausa8 Hausa-English9 foreign→native language Hausa-English English-Hausa10
Given this typology, it becomes evident why a single dictionary cannot serve equally well the needs of both groups of speakers. It is also clear that the large, reference-oriented Hausa-English dictionaries of Bargery and Abraham, compiled by and for European users, are most suitable for comprehension purposes. This has been true of most African language dictionaries (whatever their size), since they are almost always in the direction of African to European language: they therefore have a built-in disadvantage for the European wanting to speak the African language. Short of writing a companion European-African dictionary, the lexicographer’s only means of overcoming this inherent disadvantage is to maximise those features which contribute most toward the expressive needs of the user. Here we come to what can be considered the fundamental difference between Abraham and Bargery as far as English users are concerned: Abraham is by far the most useful when it comes to expressive needs whereas Bargery seems best for comprehension purposes. Parsons was certainly correct when he said of Abraham’s dictionary: ‘Lastly, the title of the book is really a misnomer. For it is essentially—more so than Dr. Bargery’s—a Hausa-English Dictionary [emphasis his]…’ (1950:528). But Parsons’ apparent disappointment that Abraham’s dictionary was not more analytical from a linguistic, structural point of view comes from his not fully appreciating what Abraham was trying to do, namely, to write what could be termed a ‘communicatively-oriented’ dictionary of Hausa. In the sections to follow, we will compare Bargery and Abraham for specific features which we feel consistently illustrate this conclusion. In the process of comparison, many features of Abraham’s innovative lexicographic methods can be appreciated. 3. Transcription There are two aspects of Abraham’s transcription where he clearly had the expressive needs of his users in mind.11 The first has to do with one of the primary functions of dictionaries generally, which is to provide guidance on pronunciation. In Abraham’s dictionary, every Hausa word is written in a complete phonological transcription. The user is thus guided in the pronunciation of entire utterances. More importantly, Abraham understood from the teacher’s point of view that in Hausa, where tone and vowel length play such an important role grammatically (e.g. to distinguish transitive from intransitive trisyllabic verbs or certain intransitive verbs from their verbal noun forms), full transcriptions are essential in helping the learner understand and acquire the grammar. His thoroughness and accuracy in marking vowel length and tone are quite astounding.12 His one disservice to the learner was his decision not to mark the distinction between the two R’s in Hausa (flap vs. trill), a task which would not have been onerous since Bargery had
8
The recently published An English-Hausa Dictionary (Newman 1990) is just such a dictionary. The small, practical Modern Hausa-English Dictionary (Newman & Newman 1977), though produced in Nigeria ostensibly for Hausa students, is perhaps more fully satisfactory for the decoding needs of English speakers. 10 Skinner’s Kamus na Turanci da Hausa (1965) was designed specifically for Hausa students and best serves their comprehension needs in English. 9
68
ROXANA MA NEWMAN
already taken the trouble to mark this difference. In this paper, the difference is consistently marked. As for full transcriptions in Bargery, these are only given for headwords and accompanying inflected forms (e.g. plurals of nouns, morphosyntactic variants of verbs), as well as about 1,000 out of some 15,000 illustrative examples—on what basis these were chosen is not clear. Bargery’s transcription is entirely acceptable for comprehension purposes, but it falls short for those who may try to use his examples to express themselves, particularly when these include morphosyntactic changes signalled only by tone and/or vowel length.13 A second pedagogical feature of Abraham’s transcription has to do with tone. Although Abraham was the first scholar to correctly analyse the Hausa tonal system and mark tones throughout his texts, he departed from a normally rigorous phonemic marking in his treatment of what he called ‘sentence or interrogative tones’, these being diacritics placed under the precise syllables in an intonational phrase where the speaker should shift into a higher pitch register, as in yes/no questions and certain expressions of contempt. These sub-phonemic ‘interrogative tones’ are consistently marked throughout the dictionary. Such pitch indications are clearly at the level of phonetic accuracy and reflect Abraham’s emphasis on illustrating spoken Hausa. The same concern for phonetic accuracy must have also motivated his decision to mark the assimilation of / ‘false accusation’ and kāshim ‘slag’, another specific departure from Bargery’s n/, cf. kāshiŋ practice. We can only interpret such departures as reflecting Abraham’s overriding concern for indicating phonetically accurate native-like pronunciation by the use of extra visual aids (but see Carnochan’s paper (this volume) for a somewhat different interpretation). 4. Alphabetisation of glottalised consonants There is one other area which further illustrates Abraham’s willingness to sacrifice a linguistically sound principle for a practically motivated one. This has to do with the expedient of intermingling the four glottalised stops with their non-glottalised counterparts in the alphabetisation of Hausa words. For example, words beginning with either of the two types of K are not separated into their own letter sections,14 but , , etc. Such a practice is a clear invitation to sequenced as follows: criticism (see Parsons’ 1950 review), as it simply misrepresents the distinctiveness of the phoneme pairs in question. In the preface to another later dictionary, Abraham recalled this criticism (of his Hausa dictionary) in defending his decision, again, not to separate the two types of D in Somali: I was unmoved by this criticism and have repeated this crime (!) here; the reason is that few if any Europeans are able to distinguish one D from another when said, far less to produce the sounds accurately. To put the sounds in different headings would lead to the same result, namely the user of the book having to hunt from one letter to the other. (1964:viii)
11
These also apply to his Hausa grammars (Abraham 1934, 1941, 1959b) and reader (Abraham 1959a). The one area where Abraham, and likewise Bargery, were misguided in their phonological analysis of Hausa concerns the length of word-final vowels: they both assumed, quite incorrectly, that low tone final vowels were automatically short (see Gouffé 1965). Thus both dictionaries are unreliable with respect to the quantity of word-final vowels. Interestingly, Abraham’s early 1934 grammar correctly marks as long the final vowels of High-Low disyllabic verbs in pre-dative position, but his 1941 grammar follows the erroneous assumption mentioned above (see also Jaggar in this volume). Another systematic error made by Abraham, but not Bargery, was in assuming that the vowels /i/ and / u/ were always long when preceding /y/ and /w/ respectively. This last error is particularly unfortunate, given the high frequency in the language of such sequences. 12
LEXICOGRAPHIC METHOD IN R.C.ABRAHAMS ‘HAUSA DICTIONARY’
69
One has to admit that there is a stubborn logic in his position. 5. Headwords and verbal derivatives One of the main reasons Abraham was able to make his dictionary so much more compact than Bargery’s relates to his decision to exclude as headwords the numerous productive verbal derivatives such as statives, past participles, intensives/pluractionals, and weak verbal nouns, so long as these were completely predictable, phonologically, morphologically, and semantically, from their roots. Another large productive class of verbal derivatives are the secondary verb grades marked by final -ō (Grade 6) to indicate ‘direction towards the speaker’ and -u (Grade 7) passive/sustentative (see Parsons 1960). Bargery faithfully records all these possible derivatives as separate headwords,15 although he cross-references them to the basic verbs, which usually have no exemplification of the derivative forms. The general exception is with certain common derived verbs with highly lexicalised meanings, such as fitō ‘rise (of sun); germinate (of seeds)’, cf. fìta ‘go out’. Bargery almost never exemplifies the more truly secondary -ō verbs such as barō ‘leave (in ‘leave’), as in nā barō shì gidā ‘I left him there at home’. Yet we know that that direction)’ (cf. derivative -ō verbs in particular are used with high frequency in the spoken language, as they contribute often subtle nuances to the basic lexical meaning. The ability to manipulate these forms, in fact, is one of the marks of native speaker fluency in Hausa. Abraham’s practice reflected his awareness of this because he very frequently used -ō as well as -u verbs in his Hausa examples under the basic verb, while generally excluding them as headwords. Typical examples are seen in (2) through (6),16 where Bargery’s and Abraham’s entries are compared: (2)
‘look for’
B: A: (3)
[lists nēmō and as headwords, with cross references to the basic verb, but no exx. of deriv. under any of these headwords] … they adduced reasons… sun nēmō ‘arrange in small heaps’
13
The major concession Bargery made was to mark the long vowel of the singular subject pronouns in the completive aspect, to differentiate these pronouns from their homographic counterparts in the other tense-aspects. 14 Actually this point is also true of Bargery’s dictionary, but we have no way of knowing how the decision was reached nor his justification for so doing. 15 The exhaustive inclusion of such grammatically predictable forms means that Bargery’s dictionary is more selfcontained than Abraham’s, which makes frequent cross-references to his grammars. Bargery never published a grammar of Hausa. 16 Throughout this paper, the citations taken from Bargery (=B.) and Abraham (= A.) are not in their own transcriptions, but in a regularised transcription where high tone is unmarked, low tone is marked by a grave accent, falling tone by a circumflex, and long vowel by a macron (except when it also has falling tone). Comments in square brackets contain the following abbreviations: deriv.= derivative; E.=English; equiv.=equivalent; ex(x).=example(s); lex.=lexical; trans.=translation. The Hausa headword is recapitulated in the examples by its first letter.
70
(4)
(5)
(6)
ROXANA MA NEWMAN
B: [no ex. of deriv.] A: hankàlinsà yā kàsu shìga ‘enter’ B: [no ex. of deriv.] shigôwā A: ìsa ‘reach’ B: [no ex. of deriv.] A: yā isō nân dà isôwā! ‘compress, squeeze together’ B: [no ex. of deriv.] A: sun matsō kusa dà shī they approached him
he is bewildered [lit. ‘his wits are in heaps’]
he will be arriving tomorrow
he reached here welcome on your arrival!
In addition to providing many more examples of derivational forms, Abraham also took care, again more than Bargery, to illustrate the various inflectional forms of verbs, such as pre-noun direct object and prepronoun direct object forms (which generally differ from each other only by a change in the length of the final vowel of the verb), as well as verbal nouns and participial forms, e.g. isôwā in (5) above. 6. Lexical equivalence and collocational information In the treatment of lexical equivalence, Abraham had a clear understanding of the fundamental aim of bilingual lexicography, namely, to ‘coordinate with the lexical units of one language those lexical units of another language which are equivalent in their lexical meaning’ (Zgusta 1971:294). The task is difficult, given the basic ‘anisomorphism’ or lack of fit between languages, simply because the same designata may be organised differently in different languages. For example, words denoting sensory qualities are nouns in Hausa but adjectives in English, and this fact has structural consequences when translating from one language to the other. An additional difficulty is that lexical equivalents between source and target languages should ideally be ‘insertable equivalents’, meaning that they can be smoothly inserted into a translation in the target language. It is perhaps in the handling of this fundamental lexicographic problem of lexical equivalence that Bargery and Abraham differ the most, particularly in their treatment of verb entries. , a verb with multiple discriminations of meaning dependant An extensive example of the Hausa verb on context, illustrates the differences in their approach to this problem: (7) Bargery: (a) beat (drum) yā k. gàngā (b) beat up, whisk (egg), churn (milk) (c) stir up (d) spin (cotton) (e) wag (tail) kàrē yā k. (f) shake thing in one’s hand, brandish (e.g. spear) shake, nod (head) yā k. kâi
[no E. trans. of ex.] [no ex.] [no E. trans. of ex.] [no ex.] [no E. trans. of ex.] [no ex.] [no E. trans. of ex.]
LEXICOGRAPHIC METHOD IN R.C.ABRAHAMS ‘HAUSA DICTIONARY’
71
cause to sway (wind) [no ex.] shake dust off garment, etc. [no ex.] Abraham: (a) yā k. gàngā he beat a drum [no E. lex. equiv.] she churned milk [no E. lex. equiv.] (b) tā k. whisk egg [no E. lex. equiv.] k. (c) stir thing of thin consistency (e.g. soup) she added kuka to the broth tā k. shake garment [no E. lex. equiv.] (d) k. (e) k. audùgā spin cotton [no E. lex. equiv.] it wagged its tail [no E. lex. equiv.] (f) yā k. (g) brandish [no ex.] wind swayed the tree [no E. lex. equiv.] (h) yā k. kâi he nodded in assent, shook head in dissent [no E. lex. equiv.] (i)
The differences between the two treatments of the verb are very clear. Bargery’s subdivisions provide an English lexical equivalent in each case, usually followed by a Hausa example but no English translation. Abraham does just the opposite, i.e. he provides no English lexical equivalents, but gives Hausa examples of the verb’s uses followed by their English translations. This particular verb has been chosen because it illustrates two important points about lexicographic was the verb that Parsons used (1950:528–29) to show how Abraham frequently method. Firstly, neglected the lexicographer’s primary duty to provide lexical equivalents. Yet to translate this verb by ‘shake, beat, stir’ etc. (as Bargery does) only creates artificial subdivisions. In Parsons’ view, the notional, semantic unity underlying the multiple senses (a) through (i) is captured by a single superordinate verb in English, ‘agitate (generally up and down)’. This verb17 does indeed seem to explain the polysemous nature . However, we have to disagree with Parsons in favour of Abraham’s methodology, given our view of of Abraham’s dictionary as a practically oriented, translation dictionary. While it may be true that ‘agitate’ , it doesn’t necessarily mean that it is a good equivalent in the sense captures the sememic essence of of translation dictionary equivalent. Superordinate terms are generally ill-suited as insertable equivalents in might be in a a translation dictionary. The only context where ‘agitate’ could possibly be translated by sentence such as ‘don’t agitate the contents of the box’ (although even here, ‘shake’ is the more natural English verb). Even the highly concrete use of ‘agitate’ in ‘the mixer agitates the cement first before pouring it out’ would require a completely different Hausa verb, garwàyā/gauràyā, which corresponds more to English ‘mix’. As for the other English contexts of (7) above, ‘agitate’ would be totally inappropriate. in a Hausa-English translation Thus ‘agitate’ is generally not the lexical or translation equivalent of dictionary. Both Bargery and Abraham understood this. We now come to the second point. Why didn’t Abraham simply follow Bargery’s solution for sorting out ? What we have to realise here is that this Hausa verb, apart from denoting an the polysemy of agitating movement, is a lexically ‘underspecified’ verb; it always draws on surrounding lexical material, usually nouns occurring as its subject or object, for its semantic interpretation. In other words, the lexical requires collocational information for its specific interpretation in a given context. It only meaning of
17
We are dealing here only with the concrete, physical meaning of ‘agitate’, not its more figurative or emotive uses.
72
ROXANA MA NEWMAN
translates as English ‘beat’ (which itself has multiple meanings) when collocating with gàngā or a similar musical instrument. Similarly, ‘spin’ (another polysemous English verb) is the correct lexical equivalent only if the noun object is a ‘top’ and, by extension, ‘cotton’ (because Hausa women spin yarn using a small top-like object). Bargery, as we saw, handled the problem of collocational information by providing it in parentheses immediately after the various ‘equivalents’, e.g. ‘beat (drum), churn (milk), cause to sway (wind)’. When he did provide examples, they were not usually translated (presumably because it would be redundant and wasteful of space). In turning to Abraham’s treatment, we can see immediately how much more helpful his is to the learner for expressive purposes; moreover, his is more economical. Abraham rejects wanting to use Bargery’s way of providing translation equivalents with context noted parenthetically, e.g. ‘churn (milk)’. ‘she Instead, he simply illustrates the total Hausa context with an English translation, e.g. tā churned milk’. Such a method accomplishes several things at once. It shows the user that there is no single English ‘equivalent’ for this word. It provides appropriate collocating Hausa lexemes, instead of parenthetically indicating these in English. Finally, in this particular example, the use of the feminine subject pronoun adds a minimum of cultural information. The one exception in (7) where Abraham actually provides a lexical equivalent but no Hausa example is sense (g), ‘brandish’, an unambiguous verb merely requiring any object which can be used menacingly as a weapon. In sum, his treatment of shows that he was using good pedagogical practice in lexicography. Abraham’s technique is of course most appropriate in cases where the Hausa word either has no lexical equivalent in English or owes its polysemy to specific collocational restrictions, often due in both cases to divergent cultural practices and norms. The four examples in (8) through (11) illustrate this point. In each one, Bargery’s English glosses are actually not lexical equivalents but phrasal or explanatory equivalents, containing all the collocational information entailed by the verb. Moreover, he doesn’t provide examples. Abraham, by contrast, doesn’t try to give equivalents or explanations, but simply illustrates the verbs’ uses in complete sentences which provide the essential collocational and syntactic information that the learner would need: (8) B: A:
spit out kola from mouth; squirt water from mouth (by elephant; by a person who has lost one or more front teeth, etc.) [no ex.] he spat out bits of kola [no trans. equiv.] yā f. yā f. ruwā he (elephant) sprayed out water (from mouth) [no trans. equiv.]
(9) B:
(10)
slide down a rock come to a dead stop yā d. dàgà kân A: d. ruwàitā B: spread news about; noise abroad A: yā ruwàità yā ruwàitā minì
[no ex.] [no ex.] the boy slid down rock in play pull horse up dead [no ex.] he spread, related the news he related the news to me
(11) B:
pull, get things out of hole or vessel (e.g. honey from hole in tree, cotton from gourd in which it has been stored, kola-nuts from
LEXICOGRAPHIC METHOD IN R.C.ABRAHAMS ‘HAUSA DICTIONARY’
A:
73
packing) by inserting whole hand or using stick or other object and not merely a finger [no ex.] (a) poke, hunt (rat, person) out of place, hole, etc. he prodded the rat out of the hole (b) dislodge thing firmly embedded in matrix (e.g. honey from tree, kola-nuts from their huhu-packing, etc.) [no ex.] (c) he picked his teeth [no trans. equiv.]
In example (11), Bargery gives a lot of detailed explanatory information for ; and yet, without any examples, it is difficult to know how much of it is subsumed in the semantical specification of the verb. For example, it is not clear from the explanation whether the information about inserting the whole hand needs to be syntactically expressed in the sentence, e.g. by a prepositional phrase. By contrast, Abraham’s much clearer subdivided entry is totally unconcerned about the role the hand may play but shows, instead, how to compress the essential semantic information (of poking or routing things out of their places) by the elegantly simple device of switching to the -ō derivational form of the verb (action in direction of the speaker). In all the verbs shown above, Abraham’s entries consistently show the reader how to use the verb directly in a phrase or sentence while deferring certain semantic information until the translations, whereas Bargery, on the other hand, attempts to put that same semantic information in the glosses but fails to show how the verb can be used. 7. Illustrative material As we have just seen, the pedagogically most relevant aspect of Abraham’s lexicographic method was to let the various meanings and nuances of Hausa words manifest themselves through full exemplification and subtle translation, with careful attention paid to the collocational potentials of both Hausa and English utterances. It is in his pervasive use of illustrative examples throughout the dictionary to show not only semantic information, but also syntactic, contextual, and idiomatic usage, that Abraham’s dictionary is most successful from a pedagogical point of view. Even in cases where the context would be self-evident, Abraham goes the extra step to provide the English user with typical illustrations. A good example of this is in (12): (12) B: A:
(a) (b) (a)
(b)
English word order and used in the same way [no ex.] a forbidden place [no ex.] he issued an order order, command yā kafà bîn he disobeyed the order yā kāyā he ordered some goods yā yi nē access to the place is forbidden wurîn forbidden place wurī mài yā shìga he entered a forbidden place
For meaning (a), Abraham’s examples give the user the collocating verbs appropriate to the two syntactic uses of English ‘order’. Meaning (b) is actually quite idiomatic, giving the sense of a place which has been
74
ROXANA MA NEWMAN
‘ordered off limits’, probably in the military sense. Such examples and translations at least help to narrow this connotation down. 8. Synonyms In providing his translation equivalents, Abraham aimed for the most direct or common ones, and assiduously avoided listing strings of synonyms in English. This is quite in contrast to Bargery, as seen in (13): (13)
fushī B: A:
anger, vexation, peevishness, bad temper, huffiness, touchiness, exasperation [no ex.; many Hausa synonyms also given] anger [many exx.; no English or Hausa synonyms]
Looking at Bargery’s undifferentiated list of English synonyms, i.e. words having the ‘same’ meaning, one is left to wonder in which contexts fushī can be used to convey ‘huffiness’ or ‘touchiness’, which in English have quite different connotations from either ‘anger’ or even ‘vexation, exasperation’. And since Bargery also lists a number of other Hausa words which are claimed to be synonymous with fushī, there is the further question of how to possibly match them up lexically in terms of shared connotative values. ‘huge’ with Bargery often goes to great lengths in listing Hausa synonyms, as seen in his entries for its list of 135 ‘equivalents’ (see note on p. 1154) and kinibībì ‘malingering’, with 95 words which ‘all bear some relation’ to it (p. 606). Some lexicographers are quite opposed to the listing of synonyms in bilingual dictionaries. As one prominent lexicographer has put it: ESL and bilingual dictionaries don’t include synonym discriminations, since such rarified distinctions are quite beyond the skills of their users and altogether irrelevant to their purpose. Their aim is not to provide exquisite sensitivity in another language but to provide the means for communicating competently with a native speaker of another language. That goal is tough enough without worrying about the fine distinctions of connotation. Synonym discriminations are strictly for the native speaker . (emphasis mine; Landau 1984:110) When taking into consideration the additional functional dimension of expression vs. comprehension in bilingual dictionaries, Kromann et al. (1984: 210) are even more critical of synonyms: The accumulation of equivalents in an active [=expression] dictionary, without any meaning discriminating glosses, is one of the deadly sins of lexicography, but accumulation is possible in a passive [=comprehension] dictionary. The fact that Bargery doesn’t hesitate to list synonyms again characterises the focus in his dictionary on comprehension. Bargery in fact is often consulted for its rich store of synonyms, rare words, and dialectal forms, in much the same way that native speakers consult unabridged monolingual dictionaries. In contrast, Abraham’s specific avoidance of synonyms, either English or Hausa, shows that he had already anticipated Landau’s warning in constructing his dictionary for pedagogical purposes.
LEXICOGRAPHIC METHOD IN R.C.ABRAHAMS ‘HAUSA DICTIONARY’
75
9. English translations of examples In many of the examples illustrated above, Abraham’s succinct translations help to pinpoint subtle semantic features of the Hausa lexemes. This is also true of cases where Abraham’s Hausa illustrations are exactly identical to those found in Bargery under the same headwords.18 A close examination of many of these shared Hausa phrases reveals that Abraham’s translations of them differ markedly from Bargery’s. The difference is that Abraham retranslated them into more direct, idiomatic spoken English, a device which he used to suggest that the Hausa counterpart was similarly vivid and colloquial, hence belonging to spoken Hausa. Compare the two English versions of each of the Hausa phrases in the examples below: (14)
(15)
(16)
(17)
(18)
cafkàtā: an cafkàcfè ruwā à àbinci B: an excess of water has been added to the food A: the food is too watery : an rabà àbin B: they have severed their friendship with mutual recriminations A: they are no longer on speaking terms ruwā: an bā kà ruwā B: by a deliberately false offer of a price which there was never any intention of paying, you have been deluded into thinking you can get a bigger price than it is worth A: he ‘led you (seller) on’ by overbidding for it cîn B: a girl who has been tampered with by men A: a girl who is no longer a virgin : an yi masà B: he’s been let off the consequences of his delinquency because of his position A: he’s been leniently treated because of his rank
In each of these examples, Bargery’s translations are circumspect, wordy, full of circumlocutions more appropriate to a formal register or written style of language. By contrast, Abraham’s renderings are direct and natural, having the convincing sound of conversation. One has only to compare Bargery’s clumsy rendering of (16) with Abraham’s masterfully concise translation to see how the latter is able to match the stylistic level of the short idiomatic Hausa expression. Thus it is by the quality of his English translations that Abraham is able to convey the really idiomatic style of spoken Hausa.19 Another area that illustrates Abraham’s adeptness in idiomatic translation is his treatment of Hausa interjections and exclamations. Compare the totally different methodological approaches in the following examples: (19) B: A:
18
ejaculation of astonishment fancy that!
Abraham’s very free use of Hausa examples copied verbatim from Bargery can be better understood when we recall his attitude about Bargery’s dictionary being ‘our joint Hausa-English dictionary’, see footnote 4 above.
76
(20)
(21)
ROXANA MA NEWMAN
gabjì B: exclamation of surprise at severe blow with stick, collision, size, quantity, use of unnecessarily large amount of household commodity, severe stumble, etc. A: what a slap! stumble! collision! prodigality! wâyyō B: an exclamation at feeling of sharp pain, at hearing disquieting news, at the occurrence of a mishap, etc. A: my word!
(22) B: A:
an expression of reproof; used also in dismissing anyone in anger or contempt (a) dà kai damn you! (b) dà shī blast him!
Bargery resorts to description or explanation as to what the exclamations mean or refer to. By contrast, Abraham translates them directly into semantically equivalent English exclamatory expressions. Here, as elsewhere, Abraham has provided completely ‘insertable’ equivalents. 10. Encyclopedic vs. strictly linguistic knowledge A final and salient difference between Bargery and Abraham concerns the kinds of information each included in his dictionary. Both authors included an enormous range of vocabulary covering almost every aspect of Hausa culture, and illustrated these words with phrases and sentences to show their semantic range and syntactic distribution. But where they again differed, in a fundamental way, was in the type of additional material they would allow, once translation equivalents and examples were taken care of. We can go through Bargery’s dictionary page after page and see that there is an enormous amount of extralinguistic knowledge of an ethnographic, historical, and even psychological nature in many of his entries.20 This need not be surprising, given the very great differences that exist between Hausa and British cultures which would be reflected in the respective languages. So, for example, under the name of a titleholder in the entourage of a traditional Hausa ruler, Bargery might very well describe the historical origins of the title, the duties of such a titleholder, or even the names of famous historical people who had held that office and in which city. Abraham, however, completely avoided such information, putting only as the equivalent the simple explanation, ‘an official title’, followed perhaps by appropriate praise epithets associated with these titleholders. Even in the relatively culture-free areas of vocabulary such as verbs, Bargery was quite inclined to add very specific details which more properly belong in an ‘ethnography of speaking’, i.e. a description of the social/cultural conditions and the interlocutors involved, under which linguistic forms are likely to be used. An example is the derived verb dākàcē, whose two treatments are compared below: (23)
19
dākàcē
This is not to deny the numerous useful examples and good idiomatic translations found in Bargery, but only to show that they are much less consistently present in his work.
LEXICOGRAPHIC METHOD IN R.C.ABRAHAMS ‘HAUSA DICTIONARY’
77
B: wait about in a place and refrain from leaving it (either through diffidence or unwillingness because awaiting another’s permission to proceed; e.g. p. waiting outside room for permission to enter; guests waiting about through diffidence and not taking part in festivities until called in; donkey refusing to enter compound) ya sâ yā d. à wàje? why has he been waiting about outside? A: wait (=dākàtā 1, 2) [under basic verb dākàtā, many exx. and syntactic patterns]
For Abraham, the type of long explanatory description given by Bargery was information that was lexically irrelevant, since it is simply impossible to predict the kinds of reasons one might have for hanging around a place. Therefore Abraham simply translated the verb with its basic lexical meaning ‘wait’. Throughout his dictionary, Abraham was very clear about excluding information that was not strictly linguistic or of lexical relevance to the word under consideration. For him, what was central to the translation of a word was its basic lexical content and its semantic and syntactic relations to other words in the sentence. The one aspect of ‘culture’ richly documented by Abraham was the area of proverbs and praise epithets. His inclusion of such expressions of verbal culture, however, remains completely consistent with his pedagogical aims, as proverbs and epithets are prime examples of the idiomatic use of language. Moreover, such expressions clearly reflect spoken language, as nothing is more indicative of expressive mastery of a language than the knowledge and the ability to use them. In other words, proverbs and epithets, while indicating much about the cultural concepts and values of a people, are essentially linguistic forms.21 Abraham’s enthusiasm for proverbs sometimes ran to whole columns—see especially under kàrē ‘dog’ (35 proverbs), kūrā ‘hyena’ (44), ruwā (43), where they are even set apart as separate subsections within the headword. Interestingly, Bargery does not give a single proverb under these common nouns. We can provide no more striking example to illustrate the fundamentally different approach to Hausa lexicography of Bargery and Abraham vis-à-vis extra-linguistic knowledge than to refer to their entries for ‘kolanut’. The only thing the entries share in common is the the specifically culture-bound word translation equivalent of the headword as ‘kolanut’. Beyond that, they are totally different kinds of entries. Bargery devotes a column and a half to describing various kinds of kolanuts, their names, sizes, colour, textures, geographical distribution, factors affecting their growth, etc. Abraham, in contrast, provides a dozen phrases showing how Hausas talk about this important cultural item: he lists its epithets, proverbs in which it appears, its metaphorical meanings, some of its immediate collocates, and only the names of those as part of their compound names. types of kolanut which actually have the word 11. Conclusion To summarise, we have examined the dictionaries of Bargery and Abraham for specific points of difference in their lexicographic methodology, differences which reflect an essentially different focus on their part. Bargery, in a quantum leap over his predecessors’ works, clearly set out to provide as rich a repository of the Hausa language as possible. Users of his dictionary cannot fail to find a wealth of varied information to help them in their task of understanding the language and its culture.22 Abraham, on the other hand, while 20
It is undoubtedly some of this material that Abraham had in mind when he complained about not having any say over ‘rejection of matter’ (see footnote 4 above).
78
ROXANA MA NEWMAN
also aiming at comprehensiveness, was more guided by pedagogical objectives. His materials are organised with the clear aim of catering to his users’ expressive/active needs. This practical focus makes Abraham’s dictionary fundamentally different from Bargery’s in both structure and substance. His pedagogical aims are embodied in his full phonological as well as subphonemic transcriptions, his preference for direct illustrations containing extensive collocational information (as opposed to the simple listing of translation equivalents and synonyms), his conversationally flavoured idiomatic English translations, and finally the general avoidance of extralinguistic information in his entries. In short, Abraham has written a very different dictionary from that of Bargery, even though they drew upon much of the same material. Undoubtedly, the individual distinctiveness and value of each dictionary has prompted the gratitude expressed in Parsons’ (1950:524) remark: ‘Hausa has indeed been fortunate in its lexicographers’. REFERENCES Abraham, R.C. 1934. The Principles of Hausa. Kaduna: Government Printer. ——. 1941. A Modern Grammar of Spoken Hausa. London: Crown Agents for the Colonies (published on behalf of the Government of Nigeria). ——. 1958. Writing African dictionaries. West Africa, no. 2152, p. 659. ——. 1959a. Hausa Literature and the Hausa Sound System. London: University of London Press. ——. 1959b. The Language of the Hausa People. London: University of London Press. ——. 1962. Dictionary of the Hausa Language. 2nd ed. London: University of London Press. ——. 1964. Somali-English Dictionary. London: University of London Press. Abraham, R.C., and Mai Kano. 1949. Dictionary of the Hausa Language. 1st ed. London: Crown Agents for the Colonies. Armstrong, Robert G. 1964. Roy Clive Abraham, 1890–1963. The Journal of West African Languages 1(1):49–53. Bargery, G.P. 1934. A Hausa-English Dictionary. London: Oxford University Press. Gouffé, Claude. 1965. La lexicographie du haoussa et le préalable phonologique. Journal of African Languages 4: 191–210. ——. 1981. Comment recueillir et éditer les proverbes haoussa: quelques suggestions pratiques. Itinérances II, pp. 115–135. Paris: Société des Africanistes. Hair, P.E.H. 1965. A bibliography of R C. Abraham—linguist and lexicographer. The Journal of West African Languages 2(1):63–66. Kromann, Hans-Peter, et al. 1984. ‘Active’ and ‘passive’ bilingual dictionaries: the Scerba concept reconsidered. In LEXeter ’83 Proceedings, ed. by R.R.K. Hartmann, pp. 207–215. Tübingen: Niemeyer. Landau, Sidney I. 1984. Dictionaries: The Art and Craft of Lexicography. New York: Scribner’s. Lukas, J. 1949. Review of R.C.Abraham and Mai Kano, Dictionary of the Hausa Language. 1st ed. Africa 19:250–252. Newman, Paul, and Roxana Ma Newman. 1977. Modern Hausa-English Dictionary (Sabon Kamus na Hausa zuwa Turanci). Ibadan & Zaria: University Press (Nigeria). ——. 1991. Lexicography of the Chadic languages. In Wörterbücher: Ein internationales Handbuch zur Lexikographie, vol. 3, ed. by F.J.Hausmann, O. Reichmann, H.E.Wiegand, and L.Zgusta, pp. 2457–2460. Berlin: Walter de Gruyter.
