You are here

D.W. Arnott (1915-2004)
The Nominal and Verbal Systems of Fula
Oxford: Clarendon Press. 1970. xiv + 432 p.

D.W. Arnott. The Nominal and Verbal Systems of Fula.

Notice. I second Pierre Francis' remark in his review below that D.W. Arnott should have adopted a less idiosyncratic transcription method in this important contribution to our knowledge and understanding of the language of the Fulɓe, which binds together a cultural and linguistic domain vast enough to inspire such nicknames as Archipel peul or Planète peule. Especially given that The Nominal and Verbal System of Fula was published four years after the 1966 UNESCO-convened Bamako meeting where experts — including P.F. Lacroix, Amadou Hampâté Bâ, Tierno Chaikou Baldé, Abdoulaye Diallo, etc. — agreed on the Standard Alphabet of Pular/Fulfulde. Had Arnott adopted that system, his book would have preempted Lacroix' complaint. And his work would have gained in legibility. In this web edition I try to compensate for Arnott's oversight by applying systematically the Pular/Fulfulde Standard Alphabet, which UNICODE embeds and strengthens. Also, whenever possible I complement Arnott's Fulfulde (eastern dialects) examples with their equivalents (labelled FJ) from my native Fuuta-Jalon Pular, a western  dialect. The absence of the (FJ) tag indicates that those examples are identical in both western and eastern dialects.  
Tierno S. Bah

Fula - Fulfulde-Pular language continuum
A. Harrison & I. Tucker. Fulfulde Language Family Report. (SIL, 2003) 

List of Tables
List of Appendices
Symbols and Abbreviations

Part I. — General Introduction

Fula is the language of the Fulani, the nomadic cattle-owners of West Africa, whose unknown origins have provided a fruitful field for speculation for those so inclined, and prompted theories of relationship to peoples as diverse as the Ancient Egyptians, the biblical Phut, the Basques, and the Dravidians of India 1, Now, after centuries of gradual movement, mainly in an easterly direction, from an early habitat which seems to have been somewhere in the eastern part of what is now Senegal or the western part of present-day Mali, they are found throughout a wide band of West Africa, roughly between the 10th and 15th parallels and extending from Senegal, Gambia, and Guinea on the Atlantic, through Mali, Upper Volta, Niger, and northern Nigeria to Chad and Cameroon, while the fringes of the dispersion are to be found in southern Mauretania, northern Sierra Leone and Ghana, in Dahomey, and even as far east as the Sudan.
The names for the people and the language present some problems in view of the variety of the terms used in the language itself. The people call themselves Fulɓe (singular Pullo), and refer to their language variously as Pulaar (Senegal), Pular (Guinea), and Fulfulde (Mali and eastwards), the common denominator being the stem Ful-/Pul-. English writers have usually referred to both people and language by the Hausa term Fulani (which properly refers to the people only, a different word being used for the language). German writers use the simple stem Ful for both, the French use Peul (sometimes Peulh), and it seems best, at least for the name of the language, to use some such simple form based on the stem. The Mandinka and Susu. name Fula, used also in the Gambia, seems appropriate as well as more euphonious in English than the plain stem Ful, and it is used here, as in previous articles listed in the Bibliography. For the people, the term Fulani, though open to objection, has been retained for both singular and plural in view of its widespread use in historical and anthropological books as well as in Nigerian government publications.
Some Fulani still live the nomadic life inherited from their forebears, moving with the seasons from wet-season to dry-season grazing grounds and back; such nomadic groups normally form a relatively small proportion of the population, outnumbered by people of other ethnic groups. Others have through the years taken to a more sedentary way of life, combining agriculture to a greater or less degree with their pastoral activities.
In some places the processes of settlement and concentration began many centuries ago, and today there are areas (such as Fuuta-Jalo 2 in northern Guinea, northern Senegal, some parts of Mali and Upper Volta, Gwandu and Gombe Emirates in Nigeria, and parts of Adamawa) where the population is predominantly Fulani, and there are long-established and fully organized Fulani communities varying in size from small villages to towns as large as Labe and Dabala, Kaedi, Matam, and Podor in the west, Djenne, Mopti, and Bandiagara, Dori and Djibo in the bend of the Niger, and Birnin Kebbi, Gombe, Yola, and jalingo, Marua and Garua in the cast. In these areas modern western-type education now complements the Islamic learning traditionally held in high respect, and Fulani are to be found in many responsible positions, among politicians and professional men as well as in the public service, playing an important part in the progress of West Africa's newly independent states, as their ancestors once did in the history of the empires of the Western Sudan.
The establishment of settled communities has usually encouraged the continued use of the Fula language, even if somewhat modified by borrowings from neighbouring languages. But in northern Nigeria, where the Holy War of Usman dan Fodio ('Usumaanu ɓii Fooɗuye) established Fulani dynasties in the former Hausa states, the relatively small numbers of the Fulani ruling class in most cases led to their being absorbed by the Hausas both culturally and linguistically. Consequently, in many provinces of northern Nigeria Fula is spoken only by the ubiquitous pastoral Fulɓe na'i (‘cattle Fulani’) and in relatively small communities of recently settled Fulani; but in Gombe, Gwandu, Misau, and Adamawa the greater concentration of settled Fulani has so far helped the language to withstand the advance of Hausa more successfully, and it is used much more widely.
Linguistically there is no difficulty in describing the speech of the Fulani diaspora as a single language, having a common basic morphological and syntactical structure and a common lexical stock which is surprisingly large and uniform, considering the geographical spread and the variety of other linguistic groups with which the Fulani have been in close contact. Such differences as exist are mainly phonetic, phonological, and lexical (especially in the cultural vocabulary, where the tendency to borrow from neighbouring languages is strong), and to a smaller extent in the shape of some of the morphological elements. These are clearly to be regarded as no more than dialectal differences, and do not bulk large enough to justify treating the different varieties of speech as separate languages. There is a considerable degree of interintelligibility, particularly between groups living relatively near to each other, and even Fulani of Guinea or Senegal and of Nigeria have little difficulty in understanding much of each other's speech, given a certain degree of intelligence and a brief period of adjustment.
The demarcation of dialects is inevitably an arbitrary process, especially in view of the mobility of the nomadic Fulani; but for practical purposes it is convenient to distinguish six main dialect areas:

  1. Fuuta-Tooro (Senegal)
  2. Fuuta-Jalon 3 (Guinea)
  3. Maasina (Mali)
  4. Sokoto and western Niger
  5. ‘Central’ northern Nigeria (roughly Katsina, Kano, Zaria, Plateau, Bauchi, and Bornu Provinces) 4 and eastern Niger
  6. Adamawa

The Fula of other areas approximates more or less to one or more of the above six dialects, that of Upper Volta, for instance, combining features of Masina and Sokoto, while the varieties spoken in Portuguese Guinea, Mauretania, and Dahomey resemble the Fula of Guinea, Senegal, and Sokoto respectively, while at the same time having some idiosyncratic features of their own 5.
As can be seen from the selected Bibliography on p. 421 ff., classified roughly on a dialect basis, Fula has been the subject of a considerable number of grammatical works in English, French, and German, the first published well over a century ago. Of the earlier grammars, the majority deal with the two most westerly dialects-those of Fuuta-Toro and Fuuta-Jalon—and that of Adamawa. Written by linguistically gifted explorers, missionaries, military administrators, educationists, and others, these are mostly on traditional lines, but contain much that is still of value, especially in view of the dialectal variations they reveal. More recently the series of articles, on phonology and various aspects of the nominal system, by the late August Klingenheben, following in Westermann's steps, have brought a scholarly precision and thoroughness as well as a breadth of coverage to the study of Fula. His work in this field culminated in his grammar of the Adamawa dialect, Die Sprache der Ful (1963), where the nominal system in particular receives exhaustive treatment. Labouret's La Langue des Peuls on Foulbé (1952) is also a useful work, less detailed in some respects, but of wider scope, its notes on dialectal variations being particularly valuable. The only recent grammars written in English are the 1953 reprint, with some revision, of Taylor's 1921 grammar of the Adamawa dialect, and Lloyd Swift's recent pedagogic Basic Course (on the Gambia variety, 1965) 6. There is a place, therefore, for a reference grammar in English, dealing more particularly with the whole scope of the nominal and verbal systems as they appear in the light of recent research, and based on the dialect of 'central' northern Nigeria-a dialect which, apart from the series of articles listed in the Bibliography, has hitherto received little attention.
The present study deals with the variety of Fula spoken in Gombe Division, the most easterly Division of Bauchi Province in the northern region of Nigeria 4. Though in vocabulary it is fairly close to the Adamawa dialect, its eastern neighbour, it is in many respects typical of the ‘central’ Nigerian dialect just mentioned. And in fact its morphology is much more typical of Fula as a whole (especially the varieties spoken in Mali, Upper Volta, and Niger) than is the Adamawa dialect, with which a number of the previous studies have been concerned. For the present purpose Gombe Fula is treated as a unity, although there are some relatively minor differences within the area (e.g. between the Fula of Dukku and that of Nafada or Kumo or Ako). These differences are chiefly phonetic and lexical, and are only marginally relevant here.
This study is based on detailed investigation of the speech of Malam Jalo Gombe (now Alhaji Ibrahim Jallo, Wazirin Gombe) and Malam Ibrahim Abubakar, followed by field work with many Fulani in Gombe Division in 1955, and subsequent conversation and research with other Gombe Fulani in London and elsewhere. Where generalizations are made about the relative frequency of various forms, and about'common' or 'normal' patterns, they are based on texts supplied by my first two informants and on personal observations of the speech habits of Gombe Fulani in many different contexts in Nigeria and elsewhere.
The complicated morphology of the nominal and verbal systems, with which this book is concerned, is the most conspicuous feature of Fula; and the two systems together affect a very large proportion of any stretch of the spoken language, since the great majority of words belong to one or other system, or to both.
The nominal class system, with its twenty-five or so classes marked by distinctive morphological elements—the exact number of classes varying to some extent according to dialect—covers not only nouns but also pronouns and pronominal elements of various types, adjectives, demonstratives, interrogatives, and even numerals. All of these undergo various morphological changes according to the class of the noun to which they refer.