21 Abraham, characteristically, does not provides the appropriate cultural contexts by which the reader might know how and when to use these proverbs and epithets. For a detailed and insightful discussion on the collection and classification of proverbs, see Gouffé (1981). 22 In our comparisons, we have omitted mention of a major feature of Bargery’s dictionary designed for users’ expressive needs; namely, his inclusion of a 73-page English-Hausa vocabulary appended to the body the dictionary. Abraham’s dictionary does not have a similar index.
LEXICOGRAPHIC METHOD IN R.C.ABRAHAMS ‘HAUSA DICTIONARY’
79
Newman, Roxana Ma. 1974. Dictionaries of the Hausa language. Harsunan Nijeriya 4:1–25. ——. 1990. An English-Hausa Dictionary. New Haven: Yale University Press. Parsons, F.W. 1949. Review of R.C. Abraham and Mai Kano, Dictionary of the Hausa Language. 1st ed. African Affairs [JRAS] 48:338–339. ——. 1950. Review of R.C.Abraham and Mai Kano, Dictionary of the Hausa Language. 1st ed. BSOAS 13:524–529. ——. 1960. The verbal system in Hausa. Afrika und Übersee 44:1–36. Skinner, Neil. 1965. Kamus na Turanci da Hausa. Zaria: NNPC. Zgusta, L. 1971. Manual of Lexicography. The Hague: Mouton.
REFLECTIONS ON R.C.ABRAHAM’S SOMALIENGLISH DICTIONARY B.W.Andrzejewski
1. Introduction Roy C.Abraham is the author of the Somali-English Dictionary (1964) and English-Somali Dictionary (1967), and though he completed the first, he died before the second was finished. The Somali-English Dictionary represents a substantial contribution to Somali lexicography, but the English-Somali Dictionary is an unfinished work, published as he left it, without any editing and checking of the accuracy of translations. Many entries in it are mere notes in English by the author to himself as to how he should deal with the translation of a particular word. It can, however, be of use to learners of Somali or to Somali learners of English if they are sufficiently advanced to cope with its idiosyncratic and incomplete state. It can also be of value, as a source of inspiration, to future compilers of English-Somali dictionaries, on account of its innovative technique of presentation. Instead of giving Somali equivalents of English words in isolation, Abraham often puts them in whole sentences, thus introducing the reader to Somali phraseology and explaining the meaning at the same time. It would be unfair to pass any judgment on a work which the author left unfinished, and as there seems little else that can be said about Abraham’s English-Somali Dictionary I shall confine my attention here exclusively to his Somali-English Dictionary. In any assessment of a scholar’s work, it is essential to place it in the context of his time. We have to examine the state of knowledge which preceded his contribution and take into account the circumstances under which his work was conducted. As I shall endeavour to show in this paper, Abraham’s work was of a truly pioneering nature and was carried out under very difficult conditions. 2. Abraham’s predecessors in Somali lexicography Abraham began his lexicographical research into Somali some time in 1948, at first as an adjunct to his work on Somali grammar, and it continued until 1960, when it culminated in the completion of the typescript of his Somali- English Dictionary. At that time, the Somali language had no official orthography and all written communication in Somali-speaking territories was conducted in Italian, English, French or
African Languages and Cultures, Supplement 1 (1992):99–109
REFLECTIONS ON R.C.ABRAHAM’S ‘SOMALI-ENGLISH DICTIONARY’
81
Arabic, even though all Somalis shared the same spoken language and had at their disposal a dialect type used as a lingua franca in all regions, accepted as standard by all radio stations which broadcast in the language. In Somalia, some private systems of writing were used, but there were hardly any publications available in them. The texts published by European scholars were very few and most of them were not entirely reliable.1 Thus Abraham did not have the possibility of using the method which is recognised as the best in lexicographical research, namely, the scanning of a large body of texts in search of words. When he began his research there were only three dictionaries of Somali in existence, those of Leo Reinisch (1902), Fr. Evangeliste de Larajasse (1897) and Fr. Giovanni Maria da Palermo (1915). None was of such a standard of accuracy that its data could be immediately incorporated in a new dictionary without being thoroughly checked with language informants. Although Reinisch was a very prominent academic linguist in his time, he did not work long enough on Somali to arrive at a completely satisfactory analysis of its phonology and tonology, and this was reflected in the transcription he used. De Larajasse and da Palermo spent a long time among Somalis but were amateur linguists, and their ways of writing Somali were far from satisfactory. In view of this, Abraham had to collect or to verify all his data from language informants, by no means an easy task. 3. Abraham’s language informants Between 1948 and 1960 there were hardly any Somalis whose command of English was of such a standard that they could act as research partners with Abraham in the full sense of the word. There were no Somalis with training in linguistics who could check his data for him, help him with editing his dictionary, or with whom he could discuss the theoretical issues involved in his work. All his research into Somali was done in London with two informants, Saleebaan Warsame and Cali Garaad Jaamac, whose names in their Anglicised spelling are Solomon Warsama and Ali Garad Jama. I had the privilege of knowing them personally; the former was a café proprietor in the East End of London who had hardly any formal education but was fluent in spoken, though not always correct, English. The latter had some government school education in the Somaliland Protectorate and came to England in the hope of getting into a university or polytechnic, but the private arrangements on which he relied failed and he had to maintain himself by taking up casual work. His English was quite good but not of such a standard that his translations of the more difficult words in Somali could be used without editing. The educational shortcomings of Abraham’s informants were richly compensated by their excellent command of Somali. Though he had lived in England for a long time, Saleebaan Warsame was in constant touch with those countrymen of his who frequented his café; they were either sailors or shore workers, many of them recent arrivals from Somalia. He was a lively and witty conversationalist and had a good knowledge of the traditional Somali way of life. Cali Garaad came from a well-known and influential family in the then British Somaliland Protectorate and was a fully integrated member of his community. After Somalia became independent in 1960 he held, for a brief period, the post of Minister of Education. Like Saleebaan Warsame, he was eloquent and had a full command of his mother tongue. As far as I have been able to ascertain, these two men are no longer alive. This is very unfortunate since their reminiscences of working with Abraham would have been of great interest to this Symposium.
1
For bibliographical guidance to these publications see Lamberti (1986) and DeLancey et al. (1988).
82
B.W.ANDRZEJEWSKI
4. Circumstances under which Abraham worked on his ‘Dictionary’ Abraham’s work on Somali began when he was on the academic staff of the School of Oriental and African Studies from April 1948 till the end of the 1950–51 session as Lecturer in Amharic. I am not sure whether he received any financial assistance from the School for his Somali research, but I doubt it. At that time the prevailing policy of the School was to subsidise research only if it was related to the subject for which the particular member of staff was appointed. After his retirement in 1951, Abraham subsidised all his research from his own resources, which do not seem to have been abundant. Only when his Somali-English Dictionary was approaching completion did he obtain a contract with the government of the British Somaliland Protectorate, which, as far as I know, covered the cost of publication but did not include any research expenses or remuneration. What seems certain, however, is that his expenses must have been considerable, since presumably neither of his two informants would have worked for him without payment. During the late 1950s, Abraham’s health deteriorated progressively, and in the final stages of preparing his Dictionary it was only his remarkable tenacity of will and indomitable enthusiasm that kept him at his work. 5. A digression on the selection of the verbal lemma in Somali To understand an important aspect of the Somali-English Dictionary, it is necessary to take into account the problem which faces all lexicographers of Somali, namely the choice of the verbal lemma, i.e. the form which represents all the forms of a verb and is used in lexical entries.2 Unlike such languages as Italian, French, German or Russian, Somali has no infinitive which by itself could constitute a meaningful utterance. There are no equivalents of such forms as cantare, chanter, singen, pyet’, or even of the English phrasal form ‘to sing’. There are verbal nouns (gerunds) such as heesid ‘singing’,3 but not all verbs have them, and there is some overlapping when two conjugations share the same type of verbal noun, e.g. baaris ‘searching’ and karis ‘cooking’, corresponding to the verbs baar (1st conj.) and kari (2nd conj.), so that one cannot predict from the form of the verbal noun what the corresponding verb will be. The only verbal forms which can by themselves constitute meaningful utterances are those of the Independent Past Tense and the Positive Imperative. The former are not very suitable to provide a lemma since they are archaic and not very frequently used. The 2nd Person Singular of the Positive Imperative would be eminently suitable to serve as a lemma since it is composed only of the root of the verb and the root extension, if any, and it has been used in fact for that purpose in the dictionaries compiled by Reinisch (1902), de Larajasse (1897), Agostini et al. (1985), and Luling (1987). This form has, however, the great disadvantage that many Somalis reject the possibility of its occurrence in numerous verbs. They find such forms as biyow ‘turn into water’, madoobow ‘become black’ or furan ‘open’ (intransitive) faintly ridiculous since they could not be addressed to a person. It is only when such forms are put in the context of fables, accounts of supernatural events, curses, blessings or concessive phrases, that they are accepted. These objections need not seem at all frivolous when we consider that early lexicographers of English also did not accept such improbable lemmas. In Samuel Johnson’s dictionary all verbal entries are preceded by the word ‘to’, which is disregarded, however, in the alphabetical arrangement of entries. The two parts of
2
This problem is discussed at length in Agostini et al. (1981) and Andrzejewski (1987).
REFLECTIONS ON R.C.ABRAHAM’S ‘SOMALI-ENGLISH DICTIONARY’
83
such entries were graphically differentiated, e.g. ‘to WALK’. It was only in the 19th century that the Singular Imperative form was universally accepted as the verbal lemma in English dictionaries, but its acceptance made it necessary for it to be accompanied by a grammatical identification label showing that it was a verbal lemma. 6. Organisation of Abraham’s ‘Somali-English Dictionary’ An important feature of the Dictionary is that it includes an outline of Somali grammar in the form of an appendix to which frequent references are made in the Dictionary itself, thus reducing the length of entries in those cases where full explanations would have made them lengthy and unwieldy, e.g. the entries relating to syntactic particles (indicators). The Dictionary is well organised and provides all the essential information about individual words. Nouns are provided with the definite articles which identify their gender, and all the plural forms are quoted, which is important since they are often quite unpredictable. Abraham’s treatment of the lemma for the verbs is highly efficient, though unusual, and its use is possible only because of the grammatical appendix to the Dictionary. He selected for it the 3rd Person Singular Present Continuous Tense, even though this form cannot by itself constitute a meaningful utterance. It can, however, be easily supplied mentally with a minimum context by the user of the Dictionary if he looks up the information about verbs in the grammatical appendix. What is more, Abraham frequently gives short sentences to explain and illustrate the meaning of his verb entries, and thus places the lemmas in contexts in which they are fully meaningful. The lemma form chosen by Abraham has the great advantage that all the other forms of a verb can be predicted from it. 7. An assessment of the ‘Somali-English Dictionary’ The Dictionary is a very important contribution to Somali lexicography and represents an advance on its predecessors. It contains some 8,000 entries and his translations of Somali words show a high degree of accuracy. The only entries which have to be treated with great caution are botanical and ornithological terms, since no entirely reliable documentation of Somali nomenclature in these fields is available even now. Abraham does not state his sources of information but it is obvious that he could not have done the requisite research in London. Somali accentual patterns consisting of combinations of tone and stress play an important role as exponents of grammatical forms. Even though the accentual patterns, with rare exceptions, are entirely predictable if the grammatical forms are identified, it is very useful to mark them in a dictionary. Abraham uses tone marks (which also imply the occurrence of stress) throughout his Dictionary, and in the main entries they are quite reliable. In the examples of combinations of words, phrases or whole sentences, however, they are often inaccurate. Inaccuracies also occur in his treatment of the final vowels of verbs, e.g. *furayya instead of furayyaa, where difference of length is grammatically important. In the length of roots of words, however, his Dictionary is highly accurate. Another shortcoming of his Dictionary is that his transcription does not recognise the difference between fronted and retracted variants of the basic vowels in those cases where it results in a divergence in meaning,
3
For information on all the grammatical terms and designations referred to in this paper, see Saeed (1987).
84
B.W.ANDRZEJEWSKI
e.g. diidayya (with front vowel variants) ‘he is refusing’, and diidayya (with retracted vowel variants) ‘he is fainting’.4 The value of the Dictionary is very much enhanced by the use of exemplifying sentences, and the explanations of meaning show a high degree of sensitivity to the ethnographic background; some of the examples are proverbs, though this is not stated. 8. Delay in publication Although the typescript of Abraham’s Dictionary was ready for the printers by 1960, there was a considerable delay in its publication. The reason was that when Abraham was assured by the Somaliland Protectorate Government that they would subsidise publication, he assigned the copyright to them. The Somaliland Protectorate was granted independence in 1960, earlier than anyone expected, and merged immediately with the former Italian Trusteeship Territory of Somalia. As a result of this historical event, the obligations and rights of the former Somaliland Protectorate passed to the newly-formed Somali Republic. Unfortunately, its government did not honour the publication contract given to Abraham. This was, in my view, mainly due to the fact that in Somalia the question of a choice of script for an official orthography, which the government had promised to introduce, was at the time a highly sensitive political issue. A large section of the population was fiercely opposed to the use of the Latin script on religious grounds, and subsidising Abraham’s Dictionary might have been construed by some people as a sign of the government’s hidden intentions. Originally, publication was going to be arranged by the Crown Agents for the Colonies in London, but with the independence of Somalia they ceased to be responsible for the matter. Abraham took his work to the University of London Press, who were about to publish it in 1962 when a new difficulty arose. As the copyright was now held by the Somali Government, the Press had to obtain permission from them to publish it. For reasons already explained, the Somali Government were afraid to release the copyright and numerous letters from the University of London Press and Abraham himself remained unanswered. The book was already printed and bound in 1964, a year after Abraham’s death, but it could not be sold in bookshops. The University of London Press approached me, hoping that because of my contacts with Somalia I could persuade the Somali Government to release the copyright. It would have been impossible for me to do anything about it had it not been for the help I received from my friend and colleague, Muuse Xaaji Ismaaciil Galaal, with whom I worked from 1950 to 1954 in Somalia and at the School of Oriental and African Studies. Muuse by then was a well-regarded scholar and was the Chairman of the Somali Language Commission appointed by the Ministry of Education. In 1966 I visited Somalia and explained to him the difficulties about the copyright. As he knew his way about the newly-established corridors of power and had friends in high places, he managed to obtain a signature on a letter releasing the copyright, and Abraham’s Somali-English Dictionary then became available in bookshops.
4
In the dictionary compiled by Agostini et al. (1985), the words with fronted vowel variants are marked by a subscript cedilla, and those with retracted vowel variants by its absence. Yaasiin C.Keenadiid uses a raised circle for the same purpose. For an explanation of these diacritics see Saeed (1987:18–19).
REFLECTIONS ON R.C.ABRAHAM’S ‘SOMALI-ENGLISH DICTIONARY’
85
9. Somali lexicography since 1966 In the year when Abraham’s Somali-English Dictionary finally became available, A.Keene Spitler and Helen Spitler published their English-Somali Dictionary (1966). Though modest in scope and written in a phonetically inaccurate transcription, it was a useful adjunct to Abraham’s work, since his own EnglishSomali Dictionary, which appeared a year later, was an uncompleted work. In 1969 Stepachenko and Maxamed Xaaji Cusmaan published their Somali-Russian and Russian-Somali Dictionary, a modest but competent and useful work, written in a reliable transcription. In 1972 an event occurred which has influenced very substantially the development of Somali lexicography. In that year Somalia introduced an official orthography in Latin script, which was the result of over two decades of linguistic research. At the same time, Somali became the sole official language of the state in all public business, and within a few years it replaced English and Italian as the medium of instruction in all pre-university education. Intensive work was undertaken by the Somali Language Commission and the Ministry of Education to create modern technical and scientific terms for Somali (see Caney 1984), and at the National University a Department of Somali Language and Literature was established, and training in linguistics was provided with special reference to Somali. As a result of these developments (see Andrzejewski 1983), there were now Somalis fully literate in their own language and some had linguistic training. They were capable of undertaking lexicographical research by themselves, and if they worked with expatriates they could act as their partners and not merely as language informants. What is more, since the introduction of the official orthography, the State Printing Agency has published numerous books and journals and a national daily in Somali has been established. Lexicographers thus had at their disposal a large body of texts which they could scan in search of words left out by their predecessors. Several dictionaries have been published since the introduction of the official orthography of Somali. In 1976 three were compiled, two of a modest size: a monolingual dictionary of Somali by Cabdulqaadir Faarax Bootaan (1976), and a Somali-French dictionary by C.Philibert (1976). The third was a major work of lexicography, an illustrated monolingual Somali dictionary by Yaasiin Cusmaan Keenadiid (1976). In 1978, two Somali scholars, Muuse Xuseen Askar and Aadan Ciise Cali, compiled a dictionary of scientific terms in the field of physics and chemistry, but unfortunately because of lack of funds it still remains unpublished. The 1980s were very fruitful years for Somali lexicography. In c. 1980 a Somali-English dictionary was published by Cabduraxmaan Ciise Oomaar and Saciid Warsame Xirsi, and in 1983 a two volume FrenchSomali dictionary was produced by Maxamed Cabdi Maxamed. In 1985 the publication of a large SomaliItalian dictionary by Agostini et al. was a landmark in the history of Somali lexicography. It was compiled by a 39-strong Somalo-Italian team, working over a period of some seven years, and was sponsored by three institutions: the Somali National University, the Somali Academy of Sciences, Arts and Literature, and the University of Rome ‘La Sapienza’. It had generous financial and logistic support from Italian Government aid funds and from the Somali Government. The team included academic linguists, both Italian and Somali, and traditional experts on oral literature and culture, as well as scanners who examined a large body of published texts in search of words, and who checked the entries in previously published dictionaries, including those in Abraham’s Somali-English Dictionary. In the same decade three more dictionaries were compiled: in 1987 a Somali-French dictionary by Maxamed Cabdi Maxamed; in 1988 an illustrated Somali-English dictionary by Cabdiraxmaan Cabdillaahi Barwaaqo, which still awaits publication; and in 1987 a Somali-English dictionary compiled by Virginia Luling, based to a large extent on Yaasiin Cismaan’s work and of comparable importance. Taken together,
86
B.W.ANDRZEJEWSKI
the dictionaries which were compiled after 1966 contain many more words than were recorded in Abraham’s work. The dictionary compiled by Agostini et al. (1985) has about four times as many entries in its 655 double-column pages. These comparisons show that Abraham’s work has many gaps, but if we take into account the conditions under which he worked we cannot but admire his achievement. We must also observe that the dictionaries compiled after the introduction of the official orthography also have numerous gaps in spite of the intensive lexicographic research which took place. These gaps are not only in the spheres of mathematics, natural sciences and technology, for which a new terminology has been created by Somali scholars and educationalists since 1972, but are even more numerous in the older layers of the language which had existed before Somali became a written language. Anyone who reads Somali novels, short stories, transcripts of oral narratives or collections of poems, or who listens to radio or television programmes on Somali traditional culture or technologies, will readily concur with this observation. The gaps in all the existing dictionaries point to the fact that Somali has very large lexical resources most of which existed long before it became a written language. 10. A postscript: some general observations My experience of using dictionaries of Somali, including Abraham’s, prompts me to offer some general reflections on lexicographical work on unwritten languages or on those which have acquired national orthographies in relatively recent times. It is an incontrovertible fact that it is impossible to compile a near-comprehensive dictionary of a language unless it has a very large body of written texts. In this context, by ‘near-comprehensiveness’ I mean the state of lexicography in such languages as English, French or Italian, when one rarely fails to find an unfamiliar word if one consults dictionaries. Even when there is an abundance of texts, lexicographical research on a massive scale is needed to reach near-comprehensiveness. It is a task far beyond the powers of a single individual and has to be undertaken by a large team working over many years. The history of lexicography in major European languages fully supports this view. From these observations it follows that it is impossible to assess with any degree of accuracy the size of the lexical resources of any unwritten or recently written language in which only a limited amount of texts is available. Claims that a comprehensive or definitive dictionary of any such language has been compiled would be totally misleading. The experience of Somali lexicography throws doubt on the general assumptions of the alleged poverty of unwritten languages which are encountered among the general public of literacy-oriented societies, and even among some scholars. A good example of this is a statement found in the otherwise well-documented book Orality and Literacy: The Technologizing of the Word, by Walter J.Ong (1985), where we find a statement: ‘A simply oral dialect will commonly have resources of only a few thousand words…’ (p. 8). To give scientific validity to a statement of this kind one would have to find several sample communities throughout the world and take a massive amount of tape-recordings from their speakers, covering all circumstances of their lives. The next step would be to employ a group of scanners for each language, who would then register all the words they encounter. I have no knowledge of any experiments of that type having ever been carried out. The realisation of the impossibility of making a near-comprehensive dictionary without research on a massive scale can have a beneficial effect on single-handed and insufficiently supported lexicographers. It can free them from the unjustified sense of inadequacy and irritation, at times amounting to despair, with
REFLECTIONS ON R.C.ABRAHAM’S ‘SOMALI-ENGLISH DICTIONARY’
87
the never-ending inflow of new lexical data even after many years of work. It can also help the users of dictionaries which fall short of the ideal of near-comprehensiveness to cope with their frustration and disappointment. They will no longer be irritated by the gaps they encounter and by the never-ceasing need for filling them by questioning the speakers of the languages concerned. They will then refrain from making adverse comparisons with the dictionaries of major European languages which are the cumulative result of three or four centuries of continuous and extensive lexicographical research. Instead, they may be filled with gratitude and respect for lexicographical pioneers such as Roy Abraham. REFERENCES Since in Somalia surnames are normally not used, names of Somali authors are given in their customary order and are not inverted. The material in square brackets is explanatory and is not found on the title pages of the works concerned. Abraham, R.C. 1964. Somali-English Dictionary. London: University of London Press. ——. 1967. English-Somali Dictionary. London: University of London Press. Agostini, Francesco (main ed.). 1981. Ragguaglio sui lavori del vocabolario somalo-italiano. In Fonologia e Lessico, ed. by Giorgio R.Cardona and Francesco Agostini, pp. 143–148. (Studi Somali 1.) Rome: Ministero degli Affari Esteri —Dipartimento per la Cooperazione allo Sviluppo, Comitato Tecnico Linguistico per l’Università Nazionale Somala. ——, Annarita Puglielli and Ciise Maxamed Siyaad (chief eds.). 1985. Dizionario somalo-italiano. Rome: G.Gangemi. Published under the joint auspices of Jaamacadda Ummadda Soomaaliyeed [Somali National University], Akademiyada Cilmiga Fanka iyo Suugaanta [Academy of Sciences, Arts and Literature] and Università degli Studi di Roma ‘La Sapienza’, with prefaces by Dr Abdisalam Scek Hussen [Minister of Higher Education in Somalia] and On. Giulio Andreotti [Minister of Foreign Affairs, Italy.] Andrzejewski, B.W. 1983. Language reform in Somalia and the modernization of Somali vocabulary. In Language Reform: History and Future/La Réforme des Langues: Histoire et Avenir/Sprachreform: Geschichte und Zukunft, Vol. 1, ed. by Istvàn Fodor and Claude Hagège, pp. 69–84. Hamburg: Helmut Buske Verlag. ——. 1987. In defence of improbable lemmas in Somali lexicography. Afrikanistische Arbeitspapiere 9:103–112. Barwaaqo, C.C. [Cabdiraxmaan Cabdillaahi Barwaaqo]. 1988. Abwaan cusub oo Af-soomaali iyo Af-ingiriisiya. [A modern Somali English dictionary.] Ms., Jeddah, Saudi Arabia. Cabdulqaadir, Faarax Bootaan. 1976. See under Somalia 1976. Cabduraxmaan, Ciise Oomaar, and Saciid Warsame Xirsi. [c. 1980]. Qaamuus Ingiriisi-Somali. [English-Somali Dictionary, published in cyclostyled form and distributed in Mogadishu by the authors.] Caney, John Charles. 1984. The Modernisation of Somali Vocabulary, with Particular Reference to the Period from 1972 to The Present. Hamburg: Helmut Buske Verlag. DeLancey, Mark W., Sheila L.Elliott, December Green, Kenneth J.Menkhaus, Mohammed Haji Moqtar and Peter J.Schraeder (eds.). 1988. Somalia. (World Bibliographical Series, Vol. 92.) Oxford (England), Santa Barbara (California), Denver (Colorado): Clio Press. Lamberti, Marcello. 1986. Somali Language and Literature. (African Linguistic Bibliographies 2), ed. by. Franz Rottland and Reiner Vossen. Hamburg: Helmut Buske Verlag. Larajasse, Evangeliste de. 1897. Somali-English and English-Somali Dictionary. London: Kegan Paul, Trench & Trübner. Luling, Virginia. 1987. Somali-English Dictionary. Wheaton, Maryland: Dunwoody Press. Maxamed, Cabdi Maxamed. 1983. Dictionnaire français-somali: Qaamuus Faransiis-Soomaali. 2 vols., published and distributed by the author, 3 Rue du Languedoc, Apt. 016, 25000, Besançon, France. ——. 1987. Eraybixin Soomaali-Faransiis: Lexique somali-français. Besançon: Imprimerie Stalactite Sucrée [UNESCO-aided publication.]
88
B.W.ANDRZEJEWSKI
Muuse, Xuseen Askar, and Aadan Ciise Cali. 1978. Qaamuuska cusub ee sayniska (Fisigis & Kimistari) [A new dictionary of science (Physics and Chemistry)]. Mogadishu: Unpublished typescript of a monolingual dictionary. Ong, Walter J. 1985. Orality and Literacy: The Technologizing of the Word. London and New York: Methuen. [Reprint of 1982 edition.] Palermo, Giovanni Maria da. 1915. Dizionario somalo-italiano e italiano-somalo. Asmara: Tipografia Francescana. Philibert, C. 1976. Petit lexique somali-français, Paris: Klincksieck. Reinisch, Leo. 1902. Die Somali Sprache, II—Wörterbuch, Somali-Deutsch, Deutsch-Somali. Vienna: Alfred Hölder, Kaiserliche Akademie der Wissenschaften, Südarabische Expedition. Saeed, John Ibrahim. 1987. Somali Reference Grammar. Wheaton, Maryland: Dunwoody Press. Somalia Wasaaradda Hiddaha iyo Taclinta Sare [Ministry of Culture and Higher Education]. 1976. Qaamuuska af Soomaaliga [Dictionary of the Somali language]. Mogadishu: Madbacadda Qaranka. [A monolingual Somali dictionary. Although the corporate author is the Ministry of Culture and Higher Education, the name of the editor and compiler is given in the preface as Cabdulqaadir Faarax Bootaan.] Spitler, A.Keene, and Helen Spitler. 1966. English-Somali Dictionary. Pasadena, California: World-Wide Missions. Stepachenko, D.I., and Mokhamed Hadji Osman [Maxamed Xaaji Cusmaan]. 1969. Kratky somali-russky i russkosomali slovar’/Abwan urursan af Soomaali iyo Rusha, Rushiyo af Soomaaliya [A short Somali-Russian and Russian-Somali dictionary]. Moscow: Izdatel’stvo ‘Sovetskaya Entsiklopediya’. Yaasiin C.Keenadiid. 1976. Qaamuuska Af-Soomaaliga [Dictionary of the Somali language.] Mogadishu: Akademiyaha Dhaqanka, Wasaaradda Hiddaha iyo Tacliinta Sare. [A monolingual dictionary.]
R.C.ABRAHAM AND SOMALI GRAMMAR: TONE, DERIVATIONAL MORPHOLOGY AND INFORMATION STRUCTURE* John Ibrahim Saeed
1. Introduction R.C.Abraham came to the study of Somali towards the end of his illustrious career as an Africanist. Although he is best known amongst Somali scholars for his lexicographical work, he also wrote a substantial body of grammatical description. In particular, in 1951 he published with Solomon Warsama, The Principles of Somali, and his Somali-English Dictionary includes a grammatical outline called ‘The Basis of Somali’ (Abraham 1964:258–332). Both of these works are based on northern Somali dialects. In this paper I will briefly discuss the nature of Abraham’s contributions to the description of the grammar of Somali. The Principles is itself a substantial piece of work: it is 481 pages long and contains 59 chapters. It is not clear to me how the division of labour fell between Warsama and Abraham, but the work is impressive given Abraham’s relatively short exposure to the language: Somali is a language of vast richness and raciness, and much still remains to be done by my successors, for two years of study are insufficient—in the case of some West African languages, I have required twenty five years! (‘Introduction’, p. 2)1 Abraham’s understanding of Somali is even more impressive given that he does not appear ever to have visited the Horn of Africa. More detailed discussion of the content of Principles follows in subsequent sections. A few more general points can be made here. One of the more difficult aspects of trying to view Abraham’s work in its historical perspective is establishing when the works were written, relative to other work in the field. This is a result of the difficulties Abraham met in finding financial support to publish his work. Principles, for example, is a very poorly produced volume, roughly reproduced and bound; the second edition has errata and addenda sheets pasted into many of its pages. This is because Abraham was forced to undertake the whole of the production and publication himself, as his rather bitter note in the introduction makes clear: African Languages and Cultures, Supplement 1 (1992):111–123 *
I would like to thank B.W.Andrzejewski for information about R.C.Abraham’s work and life, and Xuseyn Nuur for reading through the Somali language material in this paper. Thanks are also due to Peter Gurnham, Marie Fowless, and Liz Early in Toxteth, Liverpool for facilities enabling me to write this paper.
90
JOHN IBRAHIM SAEED
The entire work of typing, duplicating and binding has fallen on myself and the method adopted was the only one within my means: it might be thought that as I was Lecturer in Somali at the School of Oriental and African Studies, a grant might have been expected from their Publications Fund— previous bitter experience dissuaded me from seeking a further rebuff. (‘Introduction’, p. 2) His Somali-English Dictionary also seems to have had a chequered publication history, as B.W.Andrzejewski has related to me. The published volume appeared in 1964 by the University of London Press, but appears to have been written some years earlier. The colonial government of British Somaliland originally agreed to publish the work but Somali independence in 1960 intervened. The project lapsed until some creative lobbying of the new minister of education by Muuse Galaal and B.W.Andrzejewski gained Somali Government backing. The result is that this work belongs to the late 1950s rather than the mid-1960s. And once again Abraham seems to have received little support in preparing the work: The present burdensome task of printing—at least five pounds per page for setting up by monotype— has necessitated my typing the whole work for production by photo-offset, a gigantic task on top of the enormous task of compiling the dictionary. If in a few places the print is defective, I must crave the indulgence of the user for an author seventy years of age and suffering from heart trouble. (1964:viii) A second problem for commentators is caused by Abraham’s academic isolation from the work of other scholars in this field, possibly a result of his late entry to it. His Principles does not refer to earlier work by leading scholars in Cushitic and Somali linguistics, even major figures like Reinisch (1900/1903) and Cerulli (1919). In fact he is remarkably candid about his ignorance of some work: I should like to say how closely I am in agreement with the conclusions reached in Somali phonetics by the late Lilias Armstrong; apart from some slight misunderstandings due to lack of knowledge of the language, there are very few points where I disagree with her—my own conclusions were arrived at independently, and I was not aware of her work until one of my students called my attention thereto. (‘Introduction’, p. 2) We might note that Armstrong’s paper ‘The phonetic structure of Somali’, even now the outstanding study of Somali phonetics, was published in 1934. We can compare this, for example, with Martino Moreno’s (1955) work on southern Somali dialects, Il Somalo della Somalia. Moreno was a contemporary of Abraham’s: he started work on Somali in Rome in 1949, later visiting Somalia (1955:v). His work is centrally positioned within the Cushitic linguistic tradition, and securely based on reviews of the earlier literature. His footnotes are full of detailed comparisons of his own data and analyses with work by others on various Somali dialects. Interestingly, Il Somalo, finished in 1953, discusses details of Abraham’s analyses in Principles (see for example Moreno 1955:5, fn. 3). These delays and difficulties with publication, and his lack of discussion of other work, mean that in many cases Abraham’s work is more pioneering than its publication dates suggest and often very original, even if in some cases he is retracing ground covered by other scholars.
1
The ‘Introduction’ in Warsama & Abraham (1951) is not paginated. I have therefore for convenience used page numbers (p. 1 etc.).
R.C.ABRAHAM AND SOMALI GRAMMAR
91
Rather than review the works in great detail, I will pick out three areas where Abraham’s contribution to the study of Somali grammar seems to me to be particularly significant. These are the issues of the role of tone, the description of derivational morphology, and the marking of information structure—focus and topic —in syntax; all centrally important areas of the grammar. Before that, though, as a reader, I must note that one difficulty in using Abraham’s work is the density and fragmentation of its presentation. It seems clear that Principles, for example, was intended as a teaching grammar rather than a reference grammar: …from my experience with my pupils, a student who works conscientiously through this book has nothing to fear from any construction. (‘Introduction’, p. 1) It is tempting to add that any student who could work through the book would have nothing to fear from any intellectual exercise. There may have been exceptionally gifted and durable students in the 1940s, but the task facing one trying to ‘work through’ Principles is a daunting one. There are no exercises or texts, and the organisation of the many sections is hard to fathom, being arranged neither according to grammatical levels and categories, nor to any pedagogical principle of simple to complex structures. Despite some descriptive shortcomings, C.R.V.Bell’s (1953) The Somali Language is a far better teaching guide to Somali from roughly the same period. By comparison, Abraham’s later work ‘The Basis of Somali’, being a short reference grammar, is much clearer in structure and usable. I can speak as a satisfied customer: I have used it for years and still refer to it for one of Abraham’s strongest points: his excellent use of examples. Despite the difficulty of his work, I hope to show that Abraham displays in crucial areas a great insight into the Somali language and even when his descriptive acuity occasionally fails him, for example in the transcription of vowel and consonant length, or tone marking,2 his ideas are always interesting and often, it seems to this reader, correct. 2. Tone One important area where Abraham is on the side of the angels is the question of the role of tone in Somali, and whether Somali can be called a tone language. The behaviour of tone in relation to accent or stress has been a difficult issue in Somali grammatical description. Lilias Armstrong (1934) had noted that in northern Somali word tone seemed significant (in a complicated relationship with stress), giving pairs like (1) and (2) below (using Armstrong’s own notation for tone): (1) (2)
2
a. b. a. b.
gees\ gees • i • nan • i • nan •
HIGH FALLING TONE MID TONE HIGH-LOW MID-HIGH
‘horn’ ‘direction’ ‘boy’ ‘girl’
Abraham’s work predated the adoption of an official Somali orthography, and his own transcription changed over the years. For the convenience of readers not familiar with Somali, I have taken the liberty of transliterating all examples he used into the official Somali orthography, except where noted. Page references are given to allow the reader to check the various original transcriptions. For details of the official orthography, see Saeed (1987:13–26). In addition the following tone marks are used: á=HIGH TONE, à=HIGH FALLING, a (unmarked)=LOW
92
JOHN IBRAHIM SAEED
We might prefer to identify MID as a positional variant of LOW tone but the essential observation is of course correct. However, tonal differences are mainly put to grammatical work in Somali, marking, for example, gender, number and case in nouns, and individual inflectional paradigms in verbs. So, for example, the (a) examples above are masculine in gender and the (b) examples feminine, while (3) below shows the marking of number by tone: (3)
a. b.
èy éy
HIGH FALLING HIGH
‘a dog’ ‘dogs’
Lexical differentiation by tone can be seen as the minor case: an offshoot of the marking of grammatical information by tone. A full working out of the role of tone in Somali had to await the work of Andrzejewski (1964, 1968, 1979). The system is briefly outlined in Saeed (1987). In the 1940s and 1950s the issue was far from clear. Moreno (1955) notes Armstrong’s findings for northern Somali dialects but says, basically, that the jury was still out on tone in southern dialects—the jury being Andrzejewski, then researching the area: Occorrerà attendere i risultati delle ricerche fatte dall’Andrzeyewsky in Somalia per conoscere la e digil. (Moreno 1955:22) funzione del tono nel benādir, It will be necessary to await the results of the research done by Andrzejewski in Somalia to understand the function of tone in Benaadir, Daarood and Digil. Bell (1953) is misleading on the subject: Somali is not a tone language in the accepted sense of the term, that is words are not normally differentiated only by tone. However, the tone, that is the rise and fall of the voice in speaking is of importance just as it is in English. The sentence ‘He’s dead’ may mean (a) ‘He is dead,’ (b) ‘I am afraid he is dead,’ (c) ‘Is he dead?’ or (d) ‘Surely he’s not dead?’. The only difference is the tone of the voice. In Somali you should notice the rise and fall of the voice in the principal kinds of sentence. (1953:9) This comparison with English is unfortunately wrong: it is not possible to form a question from a statement in Somali by just changing the intonation. The role of tone in conveying grammatical information means that as a resource it is tied up: sentence-types are differentiated by specific morphemes identifying questions, various types of imperatives, optatives etc. (see Saeed (1987:69ff.) for details). Bell continues, even more confusingly: When words are the same but are different in meaning they are (in theory at least) distinguishable by tone, e.g.: (a) Common nouns inan ‘boy’, inan ‘girl’, nirig ‘baby he-camel’, nirig ‘baby she-camel’. (1953:9) Note that there is no tone marking on the above examples to help the student. In fact neither tone nor stress is marked on any of Bell’s examples. In contrast, Abraham is very sure on the importance of tone in Somali grammar. In Principles, Chapter 9 is entitled ‘Word-tone and semantical musical tone of Somali nouns’, and this chapter goes on to prefigure
R.C.ABRAHAM AND SOMALI GRAMMAR
93
in embryonic form the findings of Andrzejewski (1964) on the relationship between tone and noun declension classes. Some of the detail of his tone transcriptions in this chapter might be amended, and he identifies an extra tone, RISING, which subsequent work has not confirmed: most accounts use just HIGH, LOW and (HIGH) FALLING tones. Nonetheless, the chapter is full of interesting suggestions. For example, it has been proposed by several writers, e.g. Saeed (1987:21), that since the HIGH FALLING tone only occurs on long vowels and diphthongs, and specifically where we would expect the sequence HIGH-LOW on bisyllables, that it is possible to analyse Somali with just two underlying tones, HIGH and LOW, by identifying long vowels and diphthongs as two tone units. Going through Principles for this paper, I was interested to find Abraham making the same point, characteristically in a single parenthetic line: Monosyllables have falling tone in the singular—a falling tone being in many languages, a compound of a high tone coalescing with a low tone. (1951:46) By the time of the Dictionary, Abraham is forcefully emphasising the role of tone. The full title of the grammatical section is ‘The basis of Somali: grammar, stress and tone’, and there is a clear signal of intent in the Introduction: It has been contended that Somali is not a tonal language; in actual fact, the tonal changes given to words which are the same in form, but different in meaning, are the very background of Somali. Unless we rigidly adhere to tone, meaning vanishes and the rhythm becomes distorted to an extent where the language is no longer itself. (1964:vii) All of the language material in both Principles and the Dictionary is tone marked—an essential convention more honoured in the breach by subsequent work. In the Dictionary Abraham reiterates the importance of tone in both nominal and verbal morphology, using examples like (4) below: (4)
shúqulkan sámèe shúqulkan samée
‘do this work!’ ‘he did this work’ (1964:331)
where he is identifying a tonal contrast between the verb forms for the imperative and the independent past tense. We can speculate that it was his background in West African languages that helped Abraham be so clear on the importance of tone in Somali. If so, this is one of the areas where his position as a latecomer to Cushitic linguistics was of benefit to him and the field. 3. Derivational affixes A second area of the grammar where Abraham’s work is important is that of verbal and nominal derivational affixes. Somali has a complex system of affixes by which, for example, the argument structure of verbs may be altered along various semantic parameters. So, to use some of Abraham’s own examples, we find pairs like in (5) and (6), where a causative verb is formed by adding an affix to a base verb: (5)
a.
wuu gelayaa
‘he is entering’
94
JOHN IBRAHIM SAEED
(6)
b. a. b.
wuu gelinayaa wuu qadeynayaa wuu i qadeysiinayaa
‘he is causing (him) to enter’ (1951:334) ‘he is having lunch’ ‘he is giving me lunch’ (1951:335)
or pairs like (7) and (8) below, where an affix causes an autobenefactive interpretation—what Warsama and Abraham (1951) call ‘middle verbs’: (7) (8)
a. b. a. b.
wuu dhisayaa wuu dhisanayaa wuu hagaa jinayaa wuu hagaa jisanayaa
‘he is building (it)’ ‘he is building (it) for himself’ (1951:342) ‘he is arranging (it)’ ‘he is arranging (it) for himself (1951:345)
or a pair like (9a-b), where a change of affix produces a passive interpretation: (9)
a. b.
wuu hubsanayaa ‘he is investigating (it)’ wuu hubsoobayaa ‘he/it is being investigated’ (Warsama & Abraham 1951:360; Abraham 1964:287)
Given both his activities as a lexicographer and his background in Hausa studies, it is perhaps not surprising that Abraham would be sensitive to verbal derivational affixes, and he includes in his Principles several chapters of analysis, with a large number of examples, of causativising, transitivising, passivising and autobenefactive (or ‘middle’) affixes. Some of the detail of his analysis has inevitably been changed by later work, but his coverage prefigures an important line of subsequent investigation, including notably Andrzejewski (1968) and Puglielli (1984). Abraham also dealt with nominal derivational affixes in great detail. Chapter 43 of Principles, for example, is called ‘Noun Terminations’ (pp. 311ff.) and lists around 25 types of derivational affix, each with subtypes and a profusion of examples, showing how nouns can be formed from other syntactic categories, and from other nouns. The following are a few selections with a couple of the many examples originally given: The suffix -ad. This termination -ad forms the female of certain animals. Examples: doofaar ‘the wild boar’, doofaarad ‘sow’, baqal ‘the he-mule’, baqlad ‘the she-mule’. (p. 317) The suffix -asho. Nouns in -asho (always feminine) are formed from Conjugation 2 verbs which end in -anayaa, by substituting -sho for -tay of 2nd person singular past tense. Examples: wuu dhimanaya ‘he is dying’, way dhimatay ‘she died’, dhimasho ‘death’, wuu dhunkanayaa ‘he is kissing’, dhunkasho ‘a kiss’. (p. 322) The suffix -niin. The root of a Conjugation 1 verb+niin forms a verbal noun, in some cases masculine, in other cases, feminine. If the verb is transitive, these verbal nouns are so too. Examples of masculines: wuu duubayaa ‘is folding’, duubniin-kii (i) ‘the act of folding’, (ii) ‘the crease, fold’; wuu barayaa ‘is teaching’, barniin-kii ‘the teaching’. (p. 330) The suffix -e. This -e denotes the male doer of an act…xabaal> xabaashii ‘the grave’…xabaalqode ‘grave-digger’ (wuu qodayaa ‘is digging’). (p. 323)
R.C.ABRAHAM AND SOMALI GRAMMAR
95
The complexity and elegance of the system of lexical derivational affixes is one of the striking features of Somali grammar, and the detail given to this area in Abraham’s work is greater than in other work of the time. Bell (1953) uses verbal derivational affixes to set up verb conjugation (as have done most writers since), but his coverage is limited and he has little to say on noun derivation. Moreno (1955) covers both verbal and nominal affixes but not nearly in the detail of Abraham’s work. It has to be said, though, that Moreno’s analyses are often more elegant and more clearly presented. His ear, too, is sharper for the morpho-phonological structure of his data. For work giving as much attention to derivational affixes as Abraham’s we have to wait until Andrzejewski (1968) for verbal affixes, and Puglielli (1984) for nominal affixes—the latter being one of the by-products of the large Italo-Somali dictionary project in Rome and Mogadishu (see Agostini et al. 1985). This remains one of the most interesting areas of Somali grammar from a comparative and typological perspective. 4. Information structure The third area of Abraham’s work I would like to discuss here is information structure, by which I mean the grammatical packaging of an utterance to fit the context of a discourse. This is the level at which terms like given and new, theme and rheme, focus, topic and comment are often applied in the literature. For the Somali grammarian the importance of an uttered sentence’s information structure is clear very early on: Somali has a number of lexically empty morphemes whose role it is to mark focused elements, for example noun phrases. Moreover, we very regularly find sentences with extra NP arguments outside the predication, acting as topics or afterthought topics. We will see examples of both of these features shortly. It has emerged that there are two problems facing the linguist in this area of Somali grammar: firstly to recognise and outline the use of the basic system; and then, equally difficult, to describe the morphosyntactic complexities associated with, for example, the focus markers. For, most likely due to their historical development, focus markers are not simply neutral elements dropped into sentences but impact on a whole range of syntactic relations: agreement, case marking, the distribution of clitic pronouns, word order, etc. I have tried to describe some of this behaviour in Saeed (1984). Moreno (1955) provides a short sketch of the NP focus markers baa and (a) yaa, the cleft structures introduced by waxa, and the morpheme waa, often identified as a verb focus marker—which he takes to be a preverbal particle. He notes that the speaker of Somali has to use these markers of information structure in declarative sentences: sono indispensabili…quando il compiuto o l’incompiuto e usato in proposizione wā, yā, bā, principale affermativa. Esse sono le «particelle staffette», come io le chiamo, del verbo principale affermativo. (1955:259) waa, yaa, baa and waxa are indispensable when the complete or incomplete is used in an affirmative main clause. These are the ‘message-carrier particles’, as I call them, of the main affirmative verb. Bell has a fuller treatment. He recognises the function of these morphemes and stresses their importance in speaking Somali, for example:
96
JOHN IBRAHIM SAEED
Baa draws attention to a noun, and waa draws attention to a verb. If you use baa you know what the verb is likely to tell you but the noun (subject or object) is something freshly introduced to your notice: in an extreme case it will cause you surprise. If you use waa you assume a knowledge of the noun, probably because it had already been mentioned previously in the conversation, and what happened will be unknown to you, and in an extreme case, a surprise, e.g.: (i) Baa: The dog it was that died (but the man recovered)’ Eygii baa dhintay, ‘Ali (but not Ahmed or Farah) will go with me’ ‘Ali baa i raa‘eya; (ii) Waa: ‘The man (you expected) has come’ Ninkii waa yimi. (1953:25–26) Bell, it seems to me, does a good job of briefly introducing English speakers to the use of these morphemes; and he covers one of the main morphosyntactic irregularities associated with NP focus: that focused subjects show a reduced concord pattern with an agreeing verb. Abraham agrees with the main point: The word ba’ is the most important word in Somali and it is no exaggeration to say that if its usage is not understood, most Somali sentences cannot be explained…ba’ means ‘is the one who’ and grammatically emphasises the word after which it stands.3 (1951:72). And he goes on to point out the major concord irregularity: A pecularity of this word ba’ (which also applies to ya’ when the latter is synonymous with ba’) is that when preceded by a plural noun, ba’ (or ya’) has the meaning ‘are the group who is’. ‘Group’ here requires a singular verb to follow (not plural verb), such singular verb being in the masculine singular if the noun in the plural (subject) is grammatically masculine. Similarly, the verb is in the feminine singular after a noun-subject which is grammatically feminine. (p. 73) This prefigures Andrzejewski’s (1968, 1975) identification of restrictive verbal paradigms, though interestingly, none of these earlier authors detects the tonal difference between normal verb forms and those showing agreement with focused subjects, as described in Andrzejewski (1968). There are several significant areas where Abraham’s account of information structure in Principles is more detailed and interesting than other contemporary work. We can take, as examples, his treatment of the distribution of subject clitic pronouns, of word order and of topic structures with waa. I will look at these very briefly. In Principles Abraham investigates the effect of NP focus on clitic pronouns. When the subject NP is not in focus, a coreferential clitic pronoun may occur: (10)
Carab wuu tegay Arabi waa+uu/hei went
‘an Arab man went’ (1951:75)
where the subject clitic pronoun ‘uu ‘he’ is coreferential with the NP Carab ‘an Arab man.’ Similarly in a sentence with a non-subject NP in focus we find examples like (11) where capital letters in the translation show focus: (11)
ninkii faraskii buu dilayaa man+the horse+the baa+he beating ‘the man is beating THE HORSE’
R.C.ABRAHAM AND SOMALI GRAMMAR
97
However, as Warsama and Abraham point out, if the subject NP is in focus, no coreferential clitic may occur: …faraskii buu dilayaa ‘he is beating the horse’. *This buu consists of the emphatic ba’ (with elided -a) +-uu meaning ‘he’ [JIS=subject pronoun of the series shown in (8b)]. **The emphatic ba’…may only stand after a grammatical subject, whatever may be the logical object… Hence, ba’ after a noun=‘is the one who…’, buu after a noun=‘is the one whom’. (p. 78) Abraham is here correctly identifying the odd subject clitic distribution and, it seems to me, proposing the plausible hypothesis that the clitic pronouns are providing a back-up system for identifying the grammatical relations of the sentence. This is even more plausible given the identification in later work (e.g. Andrzejewski 1979) of a blocking of nominative case in focused NPs. Abraham is also very bold, if less successful, in trying to explain the use of focus on subjects in terms of a difference between what he calls ‘logical’ and ‘grammatical’ subjects, making, it seems, a parallel between sentences like (12) and English passives like (13): (12)
(13)
faraskii buu dilayaa horse+the baa+he beating ‘he is beating THE HORSE=it’s THE HORSE he’s beating’ the horse is being beaten by him
For Abraham the NP faraskii ‘the horse’ in (12) is logically the object but ‘grammatically’ the subject (1951:78). This parallel between English passives and the use of baa on subjects is quite revealing at the pragmatic level —and it starts the reader/learner thinking about information or thematic structure strategies —but at the morphosyntactic level faraskii is neither nominative in case as Abraham wrongly suggests, nor acts like a subject in terms of even the reduced verbal concord we find in focus sentences. Nonetheless, this suggestion, like many of his observations on NP focus, is original and serves as a flag or warning to English speakers that this area of Somali is quite unlike their own language. Abraham also recognised the influence of information structure on word order and he attempts in Principles to map out the complicated interrelationship between focus, clitic pronouns and word order in a way that prefigures recent work by Lucyna Gebert (1986). Abraham uses a variation of slot and filler tables to map out the possibilities, e.g.: Noun subject+pronoun-object+verb: noun-subject ba’ short pronoun object verb… noun-subject long pronoun buu short pronoun verb… noun-subject ba’ noun-object verb… noun-object noun-subject ba’ verb… noun-subject noun-object buu verb… noun-object buu noun-subject verb (1951:81–82)
3
Note that Abraham consistently transcribes baa and (a) yaa with short vowels.
98
JOHN IBRAHIM SAEED
This is just a tiny excerpt from his tables: in each case he provides many example sentences and discusses, though typically in a very epigrammatic way, the frequency of the various orders. In this, he was beginning work that still has to be comprehensively followed up—on the use of information structures and their consequent word orders in discourse. Finally in this area, Abraham’s work is innovative in identifying topic structures in Somali. As can perhaps already be seen from the brief examples discussed here, this is probably because of his willingness to provide semantic and pragmatic explanations at any level of analysis. He does not seem to have any of the desire to rigorously separate levels of analysis that we associate with some of his contemporaries and successors, under the influence we might assume of American structuralist-descriptive linguistics. So we continually find in his work not only tables of forms but, sprinkled around, suggestions of the meaning and use of forms. In retrospect these suggestions are often more valuable than his morpho-phonological analyses. Thus, we find suggestions throughout Principles that sentences with waa should be seen as topic structures, for example: in ninkii wuu furayaa…and naagtii way furaysaa, wuu means ‘it is he who’ and way means ‘it is she who’, so the sentences literally mean ‘the man —it is he who is opening’ and ‘the woman—it is she who is opening’, respectively. In other words, the subject ‘he’ is indicated by the -uu, and the subject ‘she’ is indicated by the -ay which is tacked on to way. (1951:73) Abraham’s identification here of the clitic pronouns as arguments of the predication, and of the sentenceexternal status of the coreferential full NPs is one early contribution to the debate about ‘double’ subjects and objects in Somali, and parallels my own conclusions in Saeed (1984). Abraham also seems to propose the label ‘accusative of reference’ for topic elements and to exclude them from the argument structure of the main predicator; see for example the discussions of comparatives in: naagta+ninkii waa ka wanaagsanyahay ‘the man is better than that woman’… In all the above sentences, the second part (i.e. what follows+) is intransitive and the first part (i.e. what precedes+) is the accusative of reference (‘as to that woman’). (1951:275) It seems that the proposal for the above example, repeated again in (14a) with a gloss, is a topic structure like that shown in (14b): (14)
a. b.
naagtaa ninkii waa ka wanaagsanyahay woman+that man+the waa more good+is [TOPIC that woman], the man is better
This proposal occurs again in Abraham (1964): L. ACCUSATIVE OF REFERENCE A noun is often in the accusative in the sense of ‘with reference to’. Examples: birtan waxaa lugu qodbay, sibidhka (‘as to this metal, the place as to which people have embedded it in, is the cement’) ‘this metal is embedded in cement’; rasiidhka magacaaga geesta kale kaga qor (‘in regard to the receipt, write your name on the other side’), ‘write your name on the other side of the cheque!’, ‘endorse the cheque!’. (1964:311)
R.C.ABRAHAM AND SOMALI GRAMMAR
99
Once again, Abraham’s interpretation of the literal meaning of Somali sentences is very revealing. So we can see that in this crucial area of information or thematic structure, Abraham’s analyses, though somewhat epigrammatic, concern themselves with both sides of the problem: with the use of different sentence-types and with the grammatical complexities of focus and topic constructions. In investigating word order, the role of clitic pronouns, and the evidence for topics, his work picks out some of the most difficult issues in Somali syntax and foreshadows some seminal work by Andrzejewski in particular. In some cases his work highlights areas which still await description: in particular we still await discourse-based studies of the use of these different information structures. 5. Conclusion As was claimed in the introduction, Abraham as a grammarian of Somali is an interesting figure. Compared with his contemporaries, he is probably less successful as a writer of a reference grammar than Moreno, and less successful as a writer of an introductory teaching grammar than Bell. Nevertheless he is often less cautious and more insightful about syntax than either. In both Principles and the ‘Basis of Somali’ section of his Dictionary, the shorter of his descriptions, his work is full of bold suggestions, based on an enviably large amount of data. When we think of the circumstances of his research: as an elderly man, often unwell, working with Somali speakers in London, we can perhaps only wonder at the scale of his achievement in compiling a major dictionary and two works of grammatical description. His glosses and translations are still a valuable aid to the syntactic structure of the language. Moreover his attention seems to have been caught by some of the most difficult and interesting areas of the language: the derivational morphology, the problem of focus markers and topic structures, and the role of tone. In each of these areas his work prefigures important lines of research by his successors. Besides, who could resist his grammarian’s personal enthusiasm, as evidenced in his introduction to the Principles volume: I shall always be glad to give personal help to those interested in the study of this absorbing subject. My telephone number is Hendon 2297. (1951, ‘Introduction’, p. 2) REFERENCES Abraham, R.C. 1964. Somali-English Dictionary. London: University of London Press. Agostini, Francesco, Annarita Puglielli and Ciise Maxamed Siyaad (chief eds.). 1985. Dizionario somalo-italiano. Rome: G.Gangemi. Andrzejewski, B.W. 1964. The Declension of Somali Nouns. London: School of Oriental and African Studies. ——. 1968. Inflectional characteristics of the so-called weak verbs in Somali. African Language Studies 9:1–51. ——. 1975. The role of indicator particles in Somali. Afroasiatic Linguistics 6: 123–191. 1979. The case system in Somali. Ms., School of Oriental and African Studies. Armstrong, Lilias E. 1934. The phonetic structure of Somali. Mitteilungen des Seminars für Orientalische Sprachen zu Berlin 37(3):116–161. [Reprinted in 1964 by Gregg Press, East Ridgewood, New Jersey.] Bell, C.R.V. 1953. The Somali Language. London: Longmans, Green & Co. [Reprinted in 1968 by Gregg Press International, Farnborough.] Cerulli, Enrico. 1919. Nota sui dialetti somali. Rivista degli Studi Orientali 7:861– 876. Gebert, Lucyna. 1986. Focus and word order in Somali. Afrikanistische Arbeitspapiere (Cologne) 6:43–69. Klingenheben, August. 1949. Ist das Somali eine Ton-sprache? Phonetik III, fasc. 5(5):289–303. Moreno, Martino Mario. 1955. Il Somalo della Somalia. Rome: Istituto poligrafico dello stato.
100
JOHN IBRAHIM SAEED
Puglielli, Annarita (ed.). 1984. Aspetti morphologici, lessica e della focalizzazione (Studi somali 5). Rome: Ministerio degli afari esteri. Reinisch, Leo. 1900–1903. Die Somali Sprache (3 vols.). Vienna: Alfred Hölder, Kaiserliche Akademie der Wissenschaften, Südarabische Expedition. Saeed, John Ibrahim. 1984. The Syntax of Focus and Topic in Somali. Hamburg: Helmut Buske. ——. 1987. Somali Reference Grammar. Wheaton, Maryland: Dunwoody Press. Warsama, Solomon, and Major R.C.Abraham. 1951. The Principles of Somali. Published by the second co-author.
THE CONTRIBUTION OF R.C.ABRAHAM TO IDOMA STUDIES* Shamsudeen O.O.Amali
1. The Idoma-speaking peoples The Idoma-speaking peoples inhabit Benue, Plateau, Cross River and Anambra States of Nigeria, with the majority living in Benue State. There are also many Idoma speakers in other parts of Nigeria and the total population runs into millions.1 In Nigeria, Idoma-speaking people occupy very important positions in local, state and national life, and there are very many Idoma university graduates, administrators, businessmen and women. There are many primary and secondary schools in Idomaland, most of them established by Methodist and Roman Catholic missionary groups. It is only recently that governments, individuals and Islamic organisations have begun to establish primary and secondary schools. 2. A brief history of the study of Idoma Before the arrival of R.C.Abraham in Idoma in 1933 there had been little study of Idoma language and culture, and in order to appreciate Abraham’s contribution, it is essential to trace briefly the history of earlier studies. According to R.G.Armstrong: The first bit of writing in Idoma appears in Clarke’s Specimens of Dialects (1848) pp. 10–11, No. 149, mislabelled ‘Gold Coast’. It is probably from the Western (Idoma) dialect, and it is written impressionistically; but it is unmistakably Idoma for all that. Professor Ardener thinks that Wm. Scott, a sawyer and a settler in Fernando Po ‘from Doma county’, whom Clarke knew in 1841, was responsible for this short vocabulary of nine words. (1985:1)
Armstrong adds:
African Languages and Cultures, Supplement 1 (1992):125–129 *
I must sincerely and warmly thank the Centre for African Studies, SOAS, University of London, and Dr Philip Jaggar for organising and sponsoring the R.C.Abraham Symposium. May the Almighty Allah reward you all and bless the soul of R.C.Abraham. Finally, I bring you very warm greetings and gratitude from the Idoma-speaking peoples.
102
SHAMSUDEEN O.O.AMALI
The first really substantial account of an Idoma dialect was printed in Koelle’s Polyglotta Africana (1854), which gives almost 300 words of Yala of Ogaja. Koelle was a very fine scholar but not a linguist or a phonetician in the modern sense. His writing of Iyala is impressionistic and the Ph.D thesis on Yala by Bunkowske (1976) shows that the orthographic problems of Yala are quite different from those of central Idoma… In 1854, Samuel Ajayi Crowther accompanied Dr Baikie on his voyage up the Benue to Garoua and collected a list of Idoma words in the pattern of Koelle’s Polyglotta Africana. These were published in Crowther’s Journal of an Expedition up the Niger and Tshadda Rivers in 1854. (1985:1–2) In the early 1920s, following the construction of the Nigerian railway through Idomaland by the British colonial administration for the transportation of human beings, coal, groundnuts, palm oil and other products, a Methodist Mission and School were established at Igwumale, and Mr Norcross produced a translation of the Gospel of Mark and the Acts of the Apostles as well as a Primer and hymnal in the Igwumale or Southern Idoma dialect. These books were widely used in Protestant schools throughout Idoma for several decades. Norcross was not a linguist, however, and his spelling was impressionistic. As far as possible he followed the current Igbo practice, writing [ε] as ẹ, [ ] as ọ, [ŋ] as , and [gb] as . Idoma words in general were spelled on the English model, and the citation forms of words and morphemes were written as separate words, without reference to the elaborate system of Idoma word-compounds. Armstrong further states that: The Catholic Mission produced much less, but chose to use the ‘phonetic’ ε and instead of ẹ and ọ. This introduced a very unfortunate sectarian aspect into the already heated debates over orthography. Many people came to regard the letter-forms ẹ and ọ and ε and as symbolic of Methodism and Roman Catholicism respectively. In 1923, Mr Judd, a missionary, published a short description of the Keana or Eastern Arago dialect of Idoma in the Journal of the African Society. Once again, this orthography was impressionistic, and the dialect, lying in Lafia at the extreme north eastern end of the Idoma distribution, cannot be taken as typical for present-day orthographic purpose in central Idoma. (pp. 2–3) 3. R.C.Abraham and Idoma In 1935 Captain (later Major and Dr) R.C.Abraham came to Idoma as an Assistant Divisional Officer (ADO). R.C.Abraham was an Administrative Officer who was very different from the other British colonial Administrative Officers who had been previously sent to Idoma.2 He combined his basic administrative duties with the study of the Idoma language and culture. His studies laid the solid foundation for the promotion of the Idoma language, literature and culture throughout Idomaland, Nigeria and the world. While Abraham laid the foundation for contemporary Idoma studies, Armstrong further developed these studies through his numerous publications. They were the founding fathers of Idoma studies, and giants in
1
I was informed by a retired Nigerian Army Officer of Idoma origin who served in a Peace Keeping Force in the West Indies sometime ago, that he came in contact with a group of people called ‘the Bush Negroes’ in Surinam, who still retained some traces of Idoma in their language. I would like to go there to investigate this claim.
2
Abraham’s cook-steward, Mr Omaaba Iwo is still alive and lives in Otukpo, Benue State.
THE CONTRIBUTION OF R.C.ABRAHAM TO IDOMA STUDIES
103
the field. According to Armstrong, who could be regarded as a strong admirer and a scholar of Abraham’s work on Idoma: R.C.Abraham, then an A.D.O., prepared a careful, cyclostyled grammar of the Otukpo dialect of Idoma. Only twenty copies of this work were produced; but it is bound and is 429 pages long and contains a number of texts. It gives complete and very careful attention to tone throughout, and has a remarkable and detailed description of the complicated Idoma system of forms in reported speech, or indirect discourse. It is not too much to say that it is the first usable grammar of an Eastern Kwa language and it is entirely comparable with the much better-known grammars of Twi and Ewe by Christaller and Westermann respectively. A second and much revised edition was published in 1950 for the Idoma N.A. and in a photo-offset printing of Dr Abraham’s typescript. Only 114 copies were printed. This book contains extensive wordlists, hundreds of proverbs and a chrestomathy (collection of texts), in addition to the grammar. The book was republished in 1967 by the University of London Press… Abraham’s orthography is essentially linguistic in the sense that it is completely self-consistent and also consistent in every sentence with his grammatical analysis. He used the letters ε, and ŋ from the International Phonetic Alphabet. His marking of tones followed the pattern one sees in his and Bargery’s Hausa Dictionaries and his Tiv works, in that the marks are placed beneath the letters. For example, is low tone, ạ is mid tone and high tone is left unmarked. This system was developed by Rev. Bargery for his Hausa Dictionary. Most linguists regard it as usable but awkward and have preferred to put the tone marks above the letters. There are advantages and disadvantages to both systems, but in Idoma, it is economical to mark the high tone and leave the low tone unmarked. One thereby saves 20 percent of the diacritic marks that would otherwise be necessary. The mid tone is the least common in Idoma and its omission would be the least economical system. (pp. 3–4) I have quoted Armstrong’s observations on Abraham extensively because they are accurate and relevant. Armstrong and I also did follow-up studies on most of these observations and remarks. Abraham’s descriptions and analyses of Idoma nouns, numerals, verbs, elision and linkage, pronouns, verbal suffixes, relative negative, ‘where’ and ‘when’, ‘motion’ and ‘rest’, imperative and subjunctive, reported speech, emphatic sentences, ‘before’ and ‘after’, agential formations, verbal nouns and prepositions, are linguistically and culturally precise, accurate, and essential to readers of his The Idoma Language, Idoma Word-lists, Idoma Chrestomathy, Idoma Proverbs (1951). Similarly his wordlists, proverbs, and chrestomathy are grammatically useful to Idoma and non-Idoma students alike. Abraham had originally intended that his study of Idoma should be used by colonial administrators working in Idomaland, but the work is also very useful to native Idoma-speakers. Abraham used the bilingual method to record oral texts in Idoma and English, and also introduced the literary and textual study of Idoma oral proverbs and culture. Incredibly, he had the ear, tongue and sensitivity of an Idoma, as well as the gift of linguistic analysis. This writer, learning from the traditions of Abraham and Armstrong, introduced bilingual literary creativity in Idoma and English (see Bibliography). The present-day study of Idoma needs the massive introduction of Abraham’s The Idoma Language, for use in Idoma and non-Idoma schools and homes. Many Idoma students and schools are in dire need of the book, but it is not available. I suggest very strongly that the book should be republished. Before it is republished, however, some corrections need to be done. Idoma primary readers are also being prepared for Idoma primary and secondary schools, based upon Abraham’s and Armstrong’s Idoma language foundations. Idoma is taught at the College of Education, Katsina-Ala, in Benue State, and I have used Abraham’s The Idoma Language with undergraduates in the Department of Languages and Linguistics of
104
SHAMSUDEEN O.O.AMALI
the University of Jos, Plateau State, Nigeria. I am also continuing work on Armstrong’s Idoma Dictionary which has its foundation in Abraham’s The Idoma Language. There is a very strong drive for Idoma to be extensively used in all Idoma Schools of Benue State. Many Idoma writers and scholars such as Onka Oblete, Ismaila Amali, Idris Amali, Usman Amali, and Stephen Obeya are using Abraham’s and Armstrong’s methods of writing Idoma. They wrote Idoma as it is spoken, fluently and naturally. I would like to conclude this paper by quoting an Idoma proverb which Abraham himself recorded: biŋ ‘the river dried up (but) its name was not lost’. One’s deeds live on—R.C.Abraham is dead but his Idoma language and cultural works live on. BIBLIOGRAPHY Abraham, R.C. 1935. The Principles of Idoma. London: Published by the author through Crown Agents for the Colonies. ——. 1951. The Idoma Language, Idoma Word-lists, Idoma Chrestomathy, Idoma Proverbs. Nigeria: Published by the author on behalf of the Idoma Native Administration. Amali, Idris O.O. 1984. Form and meaning of Idoma-Otukpo proverbs. MA thesis, University of Jos. ——. 1991. The beginning and growth of written literatures on Idoma language, literature and culture. Paper presented at the 9th Annual Conference of the Folklore Society of Nigeria, University of Ibadan, April 1991. ——. Idoma Bibliography (1847–1987). Jos: National Council for Arts and Culture. In press. Amali, S.O.O. 1970. Worlds within Worlds and other Poems. Ibadan: University of Ibadan. ——. 1972. Onugbo Mloko. Ibadan: Institute of African Studies. ——. 1981. Idoma History and Oral Literature Texts. (Jos Oral History and Literature Texts.) Jos: University of Jos. ——. 1983. The Nigerian Dreams and Realities. (Occasional Publication No. 1.) Jos: University of Jos, Department of Theatre Arts. 1985. An Ancient Nigerian Drama. Germany: Leo Frobenius Institute. Armstrong, Robert G. 1963. The verb in Idoma. Actes du Second Colloque Internationale de Linguistique NégroAfricaine, pp. 127–157. Dakar: Université de Dakar and WALS. ——. 1968. Yala (Ikom), a terraced-level language with three tones. The Journal of West African Languages V(1): 49–58. ——. 1969. An Ancient Music of The Idoma. New York: Asch. ——. 1983. The Idomoid languages of the Benue and Cross River Valleys. The Journal of West African Languages XIII (1):91–149. ——. 1985. Idoma orthography. In Orthographies of Nigerian Languages, Manual III, 1986, ed. by Ayọ Banjọ. Lagos: National Language Centre. ——. Idoma Dictionary. In preparation.
R.C.ABRAHAM AND D.ALAGOMA: THEIR CONTRIBUTION TO IGBO STUDIES Kay Williamson
1. R.C.Abraham’s Igbo materials and the Institute of African Studies, University of Ibadan I never met the late Dr R.C.Abraham. Other than through his works, I knew of him chiefly through the late Professor R.G. (‘Bob’) Armstrong, for many years Research Professor in the Institute of African Studies, University of Ibadan. Bob Armstrong used to meet Dr Abraham and his wife in London, and he introduced me to his grandson Clive, when he visited Nigeria and spent some time in Ibadan. The close friendship between Dr Abraham and Professor Armstrong led to Abraham’s leaving to Armstrong ‘the task and challenge of finishing the work [on Igbo]’ (Armstrong 1964:52).1 He noted in his obituary of Abraham that: Abraham’s Ibo materials are complete, but a long way from being ready for publication… Fortunately he has written a grammar which is a most useful guide to his intentions and which it is hoped to issue quite soon. The Ibo dictionary is a larger work than the Yoruba dictionary, and is likewise a basic contribution to the study of the Kwa languages. (1964:52) Armstrong clearly took very seriously his duties to his late friend by trying to publish as much as possible of his work. The easiest one was a manuscript eventually published as The Principles of Ibo (1967). I want to quote in full Armstrong’s ‘Foreword’ to the Principles : The Principles of Ibo was left among the papers of Dr. R[o]y C.Abraham at the time of his death, in 1963. He had already given me permission to publish the book photographically for limited circulation among specialists. It should be understood that Dr. Abraham did not regard this Grammar as complete or as being ready for publication in the normal sense. It was intended to accompany his Ibo dictionary, and the ‘Introduction’, which is reproduced here, was written for the dictionary, for which he collected a great mass of material, beginning in 1955. The present work was originally typed for photo-offset reproduction. The author then abandoned the idea of using it in its present form, and made many hundreds of changes and additions to this typescript. Nearly all of these have been edited out of the present edition photographically. The
African Languages and Cultures, Supplement 1 (1992):131–140
106
KAY WILLIAMSON
present work, then, is a historical record of a particular stage in the author’s analysis of Ibo Grammar. It would be quite unjust to complain of its defects and short-comings. If the author had lived, he would certainly have rewritten it completely. It was generous and brave of him to allow it to circulate in its present form as a contribution to Ibo Studies. Once it is clear what The Principles of Ibo is not, we can safely say what it is. It is a fresh approach to the Grammar of Ibo, and it is a first report on a very intensive study over a period of six or seven years. It is a work for specialists, and they will discover in it a fundamental contribution to the understanding of this great and difficult language.
Until I came to prepare this paper, I had understood Armstrong’s statement to mean that the Principles was originally conceived as the introduction to the dictionary; but it now seems that this was a misunderstanding on my part, and that the Principles was conceived as a companion book to the dictionary, just as The Principles of Tiv (1940a) was earlier published as a companion to A Dictionary of the Tiv Language (1940b). It is now clear to me that it is the ‘Introduction’ to the published Principles (actually called a ‘Preface’) which was originally written for the dictionary. I recall Bob Armstrong saying that to prepare the Principles for press he had found a meticulous and devoted printer in the UK (perhaps with Lowe and Brydone, the printers of the book) who had very carefully erased all the corrections in order to restore the MS to its original camera-ready shape. I do not know if a photocopy of the corrected version (i.e. a version containing the handwritten corrections) is in existence or not; it might repay study for Abraham’s final analysis of Igbo. In the ‘Preface’, Abraham discussed the controversy between the ‘Old’ Igbo orthography, which used only six symbols to write eight vowels, and the ‘New’ orthography, which distinguished all eight vowels but wrote three of them with unfamiliar phonetic symbols and changed the value of a fourth (e). Abraham gives an account of the correspondence he had had with the then Government of Eastern Nigeria over the issue and his own proposal which he claims contributed to its solution; namely, the subdots placed below ị and ụ to distinguish them from i and u. This is, of course, exactly the solution adopted by the Ọnwụ Committee in 1961, which has been generally accepted ever since. When I commented to Bob Armstrong that this was a very interesting account which threw new light on the settlement of the issue, he replied that it was his personal point of view and that there was more to it than that; I regret not having asked him to say more about what he meant by the statement. Abraham’s papers also included a large notebook containing materials for his intended Igbo-English dictionary (see Figure 1 below). The MS is in a highly unusual form; instead of a card file, which I suppose is the form in which most dictionaries have been compiled prior to the advent of the computer, it consists of a large notebook in which are pasted thousands of small slips of paper, each bearing an Igbo word or more often sentence, roughly alphabetised by Igbo headwords. Obviously, a photocopy is impossible. If work was to continue from where Abraham had stopped, it would have to be recopied into a more orthodox form.
1
In the Onitsha dictionary (1972:ix), I note that I stated as follows: ‘The large amount of material he [Abraham] had collected has been deposited with the Institute of African Studies, University of Ibadan, where it is available for consultation.’ At the time, Professor Armstrong was Director of the Institute of African Studies. It should be possible to elucidate from his papers whether the materials were given to him personally or on behalf of the Institute of African Studies.
R.C.ABRAHAM AND D.ALAGOMA: THEIR CONTRIBUTION TO IGBO STUDIES
107
Bob Armstrong considered that it was necessary to begin by identifying the dialect in which Abraham had worked. In the ‘Preface’ (unnumbered third page), Abraham makes only a brief reference to his language consultant, as follows: In this dictionary and grammar, I have had the able collaboration of Mr. Alagoma whose knowledge of English is perfect, whose education is of degree standard and whose hearing of tones is acute. For Europeans to teach other Europeans an African language without an African being always present is just an absurdity especially when, as is usually the case, the teacher is merely a learner himself and often, not even a gifted learner. Since Alagoma is not an Igbo name, this statement gives no clue as to his origin or, of course, to the dialect represented. Armstrong had not previously worked on Igbo himself, so he now proceeded to collect long wordlists which were eventually published as A Comparative Wordlist of Five Igbo Dialects (Institute of African Studies, University of Ibadan, 1967). None of them corresponded closely to the dialect of Abraham’s work, and in the ‘Introduction’ he does not in fact make any statement about Abraham’s work; he does, however, refer the reader of the Principles to the Comparative Wordlist. It was Bob Armstrong who, knowing that I was working on an Igbo dictionary (based on the Onitsha dialect) in the early 1960s, made available to me Abraham’s uncompleted Ms. Igbo-English dictionary, stressing that it was a loan only. In my own dictionary I referred to this work and acknowledged having gone through it to incorporate additional materials (Williamson 1972: ix, xv). I also referred to the original motivation for the Comparative Wordlist (1972:ix). Shortly after the end of the Nigerian Civil War, and before my own dictionary was published, I was fortunate enough to meet Mr Dagogo Alagoma of Bonny in , the indigenous language of Bonny, and learnt from Lagos as a member of a Language Committee on him that it was he who had worked with Abraham on Igbo. I told Bob Armstrong of this meeting, and he said: ‘So the dialect of Abraham’s work is Bonny koiné!’ Bob Armstrong had been deeply impressed by the complex phonology of the Southern Igbo dialects (Armstrong 1967:2–5). One of the most complex dialects in the Comparative Wordlist, that from , of the then Owerri Province, had been contributed by the late Mr Apollos A.E.Ahunanya. Mr Ahunanya had stayed on at the University of Ibadan at the time immediately preceding the Civil War, when most Igbo people from the then Eastern Nigeria had left for home, until he had finished typing the stencils for the Comparative Wordlist (Armstrong 1967:3). Bob Armstrong gave him full credit for this devotion to duty. When Mr Ahunanya returned from the East at the end of the war, Bob set him to begin going through Abraham’s material for the dictionary and retyping each word on a card in his own dialect, including full indication of aspiration and nasality as well as tone. When, in 1977, Mr Ahunanya transferred his services to the University of Port Harcourt, Bob Armstrong allowed him to carry this uncompleted work with him, and since Mr Ahunanya’s untimely death from a stroke in 1979, it has remained in my care but not been continued. It has reached the letter ‘G’. Another move towards collecting data towards a more definitive dictionary of Igbo was the employment of a number of undergraduates who spoke various Southern (Central) Igbo dialects by the Institute of African Studies, University of Ibadan, for two long vacations in the early 1970s. These students were trained by me to tone-mark and to write aspiration and nasalisation in their dialects. They went through my own Onitsha-based dictionary files and wrote out their own dialect equivalent on slips which were then alphabetised. This material is also in my care awaiting proper editing.
108
KAY WILLIAMSON
My own understanding of Bob Armstrong’s intentions with regard to the Abraham material is that it represented a koiné or even pidginised form of Igbo (Armstrong 1967:4), and that it would be more interesting to publish a dictionary based on one of the ‘deep’ dialects with a complex phonology, or on material collated from a number of such dialects. I believe that both of us intended an eventual collation of these different materials to produce a standard Igbo-English dictionary which would include the
Figure 1. Sample pages from Abraham’s Igbo-English dictionary notebook
R.C.ABRAHAM AND D.ALAGOMA: THEIR CONTRIBUTION TO IGBO STUDIES 109
110
KAY WILLIAMSON
phonologically complex dialects as well as Standard Igbo forms. I regret that in recent years my attention has been on other projects and that I have not yet seen my way clear to the proper editing and publication of these various materials. 2. R.C.Abraham and Mr Dagogo Alagoma When I was invited to give this paper, I thought it would be of interest to meet Mr Alagoma and obtain his own account of his work with Dr Abraham. Fortunately he now lives and works in Port Harcourt after a distinguished career in public service, first with the UN and later as a Commissioner in Rivers State. He kindly agreed to a tape-recorded interview which was conducted in his office at Dalag Fish Industry Ltd, an enterprise of which he is the Managing Director, on August 22, 1990. I began by asking him about his own background. He told me that his parents were missionaries of the Anglican Church and worked in the area now known as Imo State. He went to mission schools, which in those days were attached to every church, and finished his secondary school at Okrika Grammar School, which was then also an Anglican school. He went on to Teacher Training College and obtained the Higher Elementary teaching qualification and later the Intermediate BA. He taught in both primary and secondary schools until he went to the UK in 1954 to continue his studies. He did his first degree in Economics at the LSE, and two years later went to the University of Manchester for his postgraduate work until 1962. When I asked how he came to meet Abraham, he told me that one day in London Dr Abraham came to his flat and introduced himself. They sat down talking and Abraham told him that an Igbo student had advised him to meet Mr Alagoma as the best available speaker of Standard Igbo. He told him that he was not , but Abraham still wanted to work with him. They agreed to meet for an Igbo but a native speaker of two hours at a time twice a week, first at Mr Alagoma’s flat and later, when Abraham was too weak to make the journey, at Abraham’s house. They began working together in late 1957 and continued with some breaks until Mr Alagoma graduated and went back to Nigeria in 1960. When Mr Alagoma returned to England in early 1961 to do graduate work in Manchester they resumed the work, with Mr Alagoma coming down from Manchester to work with Abraham in London until May 1962, when Mr Alagoma was appointed by the UN to the Economic Commission for Africa. He left the UK and returned to Nigeria to prepare to take up his appointment in Ethiopia. He did not have much further contact with Dr Abraham until he heard of his death in 1963 and wrote a condolence letter to the family. Next I asked Mr Alagoma about how they worked together on Igbo. He explained that he would set homework for Dr Abraham which he would correct at their next session. The homework took the form of dictated Igbo sentences which Abraham would write down, translate and analyse. Sometimes Abraham would ask for the names of certain things, and when they had been given he would ask Mr Alagoma to form a sentence with the word; from there they would go on to the verb. He was particularly concerned with separating the verb from the noun and conjugating it. Mr Alagoma would correct him whenever he made a mistake, so that he could gradually find out the verbs which do not follow the usual pattern and put them in a class of their own. They began with word-building with particular reference to vowels, and then worked together systematically through verbs, nouns, pronouns and other parts of speech. Abraham typed all his answers on his specially adapted typewriter, whose type Mr Alagoma immediately recognised when I showed him the dictionary Ms. He also recognised practically all the examples as his own sentences. We looked at the dictionary manuscript together. He thought the different colours of the slips pasted on the pages of the dictionary had no special significance. I pointed out that most of the slips were carbons, and asked if he knew whether the originals were used for another work, possibly a corresponding English-Igbo
R.C.ABRAHAM AND D.ALAGOMA: THEIR CONTRIBUTION TO IGBO STUDIES
111
dictionary. He told me that the two of them conceived two books, the Principles and the dictionary, which were eventually supposed to come out and be used together. He had, however, never seen a write-up of either of them; he thought Abraham wanted to finish working with him and then do the writing. I also asked him about the list of books in Igbo mentioned by Abraham (1967, last (unnumbered) page of ‘Preface’). He thought that these were Abraham’s own reference works, upon some of which he consulted him while they were working together. When there was any deviation in the books from what Mr Alagoma had given him they would sit together and argue it out; usually where Mr Alagoma thought the reference books were wrong, and told him why, he would eventually agree with him. Mr Alagoma’s own reference books were the Igbo Bible and hymn book. This takes us back to the question of the dialect which they took as a basis for their work. I should explain that Bonny itself has two , but many people in Bonny itself speak Igbo as their best or only languages; the original language is is better preserved in some of the villages as the primary language. Mr Alagoma language. , his mother being from Finima and his father from Oloma, both confirmed that his home language was -speaking villages. I asked him how he came to speak Standard Igbo so well, and he explained being that it was the form that was used in the Anglican Church and of course in the schools run by it, based upon the Igbo Bible and the hymn book, including Sankey’s Sacred Songs and Solos. As a missionary, his father was transferred from one place to another within Igboland, but would always use the same form of Igbo whatever place he was in, and the Christians there would also use it whatever the dialect of their own place. Another matter which I discussed with Mr Alagoma was Abraham’s contribution to the settlement of the Igbo orthographic dispute. I asked whether the use of the subdotted ị and ụ was his own invention or whether he had taken the idea from somewhere; he replied that as far as he knew it was Abraham’s own invention. Mr Alagoma asked what I intended to do with the work, and I asked his own opinion on it. He observed correctly that it would be at least a year’s fulltime job to complete it, but expressed his willingness to help any time I was ready to undertake it. Finally, Mr Alagoma raised the point that although Dr Abraham had done much of the work, he felt that without him the Principles could not have been completed, and wondered whether it would not have been appropriate to have included his name as co-author instead of merely mentioning his name. I replied that the final form of the book appeared after Abraham’s death, edited by Professor R.G.Armstrong; but it is certainly true that in all Abraham’s earlier books it is only his own name that occurs.2 As Abraham refers not only in the ‘Preface’ but elsewhere (e.g. p. 53) to ‘my collaborator’, rather than ‘my informant’, I feel Mr Alagoma’s point is entirely justified. In recognition of this, I have included his name in the title of this paper. Finally, I asked Mr Alagomạ what were his own impressions of Dr Abraham as a man and a scholar. He replied: ‘He was never tired. In spite of his age, and being somewhat asthmatic—in spite of that—he would be coughing and sneezing, and he kept on working. He was a fantastic man.’ 3. The results of the work of R.C.Abraham and Dagogo Alagoma The contributions of the work of R.C.Abraham and Dagogo Alagoma on Igbo may be divided into direct and indirect. The direct ones are: (1) Their contribution to the settlement of the Igbo orthographic problem through the proposal of the subdotted ị and ụ to the Government which set up the Ọnwụ Committee. An interesting topic for
112
KAY WILLIAMSON
research would be to check the other side of the story from Government records or from members of the Ọnwụ Committee. (2) Their contribution to the development of Standard Igbo through the adoption of the form used and popularised by the Anglican Church. It is common to hear the verdict that the ‘Union Igbo’ used in the Igbo Bible was an unsuccessful attempt at a standard language. But Mr Alagoma’s glimpse of the way in which it was used for decades in all the churches and schools of the Anglican Church clearly shows that it had an enormous effect in forming the views of educated people as to what was and was not ‘standard’ Igbo. More fieldwork based on interviews with people who operated or grew up in the system could be useful. (3) Their contribution to the study of Igbo grammar through The Principles of Ibo. As Armstrong stated, this work is unfinished; hence its subtitle Archival edition of typescript. It contains a great mass of detail, and as Armstrong states is ‘a fresh approach to the Grammar of Igbo’; so fresh, indeed, that rather little use has been made of it by later scholars. I see two main reasons for this. First the tonemarking system used is complex and out of line with that used by anyone else; rather than discuss it here, I refer interested scholars to the clear critique by Armstrong in ‘A note on marking Igbo tones’, which immediately follows the ‘Foreword’ in the Principles. Secondly, the complex style of presentation, which makes it difficult to read; it is mainly through the examples that one can equate his terminology to that of others. Almost certainly there is more to be learnt and integrated into the general grammar of Igbo from a detailed study of the Principles than has yet been done. The fact that he did not show the typescript which was ‘complete in its present form not later than 1961’ (Principles : first (unnumbered) page of Armstrong’s ‘Note’) to Mr Alagoma, who worked with him until May 1962, shows him as an individualist who limited co-operation with his trusted collaborator to oral discussion and reserved all the writing for himself. (4) Their contribution to the study of Igbo lexicon through the uncompleted dictionary. It seems unlikely that this will ever be published in the format Abraham conceived. Too much remains to be done. What seems to me to be required is a project in which their material is fully utilised, but combined with the contributions of others and edited in a consistent and accessible format.
Their indirect contributions are: (5) Their stimulus to the study of Igbo dialects and phonologies through Armstrong’s Comparative Wordlist of Five Igbo Dialects. The fact that they were actually using the most standard form of Igbo available does not diminish the interest in some of the dialects with much more complex phonologies. (6) Their stimulus to the development of other Igbo dictionaries, for example my own (1972), which has already been discussed. In short, their contribution to Igbo studies, incomplete as it is, is still a major and not yet fully utilised one.
2
It is also true that Abraham’s near-contemporary Margaret Green included the name of Rev. G.E.Igwe on the titlepage of A Descriptive Grammar of Igbo, and has a graceful paragraph (1963:xiii) delimiting their respective contributions.
R.C.ABRAHAM AND D.ALAGOMA: THEIR CONTRIBUTION TO IGBO STUDIES
113
REFERENCES Abraham, R.C. 1940a. The Principles of Tiv. London: Crown Agents for the Colonies. ——. 1940b. A Dictionary of the Tiv Language. London: Crown Agents for the Colonies. ——. 1967. The Principles of Ibo (Archival edition of typescript). (Occasional Publication No. 4.) Ibadan: Institute of African Studies, University of Ibadan. Armstrong, Robert G. 1964. Roy Clive Abraham, 1890–1963. The Journal of West African Languages 1(1):49–53. ——. 1967. A Comparative Wordlist of Five Igbo Dialects. (Occasional Publication No. 5.) Ibadan: Institute of African Studies, University of Ibadan. Green, M.M., and Rev. G.E.Igwe. 1963. A Descriptive Grammar of Igbo. Berlin: Akademie-Verlag; London: Oxford University Press. Ọnwụ Committee. 1961. The Official Igbo Orthography as Recommended by the Ọnwụ Committee in 1961: Notes on Script and Spelling for Teachers. Enugu: Government of Eastern Nigeria. Williamson, K. 1972. Igbo-English Dictionary: Based on the Onitsha Dialect. Benin City: Ethiope Publishing Corporation.
R.C.ABRAHAM’S BOOKS ON TIV D.W.Arnott
Although the Dutch Reformed Church Mission had in 1932 produced a combined grammar and dictionary of Tiv (Malherbe 1931), tones were virtually ignored, and Abraham’s books were the first in which tone was systematically marked. In his early The Grammar of Tiv, printed in 1933 by the Government Printer in Kaduna, he cleverly overcame the lack of special symbols by the use of different type—low tone syllables being printed in lower case, while large and small capitals were used to indicate high and ‘mid’ tone respectively—an effective if unconventional device. In his later language books—The Principles of Tiv (1940a), A Tiv Reader for European Students (1940b) and A Dictionary of the Tiv Language (1940c)—he was able to use subscript diacritical marks (- for Low, the stylised arrows ┴ for a ‘downward glide’, and ^ for an ‘upward glide’, i.e. HL and LH sequences respectively); and the careful marking of tone throughout these books is one of their valuable features, as well as a tribute to the meticulous attention to detail which marked all Abraham’s work. Even if his awareness of tonal differences made Abraham exaggerate the intervals when himself speaking Tiv—so that some Tiv found him harder to understand than other officers whose untutored pronunciation was based on rough-and-ready imitation—nevertheless his accuracy in recording the tonal contours is such that, given his approach, in all the 515 pages of these books wrongly marked tones on individual words are exceedingly rare. It is only the tonal behaviour at juncture which seems to have escaped him, even though a few passages in the Principles show that he was aware of the ‘downstep’ feature which puts Tiv in the category of ‘terraced-level languages’ of Welmers’ classification (Welmers 1959). It is, in fact, unfortunate that Abraham’s apparent intention to publish a comprehensive treatment of the tonal system was not realised. Typically for Abraham, the list of ‘Other works by the same author’ in his three language books included ‘Tiv phonetics and tonal principles’ (1940), as well as ‘Short classified English-Tiv dictionary’ (1940), neither of which actually appeared, although publication details are given. (This may well be another instance of Abraham’s recurrent problem of getting financial backing for publication, of which he so often complained.) Such a study would indeed have been interesting, even if now superseded by later research. The Principles is an admirably succinct presentation of the results of thorough familiarity with the language in actual use and of painstaking analysis. But it is perhaps typical of Abraham’s published grammatical work that, while adept at classifying the various linguistic phenomena and describing the behaviour of the various individual categories, he less often made clear the interrelation between them. Thus the Principles gives full details of the various types of word that feature in the noun-class system
African Languages and Cultures, Supplement 1 (1992):141–145
R.C.ABRAHAM’S BOOKS ON TIV
115
(including adjectives, pronouns of various types, demonstratives, possessives etc.), but does not give a coherent picture of the system as a whole, or of the close phonological and tonological links between words of different types within each class (see Table 1 below). These include the recurrence of ng- in some classes, in others, and of labial mb- or -v in others; and cutting across these patterns is of velar k- or -gh another, one group of classes having a recurring i vowel or initial y-, another group a recurring u vowel or initial w-; another group again a recurring a or u vowel but zero-initial. Tonally too, there is consistency in the contrasting tone patterns associated with low-tone classes on the one hand, and high-tone classes on the other. These patterns are not brought out clearly by the piecemeal treatment in the Principles. Much the same is true of Abraham’s description of the verbal forms. Understandably the various tenses or aspects are treated in separate chapters, but there is no attempt to show the underlying regularity and cohesiveness of the tonal behaviour that is apparent in a review of the whole scatter of forms in the various verb classes. This treatment of the grammar in small self-contained sections is explained by the fact that the Principles, like the Reader, was intended for European officials and others needing to learn Tiv, and certainly I myself found it an invaluable help in learning the language. It does lack the graded vocabulary and exercises that one might expect in a teaching grammar, but all grammatical statements are well and fully supported by illustrative examples. The accuracy of much of the material presented has been a sound basis for more detailed studies by later scholars in the light of modern linguistic insights (see Bibliography). Their research has led to some modifications of Abraham’s findings, such as the analysis of Abraham’s ‘mid’ tone in terms of downstep, or an underlying low tone, the rejection of his treatment of the nasal element in prenasalised consonants as syllabic, clarification of the confusion in his analysis of some disyllabic same-vowel sequences as monosyllabic and vice versa, the reduction of the verb classes from eight to six, and the recognition of only one E vowel instead of the two distinguished by Abraham. The Reader begins with a considerable number of sentences illustrating the various grammatical features; they are not arranged specifically in grammatical sections, but have detailed cross-references to the relevant sections of the Principles. The second part reproduces extracts from various texts, some previously published, others collected by Abraham himself, but all fully tone-marked and with cross-references to the Principles. The Dictionary is a typically thorough piece of work, accurate in detail and fully tone-marked throughout. The coverage of verbs is particularly comprehensive, and ample details are given of their morphologically and
Table 1. The Noun Class System in Tiv
116 D.W.ARNOTT
R.C.ABRAHAM’S BOOKS ON TIV
117
tonologically differing aspect-forms, as are (in most cases) the plurals of nouns and their augmentative, diminutive and dependent by-forms. The arrangement of items, especially nouns, might be perplexing at first sight. But it is practical, and based on the sound principle—for Tiv—of listing nouns on the basis of the first consonant of the singular, the (alternating) vowel prefixes being ignored for this purpose. Phonologically based alternation in the first consonant in some words is dealt with by double entries, e.g. the plural form a-sema is listed before sembe-sembe, with cross-references to the singular i-shima, under which full details and citations are given); and double entries, with cross-references, are used to cope with the considerable number of alternative forms. Tiv’s numerous ideophones are well covered. The majority of all entries are well illustrated by citations, and glosses of entries and citations are sound; and as with his dictionaries of other languages, one is amazed at the great range of contexts from which the entries and citations are taken. Many words of religious or cultural significance, however, are not translated, the reader being referred instead to the author’s The Tiv People (1933b)—a comprehensive ethnographic survey which shows us Abraham the anthropologist, the obverse of Abraham the linguist. BIBLIOGRAPHY Abraham, R.C. 1933a. The Grammar of Tiv. Kaduna: Government Printer. ——. 1933b. The Tiv People. Lagos: Government Printer. [2nd edition, 1940, London, Crown Agents.] ——. 1940a. The Principles of Tiv.* London: Crown Agents for the Colonies. 1940b. A Tiv Reader for European Students. London: Crown Agents for the Colonies. ——. 1940c. A Dictionary of the Tiv Language.* London: Crown Agents for the Colonies. [*Both re-issued in 1968 by Gregg International, Farnborough, with a new Introduction by D.W.Arnott.] Arnott, D.W. 1958. The classification of verbs in Tiv. BSOAS 21:111–133. ——. 1964. Downstep in the Tiv verbal system. African Language Studies 5:34– 51. ——. 1967. Some reflections on the content of individual classes in Fula and Tiv. In La classification nominale dans les langues négro-africaines, ed. by M.G. Manessy, pp. 45–74. Paris: Editions du Centre National de la Recherche Scientifique. ——. 1969. Tiv. In Twelve Nigerian Languages, ed. by E.Dunstan, pp. 143–151. London: Longmans. 1980. Tiv. In West African Language Data Sheets, ed. by M.E.Kropp Dakubu. Leiden: African Studies Centre. Hyman, L.M. 1979. A reanalysis of tonal downstep. Journal of African Languages and Linguistics 1(1):9–29. Lukas, Johannes. 1952. Das Nomen im Tiv. Anthropos 47:147–176. Malherbe, W.A. 1931. Tiv-English Dictionary with Grammar Notes and Index. Lagos: Government Printer. McCawley, J.D. 1970. A note on tone in Tiv conjugation. Studies in African Linguistics 1:123–129. Sibomana, L. 1980. Aspects of Tiv morphophonemics. Afrika und Übersee 63(1): 69–77. ——. 1981. Tonal lexicalization of Tiv nouns. In Festschrift zum 60 Geburtstag von P.Anton Vorbichler, ed. by I.Hofmann, pp. 155–171. Wien: Institut für Afrikanistik und Ägyptologie der Universität Wien. ——. 1983. Tonal structures of Tiv verbs. Afrika und Übersee 66(1):149–158. 1984. Displacement and lowering of Tiv tones. Afrika und Übersee 67:277– 287. Terprstra, G. 1968a. A Tiv Grammar. Mkar: Sudan Interior Mission. ——. 1968b. English-Tiv Dictionary. Ibadan: Ibadan University Press. Welmers, William E. 1959. Tonemics, morphotonemics and tonal morphemes. General Linguistics 4:1–9.
R.C.ABRAHAM AND THE TIV PEOPLE Frances Harding
Apart from the fable, the dramatic sense of the Tiv is exhibited in his songs. Here he has a much freer field for the exercise of his imagination for the bulk of his songs are not ritual or traditional but topical and take as their theme the happenings of the day. (Abraham 1940:82) …this conservativism, this dread of change, added to the Bantu ingrained democratic attitude. (1940:33) 1. Abraham and his approach to the Tiv ethnography Robert Armstrong rightly considers Abraham to be ‘ahead of his time’ in reference to language (1964:52), but when he moves into ethnography, Abraham is very much a man of his time, associating race, language and culture to make judgements. For whereas Abraham was able to note culturally-based differences between his own experience and that of other people, he viewed them as qualitatively and inherently different, rather than as arising from learned cultural factors. In this paper, I will argue firstly that Abraham was wrong to consider that the Tiv people were restricted in their use of the narrative, and secondly that the Tiv, neither then or now, have an outlook which can be described as ‘conservativism, a dread of change’. I suggest that this is evident in contemporary Tiv drama as well as in both the everyday use of proverbs and their use in drama. Finally I suggest that it was Abraham’s own fear of a radical change in relationships of power which compelled him to deny, even to himself, the intellectual and conceptual ability of the Tiv people. Abraham came to write his ethnography not out of personal interest in Tiv society, but as David Dorward (1971:466) has stated: ‘Abraham (and Downes who wrote at the same time as him) wrote as part of the government response to the haakaa movement in Tivland in 1929.’ The primary objective was to gain an understanding of the particular religious practices of the Tivs which were rumoured to give rise to ritual murders and cannibalistic rites. This approach focused on an association between these putative practices of the Tiv and the challenge which the Tiv presented to the colonial government, and consequently it determined Abraham’s approach to the realities around him. From his writing, it seems that there were three categories of Tiv experience for Abraham. Firstly, there were those actions such as the putative ritual murders and cannibalistic practices and the related supernatural
African Languages and Cultures, Supplement 1 (1992):147–161
R.C.ABRAHAM AND ‘THE TIV PEOPLE’
119
practices which were outside his experience. These he wrote about as fully as he could and firmly condemned them. Secondly, there were practical categories like house-building, farming, child-rearing, marriage, which he simply described. In this category he included minor rituals involving the sacrifice of a goat or a chicken, as well as some of the celebratory rituals which centre mainly on feasting and dancing and music. It is, however, on the third category in which he was confronted by facts which did not correspond to what he anticipated, that I want to concentrate. On these occasions, conflict arose out of the evidence which Abraham resolved in one of three ways. Firstly, he either ignored the contradictions they set up; secondly, he merely noted them as exceptions or irregularities, or, thirdly, he fitted them into his existing framework of ‘appropriate’ attitudes or behaviour. I can find no example of him changing his mind in the face of evidence. It is in this sector that I have placed his references to the conceptual and intellectual capability of Tiv people, in particular as it is articulated in their verbal arts practice. Abraham drew for his framework of reference on the published material of missionaries, administrators and Levy-Bruhl (Abraham 1940:30), placing much greater importance on the printed word of a disparate group of people from East Africa and the Far East than on the evidence of his own eyes, ears and personal experience. Had he not done so, but chosen to follow the logic of his evidence, his contribution to a change in colonial thinking could have been significant. As we shall see, he saw all the evidence but either refused to recognise its significance, or, if recognising it, refused to commit it to writing as a public statement. 2. Abraham and the Tiv It is no surprise to find that Abraham begins his ethnography not by a description drawn from his own experience of Tiv people, far less by quoting a Tiv, but by reference to an observation made some seventy years earlier by Baikie: …wild in look and ruder in dress, greatly tattooed and carrying constantly with them their bows and arrows these men seemed perfect impersonifications of savages…at first rather shy… The Mitshis… are all wilder and less intelligent than any of the other African races…(p. 1) Though ‘shy’ and ‘wilder’, Abraham feels that by ‘the present time’, this portrays the Tiv in an unneccesarily ‘gloomier aspect’ since: …the civilising influence of contact with European officials and the agents of firms dealing in Beniseed as well as constant intercourse with Hausa…have had a civilising effect. (p. 1) Abraham’s choice of language in reference to the Tiv people continually reveals his attitudes, and never more so than when he is struggling to accommodate a theoretical point made by the Tiv into his own colonial framework. An example of this sort occurs early on in his book where he notes that the Tiv say that at a certain point in their history they had no knowledge of bows and arrows. Abraham states: ‘Considering that the Tiv words for bow and arrow are roots common to the bulk of Bantu languages, their claim is an absurdity as it stands’ (p. 13). Language adopting this dismissive approach increases throughout Chapters one and two—for example, in describing the Tiv vocabulary of direction, he writes that the Tiv have ‘only… upwards, downwards, nearness and distance’ (p. 27). Aware that the words are linked to a conceptualisation of the world as a stooping man whose topography determines the vocabulary of direction, Abraham observes
120
FRANCES HARDING
that ‘the indication of direction is a serious difficulty to Europeans’ (p. 27). Whether or not the Tiv people took this difficulty into account when their language was being formed is not clear. A negative presentation of Tiv conceptualisation is pervasive throughout The Tiv People, but perhaps nowhere more disturbingly for its sheer unfairness than in the presentation of the Tiv classification of plants and animals. First of all, Abraham introduces the Tiv knowledge of the natural world not as knowledge of a sort frequently found among rural people, but only and specifically as being ‘in common with all the Bantu people’. He describes them as having an extraordinarily accurate knowledge of the fauna and flora of their area, and describes in detail how the Tiv classify and subdivide the world of living creatures into many categories. What excites Abraham about this is best said in his own words: To those of us who are familiar with native races and their speech and modes of thought, the above classification is very striking, for whereas all native races are keen observers of fauna and trees…they very rarely possess the faculty of grouping objects by ideas and common characteristics such as we find among the Tiv. (p. 29) Having gone as far as he could in praise, Abraham does not free himself from his preconceptions to pursue the logical conclusion that it is his own perception of the ‘native races and their…modes of thought’ which need to be questioned; instead, by referring to a linguistic classification, he is able to turn the evidence around to support his own bias: …considering the rigid system of noun-classes employed in the Bantu languages, it is after all not so surprising to find the principles carried forward into the classification of living creatures. (p. 30) He observes however, that the classification is ‘not always one which is familiar to Europeans’, and citing the wide range of Tiv words available for describing the act of ‘cutting’, and comparing this with the absence of a single generic term comparable to the English word ‘cut’, he feels that it presents an interesting ‘psychological’ problem which he resolves by reference to Levy-Bruhl: If Levy-Bruhl is correct, and there seems no doubt of his correctness in view of his close reasoning and the wealth of examples adduced from many quarters, we are in face of two phenomena which mark different strata in the development of the Tiv, the first characterised by over-differentiation of shades of meaning at the expense of generalisation, and the second marking a later stage of progress and signalled by a definite working of the intellect, proceeding on lines of logical classification informed by abstract thought. (p. 31) So the Tiv are at one and the same time too detailed in their classificatory approach, denoting an ‘early stage’ of development—‘logical categories are alien to primitive mentality’ (p. 30)—and yet at the same time they show a conceptual ability which Abraham cannot reconcile with such an imagined ‘mentality’ (p. 32). In yet another example of Tiv thinking which challenges Abraham’s preconceptions and shakes him profoundly, once again, rather than use facts to explain and understand, he can only record his amazement and, as before, resort to Levy-Bruhl. On this occasion, he describes the ‘phenomenal development of the memory…and this primitive trait is very marked among the Tiv’ (p. 32). Years later, Abraham himself is described by Robert Armstrong as having a ‘really fabulous memory’ (1964:52). Two areas in which Tiv reasoning and intellect surpassed Abraham’s expectations are thus described not as opening up new ways for Abraham to think about Tiv people, nor simply as normal or brilliant, but as
R.C.ABRAHAM AND ‘THE TIV PEOPLE’
121
‘extraordinarily accurate’ and a ‘phenomenal development’. Even the positive acquires a negative veneer. Abraham sums up these diverse capabilities as puzzling contradictions which he sees as being lodged not in himself, but in the Tiv: Psychologically, one is puzzled by the Tiv mentality at every step, for side by side with their accurate classification of living creatures and the very high standard of their metal ware, we often find complete inability to think abstractly and a simple question frequently produces a blank stare; it is for this reason that Baikie described them as stupid, no doubt. (p. 32) However, Akiga Sai, writing in Akiga’s Story, tells how in 1927, when the colonial government were searching for ways to deal with the continuing administrative difficulties, a party of almost 300 Tiv elders were taken on a visit to Kaduna, some four hundred miles north of Tiv country, in order to see whether they liked the area well enough to move there. After entertaining the elders, the white man asked if they would like to live there, to which the elders replied ‘yes’. But Akiga says: When his back was turned, they said to each other, ‘What sort of a fool would leave his own people to come and lead a miserable existence far away in a foreign country? However they seem to want us to say yes.’ (1939: 397) Clearly the Tiv had their own ideas of who was stupid in the question and answer relationship. 3. Abraham and the Tiv performance arts This introduction to Abraham’s qualitative assessment of Tiv theory contextualises his approach to the verbal arts in his ethnography. If, in an assessment of the Tiv performance arts given by Abraham, one were to weigh the fairly full descriptions of the action-based performance of rituals against those which Abraham presents as being word-based, it would appear to the reader that words were merely the part of Tiv culture used to transmit elementary, simplistic narratives, aphorisms and proverbs which could be cross-referenced word-for-word or meaning-for-meaning with forms in languages such as English. It would also appear that they were carriers of information separable from the immediate performative context—speaker, occasion, audience—in which they were uttered, and further separable from the larger historical-social-religious contexts from which they emanate. In fact, as we shall see, the verbal arts articulate not only knowledge in a transmissible form, but a context of knowledge which constitutes a theoretical perspective against which contemporary experiences can be measured. In his ethnography The Tiv People, Abraham refers to several forms of performance—specifically rituals, songs, proverbs and fiction—or what he calls the ‘fable’, ‘folktale’ or ‘epic’, i.e. orally transmitted fiction. Abraham is correct to note the importance of fiction (the ‘fable’) in Tiv culture, and to associate with it a ‘dramatic sense’ in the Tiv. He is not so accurate, however, in associating ‘ritual and traditional’—in a pejorative sense of limited and unchanging practice—with fiction, whilst allocating imagination only to song. Abraham wrote on four different forms of performance—rituals, songs, proverbs and fiction. As well as simpler forms of rituals, he also described the more elaborate ivom (p. 66) and ibiamegh (p. 94). There is a change in Abraham’s attitude when he approaches those forms of performance in which the spoken word was the primary vehicle of communication, and he finds that he was unable to exercise his customary ‘for or against’ criterion. He realised that words—to which his linguistic competence gave him
122
FRANCES HARDING
access—were not the units of definable meaning that he wanted them to be, but that they carried with them as invisible baggage whole cultures and philosophies. Abraham was passionately committed to translating languages into English. To acknowledge that inherent in other languages were philosophies and systems of thought which might challenge European ways of thinking was to jeopardise the whole basis on which colonial rule was predicated. There could be no ‘different and equal’, but only ‘different and inferior’. 4. Abraham and Tiv songs Of the songs which Abraham quotes, he notes that there are three types: a dirge, icam (a song and response), and imo (a solo). About the icam, he states that: ‘To the Tiv mind the composition of such a eulogy suggests a roof built over the head of the one eulogized’ (p. 82). The ‘western’ mind, on the other hand, has no such practical metaphor but prefers to describe such singing in the realm of the mystical, as ‘heavenly’ or ‘angelic’. The songs which Abraham analyses were collected by an officer of the Education Department, and perhaps the person who provided the songs for him saw it as an opportunity to give an insight into the way the Tiv were feeling about some major issues at the time. For example, there is a song about the tough conditions of road work: See what has befallen me, what have I done to deserve it. In days gone by I farmed for my mother and when I was tired I rested. Then there was nobody to beat me and I never had to work as I am working today. Work! Work! Work! My taskmasters grind me down. (p. 83) Abraham comments: It is only fair to say that this gloomy picture has no foundation in fact; the Tiv is disinclined to any effort, physical or mental, if he can avoid it, and is paid good wages when required for manual labour on the roads etc. (p. 83) Abraham much prefers what he calls the ‘other side of the picture’ and quotes another song: The Tiv are shouting in unison ‘There is no fighting any more; The white man has come and forbidden affrays, So we fight no more.’ (p. 83) Other songs capture a series of more immediate reactions to current affairs, such as the abolition of marriage, and the banishment into exile in Kaduna of the chief, Ugba: When the Hausas speak the white man believes them. Our chiefs took many bribes and were sent away to Kaduna. We pushed the motor till it started off for the North.
R.C.ABRAHAM AND ‘THE TIV PEOPLE’
123
Baba trussed up Moji in chains and took him to prison. The white man is a scourge. This chief and that chief have been given staves of office by the European, but of what use are they? Once they used to hold courts but now they do so no longer. Favour has fallen on Gom Kpanju but now he no longer tends his paternal kin. Wait though he’ll be hanged one of these fine days. The white man has imprisoned Abagi and he is wailing… ‘Oh Jato Aka, this is the end for me, for I am to be hanged.’ (p. 83) Several writers since Abraham have referred to Tiv songs, e.g. Keil (1979), and although Abraham gives a curious selection, he is correct to note the topicality of song. 5. Abraham and Tiv proverbs in drama Proverbs are another verbal performance art which Abraham includes in his ethnography, and his way of presenting them was to equate them with English equivalents. Thus, we find ‘You have tied a bone round the dog’s throat’, or ‘I never sow groundnuts when the monkey is watching’ explained as: ‘You are tempting providence’. He himself uses as explicators proverbs which are themselves quite quaint—for example, he says: ‘The goat went foraging but did not give a share to the sheep’ is equivalent to ‘Every tub must stand on its own bottom’. Even his choice of language acquires a certain quaintness in describing proverbs: ‘When a man is away, the monkey eats up the maize and enters the farm hut which is on all fours’ (my emphasis) with English ‘When the cat’s away’ (p. 34). The use of a proverb is performative in the sense that it confers on the speaker, however briefly, a presentational role. By implying the performative conventions associated with the form, the speaker is enabled to stand back from a situation and comment on it, thereby referring the speaker and the hearer to a wider context. This action removes them from the isolating responsibility of the immediate context, and at the same time locates the specific situation firmly within a known repertoire of possible situations. There is considerable use of proverbs both as an art in itself, as well as being utilised within the performance arts. Of the 26 proverbs which Abraham selected, and which Bergsma (1970) comments on in a paper on proverbs as social control, none are to be found in the plays which form current Tiv dramatic writing. This may be a reflection of current preference, or it may indicate that those proverbs which were given to the Dutch Reformed Church Mission (Bergsma 1970), and which Abraham used, met certain perhaps unspoken requirements of the enquirer. 6. Contemporary drama/use of proverbs in drama Proverbs are used in contemporary drama, including television playscripts designed specifically as a Tiv contribution to a Nigerian national network drama programme. In a one-hour long play ‘Preserve our Culture: Square One’ by Dennis Akpede (n.d., performed 1984), there are four proverbs, three of them
124
FRANCES HARDING
spoken by senior men and the other by an older woman when she is alone. The first proverb used is that of the banana tree and, presumably because it is networked nationally, the character explains it: Papa Wanban: When the banana dies, its children succeed it. Now our forefathers are no more, it is our responsibility to take over from where they left. That is the meaning of tradition. (p. 2) The woman’s proverb, as she reluctantly takes ‘salt’ (money) from her future son-in-law and thus indicates her tacit support of their elopement, is: Mama Wanban: Well, our people have a saying that a man who does not want strange footprints in his backyard must fence it up. (p. 9) The other two proverbs in the text are spoken by the elderly intermediary in a proposed marriage arrangement which forms the basis of the play. The first comes when the intermediary, Ugese, is waiting for the father to finalise arrangements for the marriage of his daughter to the old man, Ateiwa: Ugese: Papa Wanban, it is said that a hungry man has no patience when food is placed before him. Papa Wanban: That’s a popular saying of our people. (p. 9) Later, when Ugese discovers that the girl has eloped, he says: ‘Well, it’s too late to cry when the head is off’ (p. 11). Other plays written in English, but for a largely Tiv audience, may use proverbs less often and other formal arrangements of words more often, e.g. praise names and poetic metaphors. In a play, ‘Tor Ya Tar’ (Ahura, n.d., performed 1989) which is structured as a play within a play, the Director describes the aspirations of the theatre company Director: ‘We have only hung our bags where we can reach them’ (p. 2). This is an implicit reference to the Tiv bags which are such a distinctive and prestigious feature of Tiv elders’ dress. Further on, in replying to the Councillors’ complaint that they have to sit among the ‘stinking rabble’, the Director replies: ‘Zaki, we were all born in the stench. Only some fast ones reached the soap first and used all of it, to scrub their outside’ (p. 3). Not perhaps the most profound of sayings, but an amusing one. On one occasion I was the recipient of a most delightful Tiv proverb. I had gone to meet a theatre director in Benue State, passing the elderly gateman on the way into the building. The Director was not in his office, so I waited, talking meanwhile with the young men and women who were the performers in the Company. The gateman watched us for a while and then commented: ‘She came to see the tree and is satisfied with the branches.’ 7. Abraham, storytelling and drama Abraham treated the word as a closed entity whose meaning could be defined. In approaching fiction, therefore, he privileged the narrative content, relegating style and interpretative practices to a secondary role. In a written authored text, different versions of the elements in the narrative (characters, plot, location, time) are defined in relation to a known printed text as alternatives, and consequently are subject to censorship as ‘correct’ or ‘incorrect’; but in orally-transmitted fiction each utterance is valid in its own right for the time and place that it exists. Neither is its transience a negative factor. How it is presented is in reference to style or entertainment value rather than as a means of censoring its accuracy in relation to a previous account. Indeed the whole process of shouting down the speaker in order to present one’s own version is part of an overall convention of performance and a continual process of recreation. Thus, for the period of the performance, the verbal arts become the forum for experimental social, political and religious restructuring, even sometimes satirising the figures of authority. Storytelling,
R.C.ABRAHAM AND ‘THE TIV PEOPLE’
125
seemingly canonical, contains within it the potential to be carnivalesque and to be, in temporal terms, the liminoid space where liberating action can take place. Abraham perhaps realised this and realised its implications for colonial rule, and consequently confined himself to presenting words. Fiction in Tiv is known by two names—kwagh-hir and kwagh-alom. Kwagh is the singular form of ‘thing’, hir means ‘wonderful, strange’, so kwagh-hir means ‘something wonderful’ or a ‘wonderful thing’. This refers to the stories about the supernatural people who inhabit Tivland, the adzov. The second way of referring to fiction—kwagh-alom—means ‘the things of the Hare’—alom means ‘hare’ but is also the name given to the Hare (Alom) as a character in the stories. Whilst kwagh-alom can only be used to refer to the animal stories, kwagh-hir is used to refer in a more general way to storytelling. The elaborate contemporary theatre of movable sculpted forms which is so popular in Tivland today, and whose content relies partly on familiar fiction, is always called kwagh-hir and not kwagh-alom. In the stories which Abraham includes in the chapter entitled ‘Foiklore’, he names the hare, the lion, the tortoise, the fly, the mosquito, the leopard, the fish, the night-adder, the owl, the hyena, the bush-rat, the cricket, the duiker and the spider as characters. As rural people, it is hardly surprising that the Tiv give a significant role to animals in their storytelling. Once again, however, Abraham saw this feature not as a straightforward outcome of daily life, and once again his choice of language—this time a mediaeval echo in ‘beast epic’—indicates his evolutionist attitude: ‘The Tiv share with the other Bantu the predilection for the beast epic’ (p. 64). In fact the animal stories are only one of three categories of fiction which Tiv tell—the other two concern non-human, supernatural characters called the adzov, in addition to humans. Abraham goes so far as to note that: The animals are animals in name only, and the stories are realistic dramatizations of human beings thinly cloaked in animal form, revealing for the most part, the seamy side of human nature. (p. 64) He goes on to note that a feature which distinguishes the Tiv stories from ‘their counterparts elsewhere’ is that ‘the hare…is on his mother’s side descended from the dead, thus attributing his sagacity to the accumulated wisdom of the dead and the living’ (p. 64). If this recognition of the difference between Tivs and others could have been deepened, then it would no longer have been possible for him to continue to talk about ‘Bantu predilections’ and ‘pre-logical mentalities’, or to note the contradictions as irregularities or exceptions. The novel Return to Laughter (1964) by Elenore Smith Bowen (Laura Bohannan) is the best-known published account of Tiv storytelling, and is itself a testament to the different quality of truth which emerges in fiction from that of bare ‘fact’. So important does Bohannan consider storytelling in Tiv life that not only is its performance the occasion for the laughter which gives the novel its title, but it is also recognised as a healing, binding force in the community. Storytelling tells truths and opens up new understandings; it reminds people of old obligations and shared needs, and focuses on common mystical and physical enemies which threaten them. This complex web of meanings and intentions already places it at a substantial distance from an imagined ‘Bantu predilection for the beast epic’. In Bohannan’s account there are archetypal humans and superhumans as well as some very ordinary people —blind or conceited, ambitious or loving— as well as an apposite presentation of the colonial authority: As the administrator, Ikpoom was first patient and then, with each successive complication of the old man’s tale, more and more annoyed and abrupt. As the interpreter between them, Ikpoom floundered, piled confusion on confusion, substituted women for goats and nephews for uncles. Soon Ikpoom had
126
FRANCES HARDING
created a tangle of wild misunderstandings and exasperated exhortation from which an entirely irrelevant decision emerged on wholly erroneous grounds. Ikpoom…looked perplexed…picking his teeth with a spear, fiddling with his badge, or tugging at a non-existent moustache so vividly that I saw Sackerton himself… ‘That is the way it is,’ gasped Kako… ‘It is thus indeed,’ and again he burst out laughing… ‘It is so. They judge us, but they do not understand us, and what can we do?’ (1964:292–293) Laura Bohannan understood that storytelling was about more than words. In Abraham’s collection of stories, there are no human characters or adzov, only animals. Several men, whose personal inclination and talents had won them a reputation as performers and entertainers had long been pushing the boundaries of storytelling beyond the wholly verbal. From 1960 onwards, kwagh-hir increasingly became a theatre of three-dimensional, kinetic sculptural forms. The narrative structure was based on contemporary characters or issues, as well as on ideals held by Tiv people and on the mysterious world of the adzov. It was—and has remained—an explicit challenge to the powers of the mbatsav whether they are perceived—and mockingly exposed—in their reputed cannibalistic role, or in gaining control over life and death. 8. ‘A Close Shave’ by Chris Kyoive This play (performed in 1984) meets Abraham’s observation that the animals’ names are used to cloak human characters. The characters in the play all use either the classic Tiv animal spirit (adzov) or archetypal names. There is Tor, the Chief and Wantor, his son; there is Alom and his wife Anjieke, and there is Hiinyongoga, the hero who is wronged and tricked by Alom. Basically Alom is jealous of Hiinyongoga who has just received a National Award for being ‘the one and only one with an exemplary behaviour [sic]’. Alom decides to frame Hiinyongoga for the theft of the Chief’s ceremonial robe. As Alom observes: ‘The punishment can be anybody’s guess since stealing carries a death sentence.’ He plots with his wife to remove the cloak from the Chief’s home and plant it among Hiinyongoga’s things. This is successfully carried out. Before the ‘culprit’ is known, firstly the Chief’s wife, then the elders, call for the death penalty to be carried out on the thief when he is found. The words ‘executed’ and ‘hanged’ are used. Even Hiinyongoga calls for it before the cloak is found in his premises: Hiinyongoga: You cannot get anything (here) because I am not a thief. Alom: That is not for you to say. What if we get it here? Hiinyongoga: Kill me for such is the lot for thieves if and when they are caught. (p. 17) This seems to be promotion of the official line on theft and the death penalty. This is made explicit by the chief, Tor, who says: He is to face the hangman’s rope and this must be done with some degree of ruthlessness and dispatch so that the lesson can be long-lasting and the now rampant cases of moral decadence and indiscipline completely wiped out. We have to register our protest to these happenings…for the sake of posterity, so that we don’t go down as a generation of thieves and evil-doers. This is why you are all gathered here. The execution will serve as a deterrent. (p. 19)
R.C.ABRAHAM AND ‘THE TIV PEOPLE’
127
The heartfelt songs referred to above, bewailing the death by hanging of Tor Abagi and sung in 1927, describe a different set of feelings. The play proceeds and Hiinyongoga is renamed Hiinyongo—that is from ‘guiltless’ to ‘guilty’. Just as sentence is about to be carried out, the Chief s son is brought in on the point of death from a snakebite. Hiinyongoga is named as the only one capable of curing him. Alom tries in every way not to have the execution postponed but he is outvoted. Hiinyongoga calls for the stomach of either the chief or Alom as an essential ingredient in the curative medicine, and the chief has no hesitation in selecting Alom for the task. Alom is killed and his stomach used in the medicine. The Chief’s son recovers, Hiinyongoga is rewarded, the plotters are revealed, and order is restored. The Chief makes a final statement: ‘Henceforth it should be a total war against indiscipline’ (p. 33). This particular play was performed by the state company as a Command Performance for the State Governor during the campaign ‘War Against Indiscipline’. Another play also makes use of familiar narrative and characters, but in quite a different way. Called ‘Sons of Akpe’ by Boniface Leva (performed in 1982), it tells the story of the feuding sons of Akpe and how the least prominent, educated, travelled, rich or physically impressive wins through to gain the hand of the Chief s daughter in marriage. Leva is a skilled and talented writer and performer, and his ability with words is evident in the play, as is his ability to weave a complex plot into a satisfactory narrative form. The basic story is the one which Abraham gives in his collection in The Tiv People —Abraham calls it ‘The Hare and the Lion’: Once upon a time, there was a chief who had a beautiful daughter and all the animals including the hare wanted to marry her. The girl wanted to marry the hare and all was going well until the lion strutted in saying that he had come to claim his bride, the chief s daughter. The hare saw that this was a case calling for cunning and turning to the lion suggested that they ought to make a wager together, the winner to have the girl. The lion agreed but strutted off without waiting to hear the terms, for he thought that he could beat the hare in anything. It was agreed with the Chief that the winner would be the one who came mounted as on horseback to his father-in-law’s marriage dance. When the day for the festival drew near, the hare feigned illness, lay down near the fire so that all his body became covered with ashes. Soon afterwards the lion arrived and said to the hare, ‘Come let us go to the dance.’ The hare replied that he was too ill to go although he would like to. [In Abraham’s version, the lion then offers to carry the hare on his back, hare agrees, asks to put a seat on his back so as to be more comfortably seated and then to put a bridle in his mouth, so as to have something to hold onto.] On arriving at the chief’s compound to the sound of ‘loud drumming’ by the hare’s retainers, the lion realised he had been tricked, attacked the hare, but the hare hid in the bushes, the lion was ‘nonplussed’ and lost the bet. So ends the story. (p. 65) Leva’s play ‘Sons of Akpe’ opens with the elder brother, Nyakyor, who is in the role of the lion, being confronted by his younger brother Alokyor, on whose behalf he is supposed to have asked for the hand of Monik, the chief’ s daughter, in marriage. Instead, he has arranged that he himself will be the one to marry Monik. The elder brother is a senator and wants to marry the young woman in order to seal a mutually beneficial political and economic relationship between himself and her father, the village chief. In this relationship, Leva introduces Tiv ‘unity’—one of the recurrent issues of Tiv political life—that is, the Tiv as a united political force in national politics. Referring to the central Council of chiefs, the jirtamen, a powerful representative body, the chief says:
128
FRANCES HARDING
If we agree now to revive the Ijir Tamen things won’t be the same again. No, if the unity of Tiv is not to our advantage we shall keep them divided to rule. The Ijir Tamen is too formidable a political force. (p. 17) In contrast to this, the third son of Akpe, Ubekyor, discusses with another senator the potential of kwagh-hir as a commercial enterprise, bringing development in its wake. The senator says: We saw a need to provide good roads and health centres in our local districts. But the means to meet this end, money, was lacking. It is a people that cause their own development. The kwagh-hir is a popular art form and attracts a big international audience…centres rotating (between) the various districts. It will bring not just money to fund projects. Trade. Social interaction. New discoveries. New inventions. The art will grow and the society with it. (p. 15) The playwright has made explicit a distinction between a ‘good’ senator who puts the welfare of the people first, and a ‘bad’ senator who puts his own interests first. He has also been careful not to put the old ways, represented by the reference to the jirtamen, in opposition to the new ways, represented both by kwagh-hir and the use that can be made of it. Ubekyor sees the competition between his two brothers as representative of much more than a personal quarrel: ‘…but between man and society, force and justice, truth and evil, life and death… This is what we shall do’ (p. 19). He then gives instructions to everyone. Like the lion in the story, the audience does not hear details of the plan. The plan comes to fruition on the following day when the scheming brother, Nyakyor, comes to collect his junior brother, Alokyor, who will lead him with drummers and singers to the wedding celebrations. Alokyor appears to be too ill even to get up and greet his brother. Nyakyor begs him to try and walk to the wedding place. He makes several attempts but only falls down. His young son tries to carry him on his back but he slides off. By this time the audience, knowing that the bridegroom has to be accompanied by his brother, and knowing the story of the hare and the lion, realises gleefully what is about to happen. Finally in desperation, as Alokyor once more falls to the ground apparently in pain, Nyakyor says: ‘Let me carry you on my back.’ Alokyor is seemingly aghast at this suggestion and reminds his brother that he is: ‘A member of the House of State, bridegroom and most respected person at the gathering.’ Nyakyor insists however and, apparently reluctantly, Alokyor agrees. The outcome is predictable. Alokyor rides in triumph on his senior brother’s back into the wedding ceremony and wins the bride. Nyakyor, his self-centred plans in ruins, is publicly disgraced as is, to a lesser extent, the chief, who shamelessly used his daughter as a pawn in a power game. The play is rich in literary and political material. Plays such as these, and there are many, which respond to and reflect the contemporary political reality, are the current heirs of the Tiv storytelling form. Aesthetically and theatrically, they draw on the existing forms as much or as little as they choose to, and are neither devoid of imagination nor tied helplessly to ‘ritual and tradition’. Dorward (1971) presents Abraham as a moderating and mediating voice in the accepted view of the Tiv as the antithesis of the cultured, thinking person and situated in the history of the administration of the area at the time. That may have been the case, but it remains difficult to see his work as other than negatively presenting the Tiv. After all, this is a time which revealed Akiga Sai as an acute and poetic observer of his own people. Abraham’s ethnography did nothing to alter substantially the attitudes of the colonial government towards the Tiv other than to insert a patronising tolerance of them. As he says:
R.C.ABRAHAM AND ‘THE TIV PEOPLE’
129
We should…be wrong in attributing to the Tiv a mentality altogether inferior to the surrounding peoples, and may rely on Keane’s dictum that ‘the Bantu are much lower in culture than the Negroid Sudanese, but higher than the true Negro.’ (p. 32) Can this negative concession really be considered as a positive contribution to understanding? In his assessment of the Tiv verbal arts, Abraham saw storytelling as an undeveloped form, a sort of prototype of the novel. He interpreted the form as bounded by words and found it thin in consequence. The fact that narratives were transmitted orally relegated them, in Abraham’s perception, to something less than the written word, but orally transmitted words bring enriching factors to bear other than just those which the printer’s ink can provide. Storytelling provided the carnival element in a life bounded by strict canons of behaviour imposed by elderly men, articulated in terms of tsav, and maintained by the threat of supernatural sanctions. Storytelling was one forum among many in which Tiv political and social and religious discourse took place. This is clearest in the rise of the kwaghhir as the preferred form of political opposition during the crisis years of the sixties, but is also evident in the works of the Tiv writers who articulate political concerns in written plays. Storytelling is a conceptual space in Tiv society which can be used to account for the unexpected and to articulate the desirable and attainable—‘We have hung our bags where we can reach them.’ R.C.Abraham’s own revealing words lead one to think that perhaps it was he who had a ‘dread of change’, not the Tiv; that he sought somewhere a ‘freer field for the exercise of his imagination’, just as Tiv people found it and continue to find it in the performance arts. Abraham’s own conservativism limited his vision so severely that he could see only what he thought he should be seeing. His ethnography of the Tiv people stands as a testament to a colonial figure—truly a man of his time. REFERENCES Abraham, R.C. 1940. The Tiv People. 2nd ed. London: Crown Agents for the Colonies (on behalf of the Government of Nigeria). Ahura, Tar. n.d. Tor Ya Tar. [Unpublished typescript, play performed 1989.] Akpede, Dennis. n.d. Preserve our Culture: Square One. [Unpublished typescript, play recorded 1984.] Armstrong, Robert G. 1964. Roy Clive Abraham, 1890–1963. The Journal of West African Languages 1(1): 49–53. Bergsma, Harold M. 1970. Tiv proverbs as a means of social control. Africa XL: 151–163. Bowen, Elenore Smith (Laura Bohannan). 1964. Return to Laughter. 2nd ed. New York: Doubleday. Dorward, David Craig. 1971. A political and social history of the Tiv people of northern Nigeria, 1900–1939. PhD thesis, University of London. Downes, R.M. 1933. The Tiv Tribe. Kaduna: Government Printer. Keil, Charles. 1979. Tiv Song. Chicago: University of Chicago Press. Kyoive, Chris. n.d. A Close Shave. [Unpublished typescript, play performed 1984.] Leva, Boniface. n.d. Sons of Akpe. [Unpublished typescript, play performed 1982.] Sai, Akiga. 1939. Akiga’s Story. London: Oxford University Press. [Translated and annotated by Rupert East.]
LINGUISTIC STUDIES ON TIV—BEFORE, BY, AND AFTER R.C.ABRAHAM Heinz Jockers
1. Introduction Tiv is the language spoken by most people in Benue State, Nigeria. They form the eastern part of the State with six Local Governments. A small group of Tiv-speakers live in neighbouring Cameroon. Large groups live in Plateau State, where they have expanded to the region of Lafiya, and in Gongola State, where they can be found farming up to Mutum Biyu. In areas like Wukari, known to be the centre of the Jukun, they form the majority. They expand via farming and are farmers by passion. There are an estimated three million Tiv-speakers. Tiv people, if they speak another language at all, use Pidgin (Cross-River dialect) or English, but rarely another African language. Other languages in Benue State are Idoma and Igala (each with about one million speakers). There are also a lot of minority language-speakers who very often use one of the big three languages of the State as a lingua franca, for example, people in Bassa-Komo, Bassa-Nge, Igedde, Etulo, and the Hausa (the so-called Abakwariga) and Jukun-enclaves like Abinsi, as well as groups found under ‘Unresolved Queries’ in Hansford et al. (1976). It seems to be generally accepted that there are no dialects in Tiv, or that the dialect differences are only minor. However, there are at least five distinct dialects which differ from each other in lexical items, use of suffixes in the noun class system, apocopy, tonal progressive or regressive assimilation etc. I used to have discussions with my Tiv students about ‘proper Tiv’, and a phrase like ‘my father’ produced six different translations depending on the area: tere-u, ter-ou, terugh, tere-ugh, ter-u, and ter-ough. I would get more versions taking tonal differences into consideration. Then there is the language of the elders which is not understandable to most Tiv people. The central dialect (Gboko area) which the Mission used for their publications is generally accepted as ‘Standard Tiv’.
African Languages and Cultures, Supplement 1 (1992):163–173
R.C.ABRAHAM AND LINGUISTIC STUDIES ON TIV
131
2. Early works on Tiv The first words in Tiv were published by Clarke (1848) under the term ‘Apa’ (also ‘Appa’); but only the items numbered 67, 68, and (partly) 101, 120, and 123, are in fact Tiv. The other items given under these last numbers are Jukun. A few years later, several publications appeared which gave information about the Tiv and their language. Koelle (1854) got his information from two ex-slaves who described themselves as being ‘Tiwi’. His wordlist has about 300 entries. In 1855, Samuel Crowther, the famous bishop of the Niger, published his Journal of an Expedition up the Niger and Tshadda Rivers in 1854. In one of the appendices we find a collection of about 300 Tiv words under the heading ‘Mitshi’.1 T.J.Hutchinson gave his account of the same expedition in two publications (1855, 1858). W.Baikie (1856), also a member of that party, who was responsible for the first British settlement inside the country near the Niger-Benue confluence, had also been in contact with the ‘Mitshi’. All three writers give some information about the Tiv, usually very scanty and mainly based on the judgements of neighbouring groups: ‘Cannibals, wild, less intelligent than any other African race’. This picture held for many years. The reason was mainly that the Tiv could successfully hinder any foreign penetration of their country for a long period. Travellers up the Benue were usually very happy when they had passed the Tiv area without incident. It was Leo Frobenius (1924) who first saw them in a positive light, but this was with the background of the ‘strong and noble savage’. He compared them with the Jukun, whom he saw as a declining people. In 1908, Elphinstone Dayrell, working for the colonial administration in Southern Nigeria and based in Ogoja, published a vocabulary in Jukun and Tiv for use in the colonial administration. It was published in Zungeru, the administrative centre for the Northern region to which Tivland, with the exception of a small region, had been added. But the penetration started in the South. In literature, the Tiv were generally referred to as ‘Munshi’, which is a Hausa derogative term for them. There are many unfortunate stories about the origin of that word. Due to Tiv protests it was discarded in favour of their own designation. But it was as late as 1927 that the Education Department of the Northern Provinces published a Munshi Reader, written in Tiv to be read by Tiv. In 1911, the South African Branch of the S.U.M. (Sudan United Mission) —from 1916 figuring as the D.R.C.M. (Dutch Reformed Church Mission)— opened a station at Sai in Tivland, on the border with the Jukun area. They came via Wukari, after having tried to open a station in Mbula (the Muslim region) without success. They now wanted to try again amongst the Tangale (further north, now in Bauchi State), but the colonial administration did not allow this. By chance their eyes fell upon the Tiv (who themselves saw it as a divine hint), and they started immediately to translate portions of the Bible, the first publication in Tiv appearing in 1914. The first study of the language itself was done by one of their members, A.S.Judd (1916/17). The chief of Sai gave one of his sons—Akiga Sai—to the Mission in order to have him trained there. The actual reason was that the son was a cripple and of hardly any use on the farm, but the boy turned out to be very intelligent and a year later he was able to write his language. He also became interested in the history and traditions of his people, as well as the colonial administration of the Tiv. He studied and collected notes on these subjects for about twenty years before he had finished his undated manuscript ‘History u Tiv’. About 20 per cent of the whole manuscript was eventually translated and annotated by Rupert East and
1 It was actually William Bleek (1855), the German librarian and first Bantuist, who pointed out that Koelle (1854) and Crowther (1855) had noted the same language.
132
HEINZ JOCKERS
published in 1939. Another chapter has been translated by P.Bohannan—‘The descent of the Tiv from Ibenda Hill’ (1954). Robert G. Armstrong (1917–1987) was working on a new edition but unfortunately he did not have access to the whole manuscript. Akiga Sai was later transferred to Zaria where he became responsible for Tiv publications at the Literature Bureau, now N.N.P.C. (Northern Nigerian Publishing Company). He was later dismissed by the Mission because, in spite of all the Christian teaching he had received, he was after all a Tivman. As the Mission put it: ‘Akiga was a leading Christian until the latter years of his life when prosperity and the Tiv’s polygamous tradition caused him to backslide’ (Anon 1961: 31). In 1931, the Rev. W.A.Malherbe, chief linguist at the Mission, published the first serious study of the language—Tiv-English Dictionary with Grammar Notes and Index. The ‘Grammar Notes’ include 35 pages followed by three pages of ‘Spelling Rules’. The Dictionary itself totals 207 pages, the last 37 pages being an index of all the English words which appear as a guide for the equivalent Tiv words. Malherbe is responsible for /ô/ [ ], which is still used in Tiv writings, though the Nigerian recommendation is /ọ/. But since the Mission is the greatest publisher itself and never uses tone-marking, even in their grammars or dictionaries, not to mention their readers, this is not a problem for them. Malherbe is also responsible for the writing of the diphthong /ou/ which is actually an allophone of /o/. Even Robert Armstrong could not convince the ‘Ad-Hoc Committee on Tiv Language’ (1980) to change, and they insisted on writing the diphthong ‘since it could be heard’ (p. 32). Malherbe failed to distinguish the nasal of pre-nasalised consonants from nasal syllabics. He noticed ‘combinations of consonants’ but could not organise them into a system. He was aware of the importance of tone in Tiv but only occasionally does he mention it—this meant that he was only able to distinguish nine noun-classes, since some differ only in tone. He also described the Tiv verbal system as a tense-system where many forms are identical. Even if Malherbe’s analysis was fragmentary, it was nevertheless a good starting point for R.C.Abraham. The colonial administration in Tivland started in 1913 when the first colonial administrator arrived, but it took another year before the occupation of Tivland was completed. Right from the beginning, the administration had the problem of getting into direct contact with the people. For a long time they were using Hausa-speakers (Abakwariga) as interpreters. Up to 1932 Abinsi, the Jukun enclave at the Benue, was the Tiv Divisional Headquarters. The Yoruba chief of Makurdi was the main spokesman for the Tiv for several years. It was a very difficult area to administer and the colonial period is full of riots, outbreaks of epidemic, diseases and hunger, catastrophes caused by the administration, executions of chiefs, judicial errors, etc. In 1916 G.P. Bargery, working for the colonial Education Service since 1910, was sent to Tivland where he started to work on the language and in 1915 he opened the first government school in Wannune, which was later transferred to Katsina-Ala. Unfortunately, all his notes and writings were burnt when he was on an inspection tour. 3. Abraham and Tiv studies Abraham was posted to Tivland at the beginning of the 1930s. In 1933 his voluminous The Grammar of Tiv was published. It consists of three parts: (1) ‘The Grammar’; (2) ‘Texts in Tiv’; (3) ‘Tiv Vocabulary’. Abraham must have known Malherbe’s work quite well since he thanks him in the ‘Preface’ for his ‘help and valuable suggestions’. But he also says what he is thinking about Malherbe’s work: ‘no previous grammar of the language exists’. His judgement on Malherbe’s (1931) Dictionary can be seen in his comment in the ‘Preface’: ‘the whole basis of the language rests on the tone system and if neglected chaos will result.’
R.C.ABRAHAM AND LINGUISTIC STUDIES ON TIV
133
Rather surprisingly for a grammar about a noun-class language, he begins with ‘The Verb’ (p. 1), but this is in fact appropriate for Tiv: ‘it must be said at the outset that the verb is the most interesting part of the language and at the same time, the most difficult.’ It also contains the greatest part of the grammar with 54 sections, leaving nine sections for the rest. The second part consists of texts drawn from different publications of the D.R.C.M. All the texts are tone-marked, and Abraham distinguishes between ‘H’ (High tone), ‘M’ (Mid tone), ‘L’ (Low tone), ‘Fa’ (Falling tone which he calls ‘glissando’), and ‘Ri’ (Rising tone). The third part is a ‘Tiv-English’ vocabulary which has been arranged according to the initial consonant. Malherbe had arranged his Dictionary according to the first letter of a word. Abraham states that his aim was ‘to provide a guide to the commoner words of the language with their tones, not to write a dictionary, this already existing in Malherbe’s forthcoming work’ (‘Instructions for Using the Vocabulary’). His guide contains 213 pages with hardly less information than that given by Malherbe. Seven years later, the three parts were published as three different works in revised editions. It is to these three publications that Armstrong is referring when he writes: ‘Abraham’s Tiv books are a tremendous achievement and are in fact the foundation of modern Tiv studies’ (1983:48). Abraham’s Dictionary is still the only one in use, and the collected texts have been expanded with his own material. Abraham was very accurate in his representation of tone. In his ‘Introduction’ to the 2nd edition, Arnott (1968) —with the knowledge of about 30 years in linguistics—provides a short summary of new possibilities in describing a tonal system like the one in Tiv. Abraham himself was at times unsure about the correct analysis of some phenomena, as can be seen from such imprecise statements as: ‘sometimes, when…’; ‘often a change of form and tone takes place’; ‘in many cases there is a change of form’; ‘the following are irregular…’ etc. He was not able to find an explanation for everything he encountered in the language, but he nevertheless noticed everything. The result is that only a few problems remained for later scholars to address. Abraham also announced three other works on Tiv due to be published in the same year (1940): ‘A Short Classified English-Tiv Dictionary’, ‘Tiv Phonetics and Tonal Principles’, and ‘The Bantu Features of Tiv’. None, however, were actually published, although the latter two are available in manuscript form at SOAS. Already in 1933 he had published his anthropological study The Tiv People. He is also responsible for one subject which, so it seems to me, is of vital interest to the Tiv people. He compared a wordlist of Nyanza in Malawi with the corresponding Tiv items, and concluded that there can be no doubt that Tiv was a Bantu language. It has, in the meantime, become an oral tradition amongst the Tiv that they originate from southern parts of Africa (Congo, Zululand etc.), and some serious books and papers have been written about this (see the publications by Gbor in Gundu and Jockers 1985). Abraham was subsequently awarded a D.Litt by the University of Oxford for his work on Tiv. J.Lukas based his article ‘Das Nomen im Tiv’ (1952) on Abraham’s material. He concluded that the nounclass system was closer to the ‘Gurma-type’ than Bantu, as Abraham had postulated. But Lukas relied on typological features while Abraham made a lexical comparison. Lukas pointed out that the affixes in the noun-class system and their corresponding demonstratives are remnants of a system where the forms of the suffixes were identical with the basic forms of the demonstratives. This typological comparison was later taken up by Greenberg (1977, 1978) for the whole Niger-Congo group. The Rev. Terpstra’s English-Tiv Dictionary (1959), as well as his A Tiv Grammar (1968), are based on Malherbe’s and Abraham’s publications and do not contain new insights into the language, although they do present many examples. The three articles by Arnott deal with single aspects of the language. In ‘The classification of verbs in Tiv’ (1958), he tried to simplify the verbal classification as proposed by Abraham. Both took the ‘pastform’ and the tonal structure of the verbs as the basis for their groupings. Abraham also took the number of syllables into consideration, while Arnott relied on the phonological structure and change of vowels.
134
HEINZ JOCKERS
Abraham identified eight verb-classes while Arnott had six but had to subdivide them into two classes. But Arnott provided a much better overview and clarified the different tense-aspect variations. In his ‘Downstep in the Tiv verbal system’ (1964), he dealt mainly with Abraham’s Mid tone (M). Since Abraham did not know the principle of a terraced-level system with Downstep, he introduced a M or, as can be seen in his unpublished manuscripts, lowered Hi tone. At least Abraham was aware that the tones between Hi and Lo were not all at the same level as the M. In ‘Some reflections on the content of individual classes in Fula and Tiv’ (1967), Arnott tried to find semantic categories for noun-classes. While this is possible in Fula it was not possible in Tiv, since ‘the nouns in the different classes are extremely heterogeneous’ (p. 61). Welmers (1959) identified the tonal system of Tiv as a terraced-level system in which Abraham’s Mid is analysed as a downstepped Hi. Arnott (1964) agreed in general, though he points out that Abraham’s Mid is not in any case the lowering of a basic Hi tone, but may sometimes represent the raising of a basic Lo tone. McCawley (1970) uses Arnott’s material to show that Downstep in Tiv arises from a deleted Lo which in underlying structure is part of a tone sequence on single syllables; as such they are realised as Mid tones. Carl Hoffmann, at the time a professor in the Department of Linguistics and African Languages at the University in Ibadan, started to give lessons on Tiv from 1973 onwards, and his (unpublished) seminarpapers became quite well known. He pointed out (1980) that tonal modifications in Tiv are due to ‘floating tones’ which owe their existence to apocope, a common process in Tiv. These ‘floating tones’ influence the tonal shape of the neighbouring morphemes. Other tone features are Downdrift (also mentioned by Arnott 1964), which is described as a lowering effect of a final Hi in a Hi-sequence; Falling and Rising tones (seen as phonetic realisations of segmental tones with a following ‘floating tone’); and Downstep which is analysed as a floating Lo at the phonemic level. Leben (1980) used Tiv (examples from Arnott) as one of the test languages in the generative approach to phonology. The separation of tones from segmental features and the provision of ‘suprasegmental’ features was expanded by Goldsmith (1979) into what he called ‘autosegmental’ features. Sibomana (1984) took over Hoffmann’s analysis: the ‘floating tones’, realised on the following or the preceding syllable, produce lengthening or contour tones on the receiving syllable (p. 277). He calls this ‘tone-clustering’, i.e. several tones are realised on one syllable. The displacement of tones is in most cases the explanation for the Downstep as explained by Arnott for example; a Mid tone is rarely interpreted as lexical, and even when this is the case it is not seen as distinctive (Sibomana 1980). Sibomana’s papers are another landmark for linguistic studies on Tiv. Brilliant analyses of lexical tones in Tiv were proposed by Pulleyblank (1985, 1986), where he introduced new tone rules like ‘Hi-spreading’ and ‘cyclic tone association’. Jockers’ Studien zur Sprache der Tiv in Nigeria (1991) serves as the reference-grammar for Tiv. Independent from the effects of Downstep, he identifies a synchronic Mid tone which is the result of a Floating Lo realised on a Hi syllable. Jockers also proposed a new arrangement for the verbal system. 4. The Tiv grammarians The first Nigerian and mother-tongue speaker of the language to take a degree in the linguistic analysis of Tiv was Nyiakura at Ahmadu Bello University Zaria in 1974. Unfortunately he did not pursue his academic studies but went into the Civil Service where there was, and still is, a great demand for university graduates. In the meantime a number of BA theses and ‘Long Essays’ have been written about the Tiv language, mainly at the University of Jos.
R.C.ABRAHAM AND LINGUISTIC STUDIES ON TIV
135
In 1974, Iyenge started to publish about Tiv in Tiv. Best known among his many publications are his Tiv Grammar (n.d.), written for primary school children, and his reader Ruamabera U Jimin (1983). All his publications are written in Tiv which makes it difficult to take them into consideration in the linguistic field. Every Nigerian government has had a kind of ‘National Language Policy’ in which strong demands were made for a ‘Mother Tongue Education’, and the progress of the twelve most widely-spoken Nigerian languages in particular. Tiv is one of them, but in spite of this not much progress has been made. In 1979, an ‘Ad-Hoc Committee on Tiv Language’ was appointed under the chairmanship of Nyiakura. But for Robert Armstrong, who had been asked by the Federal Ministry of Education to make a report on Tiv orthography, this committee would probably never have been formed. Their task was to prepare a ‘final orthography’ in order to facilitate the linguistic study of, and publications in, Tiv. They failed, though in 1980 they produced a quite positive report which is known only to a few insiders. And with the military take-over in 1983, the committee ceased to exist. They failed because they could not stimulate sufficient interest among students, and the field was left instead to linguistic amateurs and the churches. But this was only a mirror of the Committee itself, since only Nyiakura was linguistically trained. The others may have been enthusiastic and highly valued, but they were simply not up to the task. this enthusiasm and desire to get publications into the school-syllabus was also the reason for much of the publications from the 1970s up to now, not to mention all the unpublished manuscripts. 5. Priorities in Tiv studies It has not been possible to publish a complete edition of Akiga Sai’s ‘History u Tiv’, neither a Tiv version nor a translated version. It has not been possible to publish the theses of Makar (1975) and Orkar (1979) about the history of the Tiv. It has not been possible to publish the papers from the ‘Tiv Symposium on Tiv Literature and Culture’ in Katsina-Ala (1985), a real highlight in Tiv studies. It has not been possible to publish enough literature in Tiv to render the N.K.S.T. (the independent Tiv church, successor to the Mission) literature superfluous, so that it can be restricted to the church and its institutions. It has not been possible to persuade the ‘old grammarians’ to do useful and much-needed work on oral literature and translation for example, instead of wasting their time dealing with complicated linguistic questions. It has not been possible to establish Tiv properly in the schools according to the ‘National Language Policy’. It has not been possible to arrange proper training for teachers in their mother tongue. It has not been possible to revise and publish the remaining volumes of Hen Tiv Dedoo, a series of primers by Gosough Ikpa (1981, 1984). 6. Some comments on Tivoid When we speak of ‘Tivoid’ we are actually referring to Tiv, since hardly anything is known about the other supposed members of the group. In the article by Watters and Leroy (1989), where Tivoid forms one of the subgroups of the 19 languages described, data on Tiv and Esimbi are very limited (see also Stallcup 1980). In the Atlas linguistique du Cameroun (1983), Coupez is quoted as having made a lexicostatistic comparison which clearly shows that Esimbi is Tivoid. Watters and Leroy (1989) divide the Tivoid languages into the following geographical regions: (a) Nigeria: the relationship of the languages in the Baissa area to Tiv goes back to a statement by Meek (1931:551ff.). Meek’s impression was confirmed by Koops (1971), but Koops also relied on an impression
136
HEINZ JOCKERS
though he knows Tiv quite well. Comments on the remaining languages are based on impressionistic information from Jockers (p.c.). (b) Nigeria and Cameroun. (c) Cameroun: the relationship of the languages mentioned in these two groups to Tiv is based on an unpublished lexicostatistical similarity matrix by Breton and Nsémbé, which ranges from 1.6% to 76% (Atlas linguistique du Cameroun, p. 434). It is quite obvious that more published research is needed. This situation is surprising since Tiv or Tivoid holds a key position in the classification of Bantu languages, ‘not-so-Bantu’ languages, Bantoid languages, semi-Bantu languages, Benue-Congo languages and so on. REFERENCES Abraham, R.C. 1933a. The Tiv People. 1st ed. Lagos: Government Printer. [Revised and reprinted in 1940, Crown Agents for the Colonies, London.] ——. 1933b. The Grammar of Tiv. Kaduna: Government Printer. ——. 1940a. The Principles of Tiv. London: Crown Agents for the Colonies. [Reprinted in 1968, Gregg International Publishers, Farnborough, England.] ——. 1940b. A Dictionary of the Tiv Language. London: Crown Agents for the Colonies. ——. 1940c. A Tiv Reader for European Students. London: Crown Agents for the Colonies. ——. 1940d. Tiv phonetics and tonal principles. Ms., SOAS, University of London. 1940e. The Bantu features of Tiv. Ms., SOAS, University of London. ACCT, CERDOTOLA, DGRST. 1983. Atlas linguistique de l’Afrique centrale; Atlas linguistique du Cameroun. Yaoundé: Orstom. Ad-Hoc Committee on Tiv Language [1980 Report.] Makurdi: Benue State Government. Anon. 1927. Munshi Reader. Kaduna: Government Printer. ——. 1941. Dzua. D.R.C.M. ——. 1961. The Coming of the Bible into Tivland. Mkar: Sudan Interior Mission. Armstrong, Robert G. 1983. Tiv orthography. In Orthographies of Nigerian Languages (Manual II), ed. by K.Williamson, pp. 46–69. Lagos: Heinemann Educational Books. Arnott, D.W. 1958. The classification of verbs in Tiv. BSOAS XXI:111–133. ——. 1964. Downstep in the Tiv verbal system. African Language Studies V:34– 51. ——. 1967. Some reflections on the content of individual classes in Fula and Tiv. In La classification nominale dans les langues négro-africaines, ed. by M.G. Manessy, pp. 45–74. Paris: Editions du Centre National de la Recherche Scientifique. 1968. Introduction to R.C.Abraham’s The Principles of Tiv. Farnborough, England: Gregg International Publishers. Baikie, William B. 1856 [1966]. Narrative of an Exploring Voyage up the Rivers Kwora and Binue in 1854. London: John Murray. [Reprinted in 1966, F. Cass, London.] Bendor-Samuel, John (ed.). 1989. Niger-Kordofanian-Congo Language Family. Leiden: Foris Publications. Bleek, William. 1855 [1970]. Note by Dr William Bleek. In Crowther [1855] 1970: 234. Casaleggio, E.N. 1964. The land will yield its fruits (Lev. 25:19). [Ms. in Mkar with the Mission, a translation from the Afrikaans ‘Die Land Sal Sy Vrug Gee’, Cape Town.] Clarke, John. 1848 [1972]. Specimens of Dialects. Berwick-upon-Tweed: B.L. Green. [Reprinted in 1972, Gregg International Publishers, Farnborough, England.] Crowther, Samuel A. 1855 [1970]. Journal of an Expedition up the Niger and Tshadda Rivers in 1854. London: C.M.S. [Reprinted in 1970, F.Cass, London.] Dayrell, Elphinstone. 1908. Vocabulary and Supplementary Vocabulary of English Words, Salutations, etc., Translated into the Jukun and Munshi Languages. Zungeru: Government Printer.
R.C.ABRAHAM AND LINGUISTIC STUDIES ON TIV
137
Dorward, David Craig. 1971. A political and social history of the Tiv people of northern Nigeria, 1900–1939. PhD thesis, University of London. Frobenius, Leo. 1924. Die Muntschi. Ein Urwaldvolk in der Nachbarschaft der sudanischen Kulturvölker. Atlantis XI: 237–352. Jena: Eugen Diederichs. Gbor, J.W.T. 1978. Mdugh U Tiv Man Mnyer Ve Ken Benue. Zaria: Gaskiya Corporation. [A revised version of the author’s BA thesis Traditions of Tiv origin and migrations with special emphasis on the Eastern Frontier to c. 1900’.] Goldsmith, John. 1979. Autosegmental Phonology. New York & London: Garland Press. Greenberg, J.H. 1977. Niger-Congo noun class markers: prefixes, suffixes, both or neither. Studies in African Linguistics, Suppl. 7:97–104. ——. 1978. How does a language acquire gender markers? In Universals of Human Language, Vol. III, ed. by J.H.Greenberg, pp. 47–82. Stanford: Stanford University Press. Gundu, G.A., and H.Jockers. 1985. Tiv Bibliography. Makurdi: Government Printer. Hansford, Keir, John Bendor-Samuel and Ron Stanford. 1976. An Index of Nigerian Languages. Ghana: SIL. Hoffmann, Carl. 1980. Apocope and floating tones in Tiv. Ms., University of Hamburg. Hutchinson, Thomas J. 1855 [1966]. Narrative of the Niger, Tshadda and Binue Exploration. London: Longmans, Green & Co. [Reprinted in 1966, F.Cass, London.] ——. 1858. Impressions of Western Africa. London: Longmans, Green & Co. Ikpa, Gosough. 1981. Hen Tiv Dedoo. Lagos: Macmillan. ——. 1984. Hen Tiv Dedoo 1–3 (A Teacher’s Guide). Makurdi: Government Printer. Iyenge, Ph. Aii. n.d. Tiv Grammar. Mkar: Gar’s Correspondence Service. ——. 1983. Ruamabera U Jimin. Nigeria: published by the author. Jockers, Heinz. 1991. Studien zur Sprache der Tiv in Nigeria. Frankfurt: P.Lang. Judd, A.S. 1916/17. Notes on the Munshi tribe and language. Journal of the African Society XVI:52–61. Koelle, S.W. 1854 [1965]. Polyglotta Africana. London: C.M.S. [Reprinted in 1965, Akademische Druck- und Verlagsanstalt, Graz.] Koops, Robert. 1971. Linguistic survey of the Baissa area. Ms., University of Jos. Leben, William R. 1980. Suprasegmental Phonology. New York & London: Garland Press. Lukas, Johannes. 1952. Das Nomen im Tiv. Anthropos IIIL:147–176. Makar, T. 1975. A history of political change among the Tiv in the 19th and 20th centuries. PhD thesis, Ahmadu Bello University, Zaria. Malherbe, W.A. 1931. Tiv-English Dictionary with Grammar Notes and Index. Lagos: Government Printer. McCawley, J.D. 1970. A note on tone in Tiv conjugation. Studies in African Linguistics 1:123–129. Meek, Charles K. 1931. Tribal Studies in Northern Nigeria. London: Kegan Paul, Trench and Trübner. Nyiakura, Joseph Orban. 1974. A contrastive analysis of Tiv and English syntactic patterns. MA thesis, Ahmadu Bello University, Zaria. Orkar, John Ngusha. 1979. A pre-colonial history of the Tiv of central Nigeria c. 1500–1850. PhD dissertation, Dalhousie University. Pulleyblank, D.G. 1985. A lexical treatment of tone in Tiv. In African Linguistics: Essays in Memory of M.W. Semikeke, ed. by D.L.Goyvaerts, pp. 421–476. Amsterdam: Foris. ——. 1986. Tone in Lexical Phonology. Dordrecht: Reidel. Sai, Akiga B. 1939. Akiga’s Story. London: Oxford University Press. [Translated and annotated by R.East.] ——. 1954. The ‘descent’ of the Tiv from Ibenda Hill. Africa XXIV:295–310. [Translated by P.Bohannan.] n.d. History u Tiv. Ms. Sibomana, L. 1980. Aspects of Tiv morphophonemics. Afrika und Übersee 63(1): 69–77. ——. 1981. Tonal lexicalization of Tiv nouns. In Festschrift zum 60 Geburtstag von P.Anton Vorbichler, ed. by I.Hofmann, pp. 155–171. Wien: Institut für Afrikanistik und Ägyptologie der Universität Wien. ——. 1983. Tonal structures of Tiv verbs. Afrika und Übersee 66(1):149–158. 1984. Displacement and lowering of Tiv tones. Afrika und Übersee 67:277– 287.
138
HEINZ JOCKERS
Stallcup, Kenneth L. 1980. Noun classes in Esimbi. In Noun Classes in the Grassfields Bantu Borderland, ed. by L.Hyman, pp. 139–154. Los Angeles: University of Southern California. Terpstra, G. 1959. English-Tiv Dictionary. Gboko: Sudan Interior Mission. [Reprinted in 1968, Ibadan University Press, Ibadan.] ——. 1968. A Tiv Grammar. Mkar: Sudan Interior Mission. Watters, J.R., and J.Leroy. 1989. Southern Bantoid. In Bendor-Samuel, pp. 430– 449. Welmers, William E. 1959. Tonemics, morphotonemics and tonal morphemes. General Linguistics 4:1–9. Williamson, Kay. 1984. Practical Orthography in Nigeria. Ibadan: Heinemann Educational Books.
R.C.ABRAHAM: A LINK BETWEEN TWO PERIODS IN THE YORUBA WRITTEN TRADITION* J.Gbenga Fagbọrun
1. Introduction The publication in 1958 of Abraham’s Dictionary of Modern Yoruba marked the end of one era and the beginning of another in the linguistic history of Yoruba. It was produced at a time when Yoruba was about to be given new roles to play, including the beginning of serious academic works in and/or on Yoruba. The book itself has been sympathetically reviewed by Armstrong (1959) and Olmsted (1959), but apart from these two scholars, nobody has ever considered the light the book throws on grammatical analysis and the understanding of Yoruba linguistics. R.C.Abraham was a linguist who has never been placed where he belonged. The reason is clear: nobody has ever attempted a linguistic history of the Yoruba koiné or what is (1966:2). As observed widely acknowledged as ‘standard’ or ‘common’ Yoruba as presented in by Ward (1952:4): ‘a completely new survey of Yoruba grammar…is very much needed.’ Based on the use to which the Yoruba koiné has been put, the development of the written tradition in Yoruba can be divided into three periods (Fagbọrun 1991:325): (1)
I II III
1843–1899: the early period—basically translation from English 1900–1959: the beginning of imaginative writings and the documentation of oral literature 1960–1990: the beginning of scholarly work in/on Yoruba
The material in R.C.Abraham’s book represents the Yoruba koiné to a recognisable degree, although he claimed that his informants (or ‘collaborators’ as he called them) were all from the dialectal zone. For instance, Yoruba writers are sensitive to the use of sh as a shibboleth, and most dialects of Yoruba (including Ibadan-Ọyọ) do not differentiate /s/ from /sh/. Hence, Abraham’s dictionary, which . distinguishes the two sounds, does not represent
African Languages and Cultures, Supplement 1 (1992):175–185 *
I would like to acknowledge the help of Karin Barber (who first invited me to write a joint paper on The contribution of R.C.Abraham to Yoruba linguistics and literary tradition’), and of John Kelly. These individuals read through the draft of this paper and offered valuable comments, although they should not be blamed for any inadequacies. I should also thank Philip Jaggar for inviting me to participate in the Symposium, in which I first presented this paper, and those participants who provided feedback.
140
J.GBENGA FAGBỌRUN
Apart from his dictionary, which is encyclopaedic, no serious attention has been paid to Abraham as a linguist, although he had some insights that should be studied. The reason seems to be traceable to the linguistic exposition included in his dictionary, which for clarity’s sake needs some evaluation and rearrangement. Those who wrote on Yoruba grammar before him, with the exception of Ward (1952), were no better, although one defect in the dictionary is that many points that should have been treated independently were lumped together under the single heading ‘Tone and Tone Changes’. It is my claim in this paper that Abraham had enough data to write a simple grammar (or even a handbook) of Yoruba which would be acceptable to modern speakers of the language, since the work encapsulates what the majority of Yoruba speakers have internalised. As claimed by Armstrong (1959:90): ‘Abraham has produced a phonology (including the tone system), a morpho-phonology, and a syntax of Yoruba.’ My main concern in this assessment is to examine the way in which modern scholars can make use of Abraham’s data and analyses, many of which were later addressed by Yoruba scholars. 1.1. Lack of acknowledgements One Africanist whose work undoubtedly influenced Abraham’s linguistic analysis was Ward (1952). Both Abraham and Ward produced their work on Yoruba during the last decade of our Period II, as stated in (1). Although Abraham did not make reference to Ward, I will touch on some issues already discussed by her, which Abraham also dealt with. 1.2. Tone and vowel length Ward was the first scholar to provide full phonological marking of tone in Yoruba (using square brackets), and I would like to discuss her representation of tone and vowel length in two contexts: (a) when pronounobjects are used after monosyllabic verbs; (b) when nouns are used as subjects (1952:47). The last non-high syllable of a noun-subject changes to high-tone before a verb, as in (2): (2)
a. b.
ìwọ ‘you’; but òjò ‘rain’; but òjó
dára ‘you are good’ ‘it rains’
(Abraham, p. xix)
If the tone in (2a-b) is not changed, the sentence will be interpreted as an imperative. Abraham observed that there is no such tone-raising before negatives or before any other Auxiliary (p. xx), as (3a) below demonstrates. However, we often hear a raising of tone when certain Auxiliaries are used, as in (3e). Ward (p. 91), who earlier pointed to this tone-raising in affirmative sentences, limited the exception to ASP [+FUT], i.e. not covering all Verbal Auxiliaries, unlike Abraham. We can see that these include ti [+V, +AUX, +PAST], cited in (3c) from Abraham (p. xx). It would be rather misleading therefore, to make a general claim about this rising tonation; it may reasonably be limited to negators and ASP [+FUT], as evidenced in (3f): (3)
a. b. c.
bàbáàmi kò wá bàbáà mí wá ti lọ ?
‘my father has not come’ ‘my father came’ ‘the bird has flown’
R.C.ABRAHAM: A LINK BETWEEN TWO PERIODS IN THE YORUBA WRITTEN TRADITION
d. e.
lọ ti lọ máa sì
f.
* *
kò lọ yóó
141
‘the bird flew’ ‘the bird had escaped’ ‘the bird must escape’ ‘the bird will escape’ ‘the bird did escape’ ‘the bird did not go’ ‘the bird will go’
Other morphophonemic features, which Ward discussed, and further alluded to by Abraham without any ‘truth’ (Ward, p. 15); (b) ‘formation of reference, include: (a) consonant deletion, e.g. verbal nouns’ by reduplication (p. 16), e.g. lílọ ‘the going’
àìrí ‘invisibility’ (p. 17). All the above are discussed under vowels and tone changes by Ward (pp. 83, 93), and subsequently by Abraham (p. xiii). In the ‘Noun-Noun construction’, Ward (p. 55) observes that ‘a glide always occurred before the nouns beginning with a consonant’ if the relationship is genitival, e.g. nínú kànga ‘in the well’, realised as nínúu kànga. Abraham (p. xviii) made a similar claim. Neither of these scholars noticed that the lengthening is not ‘an offspring of a palmwine seller/tapper’, as limited to consonant-initial nouns, as we see in ọmọọ ‘an individual who drinks palmwine in excess’. against ọmọ Ward (p. 28) was the first to reject the use of a tilde (˜) which was formally adopted from the proceedings of the 1875 Y(oruba) O(rthography) R(ecommendations) (CMS Archives, 1875, CA2/096/113), and to represent a gliding tone or the ‘length of vowel’ as in (4). Abraham described the adoption of the tilde as being ‘worse than useless, as it has been used to indicate doubled vowels on four separate levels’ (p. xi): (4)
õrun=/òórùn/ ‘odour’, /oòrùn/ ‘sun’, or /oorun/ ‘sleeping’
The rest of this paper will focus on several issues which Abraham raised, including Yoruba orthography (§2), phonology and morphophonemics in relation to Yoruba writing (§3), and syntax (§4). 2. Orthography 2.1. Tone-marking in Yoruba orthography One major contribution I think Abraham made to the Yoruba writing system was the use of tone-marking. He was the first to employ total tone-marking, indicating the tone directly above a segment, including the mid-tone which should normally be left out. Ward (1952) placed the tones in square brackets, e.g. ẹjẹ [_¯] in Abraham’s transcription). ‘now’ (= The tonal system in normal Yoruba orthography is not as cumbersome as presented by Abraham. He attempted to mark every tone because, according to him, ‘a completely unmarked text causes endless hesitation to Yoruba readers’ (p. x). Most Yoruba scholars have followed this idea, although there has never been any formal recommendation. Nor has there ever been any recorded experimental test to verify Abraham’s claim that total, as against selective, tone-marking aids reading and comprehension. For instance, the Yoruba Bible is not exhaustively tone-marked, and yet most readers find it comprehensible.
142
J.GBENGA FAGBỌRUN
2.2. The use of the hyphen The hyphen, introduced in the ‘Report of the Yoruba Orthography Committee’ by the Western State Ministry of Education (1969, Nigeria) for words such as gan-an ‘really’ and márùn-ún ‘five’, was overused by Abraham. For example, he used the hyphen in some words that have high or low tone as in (5a). This use of the hyphen, as employed by Abraham, is now confined to a mid-tone syllabic nasal which may be confused with the that nasalises vowels, as illustrated in (5b): (5)
a. b.
‘drum’; té-ń-té ‘top’ ‘join with me’; ‘exceedingly’;
‘loosely’ (Abraham, p. x)
2.3. The use of and nasalised vowels Another instance of superfluity in Abraham’s orthographic representation can be seen in his use of after a vowel preceded by a nasal consonant, which the force of progressive assimilation has rendered redundant, as in (6). Logically, Abraham should have written nín ‘to have’ and not ní which he used throughout his dictionary (e.g. pp. 438–441): (6)
‘to know’, written as
; mu ‘to drink’, written as mun
2.4. The use of un for the syllabic nasal Abraham proposed the use of un for the syllabic nasal which is still represented in various ways. He was not consistent in his usage however, as illustrated in (7), although he seems (p. 690) to have admitted the possibility of ún/ùn/un. For example, he used ń as an aspectual marker throughout, and also wrote ‘open space’, not gbaungba (p. 235): (7)
‘stamp’;
‘thirst’ (p. xii)
This use of un for all instances of a syllabic nasal in (7), according to a report by Fagbọrun (1989), will make unnecessary the use of a macron, i.e. , which most Yoruba writers usually leave out. 2.5. The use of an/ọn Abraham tried to remedy some inconsistencies associated with the use of an/ọn and chose to use ọn. Yoruba orthography reformers have never made a strong decision on which one to choose. Both an and ọn were recommended in the YOR (1969), and also in 1974 by the Joint Consultative Committee on Education (JCCE). Both an and ọn have been used inconsistently until now, especially by young writers and students, as investigated in Fagbọrun (1989). We see no justification whatsoever for accepting a dual spelling as displayed in (8):
R.C.ABRAHAM: A LINK BETWEEN TWO PERIODS IN THE YORUBA WRITTEN TRADITION
(8)
àwọn and àwan ‘they’; ìtàn ‘history’
143
‘to be clever’; gbàngàn ‘hall’;
The phonetic reason often given by some scholars for using ọn and an in different domains may not work for non-linguists or those who cannot differentiate labial from velar or alveolar sounds. After all, writing in any language is not meant for linguists alone. 2.6. Word division Although, until the YOR, it was not recommended that the particle [+V, +AUX] should be separated from the following verb, as in (9), Abraham wrote it as a single word. He also noted that is often used as an independent syllable which has its own tone. This idea of lexicalising the verbal auxiliary seemed to be unknown or overlooked before Abraham’s dictionary. (9)
a. b. c.
ó ń 3sg PROG write a ń 1pl PROG hear ó ń sùn 3sg PROG sleep
d.
‘he is writing’ ‘we are hearing’ (p. xxii) ‘he is sleeping’ (Ward, p. 30)
‘I am taking yams’ (Ward, p. 171) 1sg PROG take yams
Abraham used irrespective of the initial consonant of the following verb (e.g. p. 433), unlike Ward (p. 21) who used a ‘double nasal consonant’— and — as we see in (9c, d). 3. Phonology and morphophonemics I will discuss in this section some points that are relevant to our understanding of the Yoruba writing system. These are those that support the notion of phonemic writing in terms of tonal changes and the use of vowel lengthening. Only the issues earlier neglected by scholars or not well discussed will be described here. 3.1. Nominalisation Abraham discussed prefixation with reference to the following items: ì-, a-, àì-, aláì-, and olù-. He also discussed reduplicated verbal nouns. There is little or no difference between Abraham’s treatment and what ‘a Yoruba scholars still construe about nominalising processes. On reduplication, he cited ìkòkò gbì ‘the pot is hot’. Abraham assumed that ‘to be hot’ hot pot’, which is derived from ìkòkò . This is a morphological error which often shows up in the writings of some Yoruba was from gbó
144
J.GBENGA FAGBỌRUN
users. Gbó is not a meaningful unit when separated from gbóná. Gbóná is a member of the ‘fixed verb(1967:33), as against the ‘separable verb-nominal collocation’. nominal combination’ of (10)
dú ‘darken’>dúdú=dídú ‘darkness’; dín ‘fry’>díndín= dídín ‘the frying’; mu ‘drink’>múmu=mímu ‘the drinking’
The use in (10) of the fully reduplicated forms dúdú ‘darkness’, díndín ‘the frying’, and múmu ‘the drinking’ (instead of dídú, dídín, and mímu respectively), may be seen as relics of dialectal variation which are not easily discarded from the language. Abraham did not realise that this variation should be attributed to dialectal influence. Nevertheless, he knew that wíwín and wínwín ‘the borrowing’, both derived from ó wín ‘he borrowed’ (p. xxii), are variants. 3.2. The use of aláìIt could be an oversight for Abraham (p. xiv) and some writers before him to have claimed that aláì- was a single prefix, as in (11), and it is possible that he was confused about the phonological change of alveolar nasal to alveolar lateral before an oral vowel, as explained in Rowlands (1965:105): (11)
a. b.
‘to have offspring’ ‘lack of offspring/childlessness’
c.
‘an individual who is childless’
There are two nominalising prefixes in aláì-, as illustrated in (12). The first is oní- ‘one identified with X’ and the second is àì-, a negator which denotes ‘lack of X’ (X is a variable which is [+N] after oní- but [+V] in (11) is made up of two prefixes as shown in (12): after àì-). It means that (12)
àì- [+NEG]+ oní+ >
[+V]>
[+N]; àì-+V>N ; oní-+N>N
Abraham seems to have confused the feature of àì-, regarding it as ‘a double prefix, consisting of negative à-+the ì-’ (p. xxi). We now know that àì- itself is a negator, as in (12), and that à- and ì- are two different prefixes which have nothing to do with àì. 3.3. Deletion, assimilation and Yoruba writing What Abraham discussed under ‘diminished breathing’ could be described in terms of consonant deletion which allows for vowel lengthening, a common feature in Yoruba phonology and often reflected in writing. He was not certain about the origin of the multiple vowel in Yoruba, as in examples (4), (13), and (14). One of the reviewers of Abraham’s dictionary (Olmsted 1959: 341) described this as a ‘geminated vowel’, but I am not convinced that this term could be used for consonant deletion that results in vowel lengthening. It may probably be regarded as an instance of ‘superficial vowel gemination’, as pointed out to me by John
R.C.ABRAHAM: A LINK BETWEEN TWO PERIODS IN THE YORUBA WRITTEN TRADITION
145
Kelly (p.c.). We have noticed that Abraham did not differentiate between the linguistic processes operating in (13a-b): (13)
a. b.
aago ‘bell’
The item in (13b) is derived by nominal reduplication, while aago in (13a) is a result of consonant deletion. It could be emphasised here that the co-occurrence of two or more vowels in Yoruba is not based on one and the same process, likewise the deletion of sounds. This issue is exemplified by Ward (p. 15ff.) but not fully analysed. Abraham does not go beyond the description in Ward. The structure displayed in (14a) is principled, although Abraham did not specify this: if two identical consonants are present in a monomorphemic noun, the first one may be deleted. On the other hand, as illustrated in (14b), if either y, r, or w occurs in the middle of a word (Ward, p. 25), it can be deleted, and vowel assimilation may take place to conform with the standard realisation, as against a dialectal form such . as (14)
a. b.
akika>aaka ‘type of tree’; egungun>eegun ‘bone’ òyìnbó/èyìnbó>òìbó/èèbó ‘European’ ‘deities’; ‘morning’
3.4. The sound ‘u’ Another phonological feature that stemmed from deletion and assimilation processes, later interpreted as coalescence by some Yoruba scholars (e.g. Awobuluyi 1987), was discussed by Abraham under ‘other tone changes’ (p. xxi) and ‘the sound u’ (p. xxv), although he did not go beyond what was already in the literature on this. (15)
a. b.
‘thank’
As he put it, ‘in frequent cases, i+i results in the sound ‘u’, as in ibi+ ìgbé>ibùgbé “place of living”’ (see (16)). ‘This often happens when ibi ‘place’ is followed by prefix i-’ (p. xxv), as in (15b) above. (16)
a. b.
aláyé+ní+ìwà>aláyélúwà ‘his/her majesty’ ìlòkíìlò>ìlòkulò ‘misuse’
The explanation of the sound u offered by scholars, including Abraham, is not consistent. For example, if > ‘to lie’, what prevents from being derived from pa+ ‘to a+i>u in all cases, as in pa+ ‘to prepare ground’? The presence of u might be attributed to some historical accident— probably and sá+uré respectively. That is, at an earlier stage, some Yoruba thank’ and súré ‘to run’ are from dá+ nouns would have started with the sound u as we still hear in some eastern dialects. If this is the case, it and suré (and others like them) are derived by deletion and could reasonably be supposed that contraction, as illustrated in (17). Abraham would not like to use an elision approach to account for the
146
J.GBENGA FAGBỌRUN
presence of u, since he declared that ‘there is no underlying principle. Whether the first vowel is elided or the second, when two vowels stand in contact, depends on each particular case’ (p. xxxi). There is no gainsaying this assertion. It is interesting to notice that this problem remains unresolved until now. As shown in (17), Abraham noted that when two words come into contact either of them may lose its vowel, without any semantic difference, only the use of u is unpredictable: (17)
dín ‘fry’+ẹrọn ‘meat’>dínrọn or
If we represent the last vowel of the first word with V1 and the first vowel of the second word with V2, the so-called coalescent vowel will be V3. This V3 is the vowel u which does not occur in either of the is from pa+ . words that come into contact. V3 would be V2 in its prototypical form, assuming that This is a controversial issue which is still unresolved. 3.5. Desuetude of morphosyntactic vowel harmony Abraham (p. xxv) noted that ‘sometimes, o=ọ’, as in (18): (18)
kò lọ=
‘he did not go’; yíóò lọ=
‘he will go’
This alternation between ọ and o in (18) is no longer in the Yoruba koiné. What remains is kò lọ or yíóò lọ. However, in most dialects, and at times in some unplanned discourse texts, yóó or kò are used when the or go with vowels word that follows begins with a vowel of high tongue height, i.e. i, e, o, u, while of low tongue height, i.e. ẹ, a, ọ. Abraham was the first scholar to call our attention to this alternation. This morphosyntactic vowel harmony has been accidentally removed from the Yoruba writing system through levelling of sounds. As we see in (19) from Crowther (1852: 18–19), it used to affect mo/mọ (1sg), o/ọ (3sg), in addition to the V[+AUX] in (18): (2sg), and (19)
a. b. c. d.
on gbọddọ wá kí ma wá mọ wá iwò nyin o lè isure
‘he dare not come’ ‘he may come’ ‘I come to see you’ ‘he can/may run’
It is unfortunate that grammarians have never considered the variation illustrated in (19) as an instance of vowel harmony. That Abraham pointed to this is striking and incredible, and we should not overlook it, since the form continues to feature in the spoken language at least.
R.C.ABRAHAM: A LINK BETWEEN TWO PERIODS IN THE YORUBA WRITTEN TRADITION
147
4. Syntax 4.1. Pronouns One major syntactic issue which Abraham approached quite differently is the nature of Yoruba pronouns, and this is the only syntactic point I will discuss. Abraham (p. xxix) seems to have been the first to divide Yoruba pronouns into two categories, an approach which was later explicitly discussed by (1966, 1967:18–19). Ward (1952:78–79) assumed, like most earlier grammarians, that the items in (20b) were abbreviated from those in (20a). Both are, according to her, pronouns, one ‘disyllabic’ (p. 91), the other monosyllabic. Abraham, however, gets the credit for seeing that the items in (20a) are independent of those in (20b): (20)
a. emí lọ òún rí mi àwá wí wí wí
b. mo lọ o wí ó rí mi a wí ẹ wí wí
‘I went’ ‘you said’ ‘he saw me’ ‘we said’ ‘you said’ ‘they said’
Scholars have not agreed on what name to give to both categories. Abraham called those in (20a) ‘independent-pronouns which are nouns’, while those in (20b) were regarded as ‘subject-pronouns’. The . (1967:19), since they possess some characteristics items in (20a) are labelled ‘pronominals’ by of both pronouns and nouns. As he puts it, they have ‘a system of number and of person…they behave exactly like nouns (and not like pronouns)’. Pulleyblank (1986) calls them ‘strong pronouns’. On the other hand, those in (20b), which Abraham called subject-pronouns, are regarded as ‘pronouns’ by and clitics by Pulleyblank. There are some scholars who dogmatically perceive ‘strong pronouns’ as nouns, despite all the pronominal features these possess. My aim here, however, is not to open up further the controversy over the notion of ‘pronoun’ in Yoruba, but to highlight how observant Abraham was, and how he used his knowledge to absorb and catalogue linguistic information, most of which still attracts the attention of scholars. 5. Conclusion It is my view that if a linguistic history of Yoruba does not demarcate between Crowther (1843, 1852) and some of the earliest grammarians, and the influence of Abraham, who was very close to modern and Awobuluyi), many salient points will be confused. On the other hand, grammarians (such as it is also necessary to look closely at the difference between Ward and Abraham who had many things in common, since the latter did not refer to previous grammatical analyses in/on Yoruba. We may not know who did what. What has long existed will be doggedly regarded as a new discovery by modern linguists.
148
J.GBENGA FAGBỌRUN
The title of Abraham’s dictionary suggests that it contains ‘modern’ Yoruba, indicating that there should be ‘old’ Yoruba. Which Yoruba is old? Who spoke (or speaks) it? To scholars of generative historical linguistics and literary tradition, this term ‘modern’ shows that as far back as 1958, scholars had perceived some changes in the structure of Yoruba. Unfortunately they have not come up with reasonable points in support of the term ‘modern or refined’ Yoruba. We have noted that most structures so regarded have been in use since Crowther. Probably what is ‘modern’ is the analysis and not the corpus of data. If we were allowed, Abraham’s dictionary would be renamed ‘A Dictionary of the Yoruba Koiné’, since it contains most of what we know of so-called standard Yoruba from the onset. Moreover, most of what Abraham proposed or alluded to in Yoruba grammar is yet to be fully investigated by scholars. REFERENCES Abraham, R.C. 1958. Dictionary of Modern Yoruba. London: University of London Press. Armstrong, Robert G. 1959. Review of Dictionary of Modern Yoruba by R.C. Abraham. Africa 29(1):90–92. Awobuluyi, A.Ọladele. 1987. Towards a typology of coalescence. The Journal of West African Languages XVII(2): 5–22. , Ayọ. 1966. A Grammar of Yoruba. (West African Language Monographs 5.) Cambridge: Cambridge University Press. ——. 1967. A Short Yoruba Grammar. Ibadan: Heinemann Education Books (Nigeria) Ltd. Church Missionary Society (CMS) Archives, CA2/096/113. Minutes of a Conference on the Yoruba Language, December 1875. Crowther, S.A. 1843. A Dictionary of Yoruba. London: CMS. ——. 1852. A Grammar and Vocabulary of the Yoruba Language. London: Seeleys. Fagbọrun, J.Gbenga. 1989. Disparities in tonal and vowel representation: some practical problems in Yoruba orthography. The Journal of West African Languages XIX(2):74–92. ——. 1991. The analysis of linguistic innovation: historical and grammatical perspectives on English loan-translations and the Yoruba koiné. DPhil thesis, University of York. Joint Consultative Committee on Education (JCCE). 1974. Yoruba Orthography Recommendation (YOR). Ms. Olmsted, David. 1959. Review of Dictionary of Modern Yoruba, by R.C.Abraham. American Anthropologist 61: 341–342. Pulleyblank, Douglas. 1986. Clitics in Yoruba. Syntax and Semantics 19:43–64. Report of the Yoruba Orthography Committee. 1969. Ibadan: Ministry of Education. Rowlands, E.C. 1965. Yoruba dialects in the Polyglotta Africana. Sierra Leone Language Review 4:103–108. Ward, Ida C. 1952. An Introduction to the Yoruba Language. Cambridge: Heffer and Sons Ltd.
ABRAHAM’S ‘MAGNETISED TONE’ IN YORÙBÁ* Benjamin Akíntúndé Oyètádé
1. Introduction Abraham’s (1958: x–xxi) description of the tonal system of Yorùbá is one of the detailed explanations available on tonology. It has proved very useful because it has served as a solid foundation upon which other (1966a), Courtenay (1969), Oyèláràn (1971), tonal analyses of Yorùbá such as those of Pulleyblank (1986), Akinlabí (1985) and Oyètádé (1988), are built. A few of the errors in Abraham’s treatment have been addressed in some of these studies. Some of the errors are quite understandable in view of the fact that Abraham was not a native speaker, and so had to rely on written material and informants. Some of the mistakes are thus not his but those of his sources, and some turned out to be purely typographical. In this study, I examine one of the types of tones that Abraham identified. Of particular interest to me is what he called the ‘magnetised tone’, and I subject his description to my own native speaker judgement and provide an analysis. In §2 I present Abraham’s description of the data, and assess his description in §3. In §4 I provide a dialectal clue to Abraham’s view, and in §5 a general rule is proposed to account for the relevant data. In §6 I offer a conclusion based on the facts presented. 2. Abraham’s description of the ‘magnetised tone’ Abraham’s description of the magnetised tone (hereafter MT) is as follows: There are certain classes of words which are temperamentally inclined and whose tones fluctuate under the influence of their environment, some succumbing to the influence of words preceding them, while others are affected by what follows. (1958:xvii) It is in these circumstances that the MT, which Abraham recognised as the ‘fourth static tone’, comes into existence. The other three static tones are high, mid and low tones represented with the following diacritics: (΄) for high, (¯) for mid and (`) for low. The mid tone is usually unmarked except on syllabic nasals or when it becomes necessary to do so to clarify an illustration.1
African Languages and Cultures, Supplement 1 (1992):187–195
150
BENJAMIN AKÍNTÚNDÉ OYÈTÁDÉ
Abraham claimed further that the MT has no independent existence but that it only arises positionally. He perceived its position in pitch as being between mid and low. He went further to give the name ‘midlow’ to the tone and marked it with the symbol (˘). Abraham (pp. xvii–xviii) noted that the most important case of the realisation of the MT ‘is where two words stand next to one another and such combinations consist solely of mid tones’, e.g.: (1) (2)
ìwọ’ a. b.
abẹ kọn2 kọn kejì kẹta fŭnfŭn funfũň kọn
(3) (4) (5) (6) (7)
He described the tone rule reducing (8)
‘you are kind indeed’ ‘razor’ ‘one razor’ ‘one husband’ ‘the 2nd razor’ ‘the 3rd razor’ ‘a white cloth’ ‘a white cloth’
to
as follows:3
‘when the noun before kọn has two or more mid tones, then the syllable immediately before kọn falls from mid to midlow.’ (p. 513)
The examples cited in (2–4) and (6) are again used to support this rule. However, Abraham noted that certain mid-toned words such as ohun ‘thing’, ẹni ‘person’ and ibi ‘place’ are exceptions to this rule. Instead of their final mid tones becoming mid-low before kọn, he observed that the mid tone actually becomes low, e.g.: (9)
a. b. c.
ohun ẹni ibi
kọn kọn kọn
→ → →
ōhùn ībì
kọn kọn kọn
‘one thing’ ‘one person’ ‘one place’
The present writer does not see the examples in (9) as exceptions at all. In the next section, we shall see how the same analysis that handles the data in (2–5), and (8) handles the examples in (9). Another word in which Abraham observed the MT is èwo ‘which one/what kind/what’ (pp. 168–169). He rightly noted that the item is often shortened to wo, but claimed further that (just as it was described in the case of kọn above): ‘when the noun preceeding wo ends in mid tone, wo becomes wŏ.’ The following
*
I am grateful to Philip Jaggar for inviting me to present an earlier version of this paper at the R.C.Abraham Symposium, and also to the participants at the Symposium, especially Jack Carnochan and Kay Williamson, for their comments. 1 These clarifications are necessary in dictionaries such as Abraham’s, in phonetic transcriptions of literary genre and in manuals for teaching foreigners. Other diacritics that are in common usage in Yorùbá orthography include the sub-dot for [š] or [∫], sub-dot ẹ for [ε], and sub-dot ọ for [ ]. Nasalisation is indicated in the =ọn, [ĩ]=in, = ẹn, [ã]=an and [ũ]=un. orthography by an ‘n’ following a nasalised vowel, e.g. The nasalised vowel ọn and an are dialectal variants. I have used on in most of the examples in this study because that is the variant that Abraham consistently used. The other—an—is preferred in S(tandard) Y(orùbá) for ìkan/ ‘one’. 2
ABRAHAM’S ‘MAGNETISED TONE’ IN YORÙBÁ
151
examples in (10a) are Abraham’s examples of when èwo becomes wŏ if preceded by a mid tone, and (10b) when èwo is simply wo if not preceded by a mid tone: (10)
a. b.
ẹyẹwŏ ‘which bird’; òkunwŏ ‘which sea’; irú arúùfinwŏ ‘what type of criminal’ ìgbàwo ‘what time’; oyèwo ‘what kind of Official Title’; ‘what day of the month’
Abraham (pp. 421, 446–447, 513, 562) referred to other magnetised words apart from the use of midlow tone, using the same symbol for them all, but his claims about these words are not of crucial significance to the main thrust of this paper, and so I will not pay particular attention to them. Let us now turn to an assessment of Abraham’s description based on my perception of the data presented above. 3. An assessment of Abraham’s description One general comment about Abraham’s description of the data is that he rightly perceived a different pitch, distinct as it were, from the three well known static tones of Yorùbá. When one examines the description critically, one discovers that a number of claims regarding the MT are inadequate, at least from my own point of view. Let us now look at these inadequacies one by one. 3.1. Fourth static tone From Abraham’s description, it is clear that the midlow MT, which he rightly identified, could not be a static tone of a similar status as the high, mid and low tones of the language. Abraham himself said that ‘its position in pitch is between mid and low…it has no independent existence, but only arises positionally’ (p. xvii). The MT is thus significantly different from the other three static tones which have an independent existence and are not positionally conditioned. As a matter of fact, positing a fourth static tone cannot be substantiated. 3.2. Environment for the ‘magnetised tone’ As noted earlier, Abraham claimed that the environment that triggers the realisation of the MT could be the influence of words either to the left (i.e. preceding) or to the right (i.e. following) of the tone that is magnetised (p. xvii), and that the trigger is unidirectional. As we shall see in §5, I make the following claims: (11)
3
a.
a unified account takes care of all the relevant data presented by Abraham. The non-relevant data are handled separately.
This item is rendered as ìkan or in SY (cf. fn. 2 above). I use the symbol ( ) as in downstep of the mid instead of Abraham’s (˘) for the midlow tone.
and
to indicate
152
BENJAMIN AKÍNTÚNDÉ OYÈTÁDÉ
b. an analysis of the phonological/tonological processes involving and èwo before they are reduced to kọn (my ) and wo (my ) provides a solution to the problem raised by Abraham’s description. c. the ‘two or more mid tones’ as claimed in (8) has nothing to do with the tone change from mid to midlow as Abraham perceived it. d. the examples given in (10) need a more rigorous explanation. e. the difference shown between the examples in (10a) and (10b) based on the tone of the word preceding wo is unnecessary.
3.3. Non-relevant data Some of the examples cited by Abraham do not support his claims about the MT at all. Such examples are the ones in (1) and (6), and I will attempt to place them where they belong in §5.4. 4. Dialectal clues There are at least two ways of viewing Abraham’s data. One can assess it from the Standard Yorùbá (SY) point of view,4 or from the point of view of influences from other dialects. This gives us a clue about Abraham’s perception and also enriches the analysis provided in §5. 4.1. Standard Yorùbá forms The items presented in (1–7) are pronounced in isolation as follows: (12)
a. b. c. d. e.
‘you sg.’ ‘did something’ ‘indeed’ ‘razor’ ‘one’
f. g. h. i. j.
‘husband’ [ìkējì]/[èkējì] ‘second’ ‘third’ ‘cloth’ ‘white’
All these items are retained in the forms shown above when rendered in isolation. This does not rule out minimal idiolectal phonetic variations. There is, however, a significant difference when these forms are or with a rendered in a sentence or in a genitival clause. In SY, (2b) will be rendered as or which is a case downstepped mid tone on the last vowel of the qualifier, and not as of vowel assimilation after the low tone of the qualifier has downstepped the following mid tone. The same applies to the items in (3), (4), (5) and (7) which are rendered as shown in (13): (13)
b.
a.
abẹ kėjì
ọkọ
not
not
*
kėjì
ABRAHAM’S ‘MAGNETISED TONE’ IN YORÙBÁ
c. d.
abẹ
not not
funfun
* *
153
funfunùn
Notice that what is happening here does not apply to the items in (1) and (6)—the non-relevant data (cf. §§3.3 and 5.4). 4.2. Dialectal influence dialect spoken in township, In certain dialects of Yorùbá, especially in the version of and environs, in areas of dialect, and in parts of dialect-speaking areas, the data presented by Abraham in (2b-6) and (8) are acceptable. What one can deduce therefore is that Abraham’s sources for these data must have been strongly influenced by these dialects.5 Let us now turn to an analysis of these facts. 5. Analysis In Oyètádé (1988:213ff.), I claimed that mid tone downstep in Yorùbá is sensitive to both the underlying (linked) and floating (unlinked) low tones which other scholars, e.g. Stahlke (1974:141), have called ‘superficial low tones’. By this I mean that a low tone—either linked/associated or unlinked/delinked/ floating—conditions downstep on the following mid tone or sequence of mid tones. See also (1966b:1–13) earlier view on the effect of a preceding low tone on the tone of a following syllable—my analysis of Abraham’s data is based on this claim. 5.1. The data The data presented in (2b–6), (7), (9) and (10) have one thing in common— they all have one of the following items used as qualifiers in the concatenations: (14)
a. b.
kọn kejì
< <
ìkọn/ ìkejì/èkejì
‘one’ ‘second’
4 The so-called SY (Standard Yorùbá) or ‘Common Yorùbá’, as some scholars prefer to call it, is a controversial issue in certain circles. However, in this study, it is used to refer to the kind of Yorùbá used in written texts, i.e. books and newspapers, for teaching in schools and for radio/television broadcasts. 5
Peter Morton-Williams, a colleague of Abraham’s, working in Ìbàdàn when the data for the dictionary was being collected, confirmed at the Symposium that one of Abraham’s major informants was a native of . This goes a long way to confirm my speculation. But notice that this is contrary to the view expressed by Abraham in his ‘Preface’ (p. iii), where he asserts that Daniel Fágúnwà is from Ilé- , and that the three other informants—Samuel Wínjọbí, Dúró Ògúndìran and Lápàdé —are from Ìbàdàn. We know that the late D.O.Fágúnwà is a native of ÒkèIgbó. There are records to the effect that he worked in , which used to be classified as a quarter in Ilé- . The fact that the three other informants were studying or working in Ìbàdàn does not mean that they were natives of Ìbàdàn. They may have come from any of the towns in area or from the north of .
154
BENJAMIN AKÍNTÚNDÉ OYÈTÁDÉ
c. d.
kẹta wo
< <
ìkẹta/ èwo
‘third’ ‘which/what’
The full forms of the items before they become contracted are given in stages (i) of examples (15a-e). The phonological processes, specifically vowel assimilation and vowel deletion, which produced the output apply at stages (iii) and (iv) respectively. (15)
a. → b. → c. → d. → e. →
i. iii. i. iii. i. iii. i. iii. i. iii.
abẹ ìkọn abẹ ọkọ ìkọn ọkọ abẹ ìkejì abẹ abẹ ìkẹta abẹ funfun ìkọn funfun
→ → → → → → → → → →
ii. iv. ii. iv. ii. iv. ii. iv. ii. iv.
abẹ abẹ ` =abẹ ọkọ ọkọ ` =okọ abẹ ìkėjì abẹ ` kėjì=abẹ kėjì abẹ abẹ ` =abẹ funfun funfun ` =
funfun
In (15a-e), stage (i) is the underlying form. Stage (ii) shows the downstepping effect of a low tone on the following mid. Stage (iii) is the dialectal version mentioned in §4.2. What takes place here is a vowel assimilation process in which the initial vowel of the qualifier is assimilated to the last vowel of the preceding noun. As a general rule, it is the vowel of the qualifier that assimilates to those of the preceding nouns. Stage (iv) is the acceptable form in SY, arrived at as a result of vowel deletion which deletes the assimilated vowel of the qualifier, leaving the low tone floating. The final output of stage (iv) appears after the=sign, where the floating low tone is no more visible but the downstep that the low tone conditions in stage (ii) is still evident. It is a combination of vowel assimilation and the downstepping effect of a low tone on the following mid that I think Abraham perceived and tried to present as MT in Yorùbá. In actual fact, no 1966b:1–13). tone is magnetised (see I have shown in (11a-c) that a closer look at the qualifiers in relation to the nouns they qualify offers us a unified account of the facts. The issue of directionality is thus solved once and for all. Also, notice that the mid tone of the nouns has no input whatsoever to either the downstepping, the vowel deletion or the vowel assimilation that take place. Hence, Abraham’s claim that the most important environment for the realisation of the MT ‘is where two words stand next to one another and such combinations consist solely of mid tones’ is not valid. 5.2. An alternative explanation For the examples given in (9), the forms on the left hand side of the arrow are acceptable in this context in . The low tone on the last syllable/vowel of the nouns SY with a perceptible downstep on the qualifier (as indicated on the right hand side of the arrow) is reminiscent of the initial low toned vowel of the qualifiers which had been deleted, leaving the floating low tone to dock on the mid toned final vowel of the nouns. But notice that the forms could have other meanings in SY. The example in (9a) which Abraham presented as ohùn kọn ‘one thing’ could also mean ‘one voice/tone’. To make a clear meaning distinction, ohun ùn kọn
ABRAHAM’S ‘MAGNETISED TONE’ IN YORÙBÁ
155
in deliberately slow speech is preferred for the meaning ‘one thing’, while ohùn kọn for ‘one voice/tone’ is unambiguous. In (9c), ibì kọn has no other interpretation. As a matter of fact, it is the only acceptable form for ‘one place’. The form on the left hand side of the arrow, ibi kọn, could mean ‘one/any evil’.6 In the case of (9b), there is no conflict of meaning. However, ẹnìkọn or ẹnìkan is preferred, even in SY, with the meaning ‘one person, somebody’. To summarise, ibi kọn and ẹni kọn sound somewhat artificial (apart from for ‘one/any place’ and ẹnì kan for ‘one person, the context, i.e. you will commonly hear ibì somebody)’, but ohun kọn does not. The issue raised in (11d), i.e. that of a more rigorous explanation of the examples in (10), is thus also dealt with. Now let us turn to the forms in (11e). 5.3. Preceding mid tone(s) and ‘magnetisation’ The reason for the distinction made between the data in (10a-b) in Abraham’s view is that the forms in (10a) are subject to magnetisation because the tones of the last syllable of the preceding noun are mid, and those in (10b) are not magnetised because the tones of the last syllable of the preceding noun are not mid. I have shown in §5.1 that the mid tone(s) of the preceding noun contributes nothing to the vowel deletion, downstepping of the following mid or vowel assimilation that characterised what Abraham analysed as MT. Based on the analysis presented in §5.1, Abraham’s claim in (8) and the proposal that examples (10a-b) were meant to support are not valid. If the items in (10a-b) are taken through the same stages as the ones in (15), the same procedure will give us a unified account of the data. 5.4. Non-relevant data The examples in (1) and (6) are the non-relevant data as mentioned in §3.3. They are non-relevant because jọjọ the tonal patterns that distinguish the qualifiers in the rest of the examples are not present here. In funfun we find a pattern that is not modified by the type of qualifiers used in the other examples. The and data are thus not subject to the processes that apply to the rest of the data at their different stages in (15). They are therefore unanalysable within the analysis that I am proposing here, and so I ignore them. and funfunùn, a clue about where they belong If what Abraham perceived was may be found in the fact that there is an emphatic use of tone similar to this in Yorùbá. In this case, (1) and (6) will still mean ‘you are kind indeed’ and ‘a white cloth’ respectively, but with a sort of sarcasm. Notice , and (6) as however, that with Abraham’s tone markings, (1) will be rendered as funùnfunùn. These pronunciations are very strange both in SY and in many dialects of Yorùbá known to me. 6. Conclusion What I have done in this study is examine Abraham’s description of what he termed the ‘magnetised tone’ in Yorùbá. I found that Abraham truly suspected a type of tone change, though there are certain problems in his description and explanation of the process. I was able to show that a couple of phonological/tonological processes are involved in what Abraham rightly perceived. My analysis, however, does not support 6
See for example Psalm 23, verse 4 in Scofield’s (1967:455) edition of The Holy Bible, where ‘I will fear no evil’ is rendered in the Yorùbá version as èmi kì yíó ibi kan.
156
BENJAMIN AKÍNTÚNDÉ OYÈTÁDÉ
Abraham’s claim that any tone is ‘magnetised’. The processes I identified include low tone conditioning downstep on the following mid, vowel assimilation in certain dialects of Yorùbá, and vowel deletion in SY. I also showed that two of the illustrative examples used by Abraham are in fact not relevant to his claims about the so-called ‘magnetised tone’. REFERENCES Abraham, R.C. 1958. Dictionary of Modern Yoruba. London: University of London Press. Akinlabí, A.M. 1985. Tonal underspecification and Yoruba tone. PhD thesis, University of Ibadan. , A. 1966a. A Grammar of Yoruba. (West African Language Monographs, 5.) Cambridge: Cambridge University Press. ——. 1966b. The assimilated low tone in Yoruba. Lingua 16:1–13. Courtenay, Karen. 1969. A generative phonology of Yoruba. PhD dissertation, UCLA. Oyèláràn, O.O. 1971. Yoruba phonology. PhD dissertation, Stanford University. Oyètádé, B.A. 1988. Issues in the analysis of Yorùbá tone. PhD thesis, SOAS, University of London. Pulleyblank, D. 1986. Tone in Lexical Phonology. Dordrecht: Reidel. Scofield, C.I. (ed.). 1967. Holy Bible (The New Scofield Reference Bible). New York: Oxford University Press. Stahlke, H. 1974. The development of the three-way tonal contrast in Yoruba. In Third Annual Conference on African Linguistics, ed. by Erhard Voeltz, pp. 139– 145. Bloomington, Indiana: Indiana University Press.
FOND RECOLLECTIONS OF ROY P.E.H.Hair
I only knew Roy Abraham in his later years, and therefore what I recollect about him may caricature his earlier life. But I doubt it, since I suspect that he was a born eccentric. He was also a scholar of very great distinction, and so impressed me on a fairly brief acquaintance that, although I was not informed in his speciality, the field of African linguistics, I began to compile a bibliography of his writings. I have found among my papers an early version which I showed him and which bears his corrections, in red ink, and his characteristic writing (he wrote an ‘a’ more like a ‘z’). But he died in 1963 before I could complete it, and when I published the bibliography it had to be in his memory (The Journal of West African Languages 2(1), 1965). During the next few years up to her death, I kept in touch with Sadie, his elderly widow, and on only annual visits a degree of affectionate regard grew up between us, so that I was privileged to spend a couple of hours with her and be of help to her during her final illness. I met Roy in Ibadan in 1952. I was in criminal exile. Having gone out to Nigeria to undertake field research I had had a row with my elderly boss—he possessed the senescent narrow-mindedness of the age which I have now attained—so I was ordered to stay away from my field lest I overturn the British Empire. Instead, I was to occupy myself in less subversive activities at the institutional headquarters at Ibadan University. On a trip to Lagos I had met Bob Armstrong, then researching in Idoma country, and Bob told me, perhaps warned me, about the forthcoming arrival of Roy. (Bob was in closer touch with both Abrahams than I ever was, and right up to her death warmly and thoughtfully supported Sadie. In return, she was very fond of Bob, despite some half-hearted interrogation of me over his life-style. But Bob is now dead too, so cannot reminisce.) It was Bob who had arranged for Roy’s visit to Nigeria to be conducted under the wing of the newly-founded West African Institute of Social and Economic Research, a grand-sounding entity which was mostly hunting around for things to do, at least things that the aged Director thought proper and nonirritating to his peers in the Lagos Administration. However, Roy chose to arrive in Nigeria and at Ibadan some time after my moment of disgrace. The Director and Secretary of the Institute could not forbear taking their vacations in Britain and therefore could do nothing else but leave me, with very strict instructions, to hold the fort in Ibadan, to open letters, to keep the clerks in order and things generally ticking over till they got back—and to behave myself. This is why it was me who looked after Roy. A dozen or so unmarried academics and technicians not on vacation ate in a mess (on the ‘New Site’, the present site of the university, but then only minimally occupied); and one night Roy arrived. We happened to have another visitor that night, a talkative man fond of describing his global exploits. He was chirruping away merrily when he chanced to begin an anecdote with: ‘When I was in Kenya…’. The other stranger
African Languages and Cultures, Supplement 1 (1992):197–202
158
P.E.H.HAIR
(Roy) present instantly leant forward and intervened enthusiastically: ‘You’ve been in Kenya?’ Visitor: ‘Oh yes.’ Roy: ‘Was that in the western province?’ Visitor: ‘No.’ Roy: ‘In the northern province?’ Visitor: ‘No.’ Roy: ‘In the southern province?’ Visitor: ‘Not quite.’ Roy: ‘Near Nairobi? Near Thika? Up in the hills, beyond the river, across the plains?’ Around the table embarrassment grew as the luckless man was pursued relentlessly but in a tone of mounting angelic curiosity—Roy just wanted to know, he having been there— all over Kenya. (Roy was there during the Ethiopian campaign, which generated his work on Amharic and Somali.) Eventually it transpired that the man had spent one day in Mombasa. This was my first acquaintance with Roy. It was transparent that at one level he operated with no sense of malice, that his curiosity was childlike, and that he simply lacked the social instinct to hold back in a situation when carrying on may rouse the ire of a heavily-muscled victim. My very last sight of Roy was some years later. He had pushed his way through a crowd assembled around a wall timetable at Cambridge Station and was blithely addressing me on the train time through the bodies of the bystanders whom he ignored. He had more than the normal provision of a one-track mind; and this, after all, was what made his scholarship possible. At Ibadan, Bob and I organised Roy. Bob found him Yoruba informants, I laid on a bungalow, with all services, a servant, mess-arrangements, and transport. I heard Roy interrogating his first informant, a girl called Victoria, whom he called ‘Fikkki’—Bob opined that with his knowledge of German Roy was being naughty. Maybe, for Roy was given to being naughty, or at least thinking himself to be naughty, and his laughter was a bit like a giggle. It was difficult at times to remember that this elderly brilliant adolescent had begun as an officer in the Indian Army, let alone continued as a Colonial Administrator. But Roy in the colonial service had had his difficulties—indeed I suspect that he was often sent to far-flung places as linguist and anthropologist in order to get him out from under the feet of the routine administration. One governor of Northern Nigeria was obsessed with the notion that traditions and languages in his region had developed from Arab origins, and he published on these lines. Knowing that Major Abraham had Arabic, he sent him some of his pet Arabic-derived etymologies; Roy despatched them to his professor of Arabic in Oxford who wrote below ‘Rubbish’, and Roy passed the material back to the governor without further comment. Roy told me this story, gleefully. At Ibadan Roy got through a lot of Yoruba in a very short time. In the mess they put his eccentricities down to overwork, with some justice. In particular, he recruited little Mr Sutton, a dedicated, misogynist, botany technician whose trousers came up to his armpits and who spent all his spare time wandering through the countryside collecting and then drawing flora and fauna specimens—he did the illustrations for Roy’s Dictionary of Modern Yoruba (1958). We took Roy about a bit, to keep him happy. One week we were visited by a senior academic from Britain, Lucy Mair. Lucy, not so young, did not help herself by arriving in clothes that appeared to have been picked up at random at a church jumble sale, and she was very stiff. (Sally Chilver, our contact woman in London, told us however that Lucy thawed out after a dram or two.) One night I took Roy and a sober Lucy to the open-air cinema in Ibadan. On arrival Roy stopped back to gossip in Arabic with the Syrian proprietor; but when he joined our seats he sat next to Lucy and tried to vamp her. Lucy grew starchier and starchier, until Roy turned to me and said in a very loud voice: ‘Paul, that woman is an ABORTION!’ Being young and unconfident I worried that I was not doing too well in looking after Roy. And I was right, Roy was not happy. He didn’t like his nice bungalow, so I moved him to the Old Site where there were more people to talk to. Bob had now gone off to his beloved Idoma and I was left coping with Roy. Since Roy arrived we had heard much about Sadie, his wife, who had seized the chance to visit South Africa. Letters flowed from Roy to Sadie, and soon almost daily cables (I had to send a driver with each). Clearly Roy was finding it difficult to exist without Sadie. The mess was inclined to think that at home
FOND RECOLLECTIONS OF ROY
159
she did up his buttons and shoe-laces. Suddenly one morning I had a message. Roy had run amuck and was roaming around the Old Site telling passers-by that I was persecuting him and making his life a misery, and he must go home immediately. In fairness, I suppose I had rather discouraged his retreat when he hinted at it, on the grounds that he had an obligation to the Institute. Roy did depart, after more urgent cables to Sadie, leaving me more than a little upset at losing what had been entrusted to me; but when I consulted the senior academic in situ, he just laughed and said it was probably all for the best. I got no sympathy either when Bob re-appeared, since he reminded me that he had always said that Roy was a prima donna (Bob had once been an opera-singer so didn’t use the term lightly). To my astonishment, once Roy had collected Sadie and was back in Hendon he sent me warm messages— not a word ever again about the contretemps—and looking through his books I see that in the ‘Introduction’ to his Yoruba Dictionary (1958) I get flowery thanks for this and that. Actually Roy overdid it, for it was Bob who was the instigator of the Yoruba activity. However, the thanks to me in the posthumous The Principles of Ibo (1967) were more deserved: I forgave him for the scare, and while still at Ibadan set things in motion to get him to work on Igbo, partly because I myself was supposed to be doing fieldwork among Igbo-speakers. But of course he died before he could complete this work. Back in Hendon, Roy went through several Yoruba informants (it is worth noting that he had previously researched languages in their homeland, but he seems to have done Yoruba as well at Hendon as if he had stuck it out in Yorubaland). I kept in touch because the first informant, whom we had recruited for him at Ibadan, had a nervous breakdown and returned to Nigeria, believing that he saw white men pointing their fingers at him (probably not Roy but the examiners of the college course which he had failed). I spoke to Davidson Nicol about the medical aspects and we inclined to the view that Roy’s quickfire interrogation might not have helped. However, all ended happily: the man was sent to his home village where the elders prescribed a wife, and this prescription taken, he recovered. As for Roy, he worked day and night at Yoruba, seven days a week, with new informants; and Sadie must have been now, as ever, uncommonly devoted to provide the comforts and attention he needed, she herself being no push-over but a very sharp person in her own right. They lived at Hendon, in a semi, in a street just behind the synagogue, although I think neither attended it. Perhaps Roy’s extraction had something to do with his marginal position in the administrative service— beneath the public-school manners of the Colonial staff lurked various prejudices— although in any case linguists and anthropologists had their uses but didn’t get promotion. I am amused when jejune NorthAmerican historians parrot the cliché that colonial anthropologists were merely ‘puppets of colonialism’. Still and all, Roy had the mental world of his upbringing and service. Having spent a lifetime unravelling the intricacies of African languages, and occasionally the intricacies of African cultures, he did not patronise their speakers and practitioners, but was instead distinctly dismissive of Black Africans’ lifestyles, being inclined to giggle when he thought of their funny ways. Perhaps he thought the same about the French, Germans, Russians, etc.; be that as it may, he did not share the relativist conceits of the world-view of the 1990s. The Abrahams were friendly with the Meeks, C.K.Meek having been Roy’s contemporary corresponding number in South Nigeria. Presumably they knew each other in Nigeria, although I never learned how much time Sadie spent out there. I didn’t hear either Roy or Sadie speak of other Nigerian ex-colleagues as friends, but this may have been because most of them were dead by now. After the War, Roy had a short spell at SOAS which ended, I believe, with a bang. Roy was difficult. But he remained on good terms with Archie Tucker (Sadie said he was the only SOAS person who backed Roy), and after Roy’s death Sadie kept in touch with the Tuckers. She spent a lot of energy seeing to the re-publication of some of Roy’s books,
160
P.E.H.HAIR
and she was advised in this by Archie, as well as by Bob and myself. Sadie was well educated and intelligent, but I have no idea whether she made any contribution to Roy’s linguistic activity. After Sadie’s death I lost touch with the Abrahams. She had talked regularly and concernedly about her grandson who was in England, her only grandchild, but I never met him and he later died. Roy and Sadie’s only son, Donald, was by then in America (I know him only from correspondence), and she said she spent ‘a fortune’, ringing him up in the days before cheap calls. I had commemorated Roy in print, giving, as far as I could, a scholarly assessment of his achievement and character; his immense industry, his linguistic capacity and perspicacity, and the range of African languages he had worked on. And I hinted at the personality that appeared in the all-too-revealing prefaces to many of his books: tetchy, quarrelsome, dismissive of other scholars, egotistically self-assured. While correctly putting much of this down to the way he had been treated, the lack of appreciation of his abilities and the frustration engendered, I was unable while Sadie lived to say more. I saw something of Roy in action. In the Ibadan mess, when old Professor (not ‘Freddie’!) Parsons reminisced about his time in Russia, Roy chipped in, because in 1945 he had been an army officer escorting Russian ex-POWs and deserters back to their unwelcoming homeland. And I think it was Roy who told me that Russian was an easy language—he, after being instructed to acquire it, having done so in six weeks. I have mentioned Roy chatting in Arabic; he lapsed into Urdu when taken to the Indian shops at Ibadan, and I once took him to the Sabon Gari there to chat to the traders in what seemed to me fluent Hausa. Perhaps best of all, I heard him tick off a Yoruba informant for getting a tone wrong, this after about three weeks study of the language. The other aspect of Roy’s activity I admired was his typewriting. He was quite obsessed by the magic of transforming sounds into print. I sometimes thought that he only did languages in order to produce typescript. He talked incessantly about the mechanical problems involved, and on one visit to Hendon attempted to convey to me his own enthralment with having procured a machine with two alternative platens, enabling him to get two typefaces on the same page. His later books were produced photographically from his own typescript, and he exercised much ingenuity in laying out the pages in complex patterns. His method of working was to go over and over the material, adding to or deleting the first draft, new material being typed on separate scraps of paper and stuck in, until the typescript was a mixture of pages, half-pages and extended pages. Sadie delighted to tell the story of how they once showed a typescript to a printer at Stephen Austen’s: he held the typescript out of an upper floor window and the longest of Roy’s extensions to a page—a sort of super footnote—fell down and reached the ground. I could never understand Roy’s obsession with managing typescript—until recently I acquired a word-processor and it was revealed that half the fun and two-thirds of the labour in desk-top publishing is not the research and writing up but the production of camera-ready copy. What would Roy have done had he lived into the WP age! Thanks partly to my contact with Roy, my research and publications in the 1960s dealt mainly with the history of African languages and African-language linguists. Then I moved off in other directions, and in the last two decades it had escaped my notice that Roy Abraham’s work was not forgotten but was being increasingly respected by younger scholars. Therefore I was surprised when I learned that a symposium in tribute to his memory was to be held, and even more so when I noted the extent of the respectful interest. But a very pleasant surprise. For it is good that a man so devoted to scholarship, so hardworking, so singleminded, should not only be commemorated but should have his work closely examined, nearly 30 years after his death, and have it found worthy of being discussed, criticised, amended and built on. I do not go back on my view that Roy was an eccentric, on the whole rather an appealing eccentric, but to some a very
FOND RECOLLECTIONS OF ROY
161
irritating eccentric; since no doubt it was just this departure from our normal mediocrity that enabled him to do what he did. REFERENCES Abraham, R.C. 1958. Dictionary of Modern Yoruba. London: University of London Press. ——. 1967. The Principles of Ibo (Archival edition of typescript). (Occasional Publication No. 4.) Ibadan: Institute of African Studies, University of Ibadan. Hair, P.E.H. 1965. A bibliography of R.C. Abraham—linguist and lexicographer. The Journal of West African Languages 2(1):63–66.