The verbal system comprises not only fifteen different ‘tenses’ marked by different verbal suffixes and by various other morphological features, but also three distinct ‘voices’ or series of tenses—Active, Middle, and Passive—each containing a complete or almost complete set of tenses. It also includes a series of nineteen radical extensions (similar in general function, though not in detail, to Bantu radical extensions), which can occur singly and in combination. These nominal and verbal systems overlap in so far as, firstly, subject and object elements occur within the compass of the verbal ‘complex’ (see Chapter 28), but are at the same time integral parts of the nominal system; and secondly, infinitives and participles (here together termed verbo-nominals) are hybrid forms, each having two suffixes, one belonging to the verbal system and one to the nominal system.
The morphological systems referred to above, however, cannot be fully described without reference to various morphophonemic patterns which play an important part in both nominal and verbal systems, and to the underlying phonological features of the language. Included in the latter are certain aspects of sentence intonation patterns (Fula not being a ‘tone language’ in the sense in which that phrase is normally used); for some intonation features play a significant if subsidiary role in the verbal system, and to a smaller extent in the nominal system also. The bulk of the book (viz. most of Parts IV, V, and VI) is therefore concerned with the analysis of the various types of nominal, verbal, and verbo-nominal forms and their constituent parts, and the various permutations and combinations which occur, with greater or less regularity, in the two systems. This analysis is preceded by an outline of the basic phonological features of consonants and vowels and the morphophonemic patterns in which they are involved, together with a brief account of the pausal feature here called ‘final glottality’ and of the relevant aspects of sentence intonation patterns (Part III).
But these phonological and morphological features, fundamental though they are, are no more than the dry bones of the systems. Of equal importance in a full description is the way in which the systems operate in the everyday use of the language, and the roles that the various nominal and verbal forms play in this process.
To some extent these roles can be described in formal terms by reference to the positions that nominals, verbals, and verbo-nominals of various categories can occupy in various types of sentence. Accordingly the grammatical introduction (Part II) contains, besides a summary of the main features of the two systems and of the various syntactic categories and form-classes found in the language, an account of the way in which nominals and verbo-nominals can combine in nominal and adverbial groups, an analysis of the structure of various types of sentence, and an indication in general terms of the positions which nominals, verbo-nominals, and verbals can occupy in these structures. This general account is supplemented later by paragraphs in the relevant chapters in Parts IV-VI, giving more specific indications of the positions in sentence structure occupied by individual types of nominals and verbonominals (Chapters 22-7 and 60), of the syntactical behaviour of the three voices (Chapter 45) and the various tenses (Chapters 46-56), and of the syntactical implications of certain radical extensions (Chapters 57-8).
Language, however, is essentially a means of communication, at least of self-expression; and, given a variety of possible forms of various categories, the actual use of one form rather than another depends only partly on strictly grammatical factors. Very often the choice is determined by the meaning which the speaker intends to convey. The meanings of individual forms, of course, normally belong to the lexicon; but where generalizations can be made about the meanings associated with particular formal categories, such generalizations have a place in a description of the grammar of the language. Formal characteristics are, of course, of primary importance, and the grammatical categories must be based essentially on formal, not semantic, criteria. But to limit the description, in this instance the description of the nominal and verbal systems of Fula, to an account of their formal characteristics only, without a complementary account of the regular semantic associations, would be to give only a partial picture of the systems as they operate in the living language.
Accordingly some account is here given, at appropriate points, of the meanings associated with various categories of nominals and verbo-nominals (Chapters 22-5 and 60), and with individual nominal classes (Chapters 13, 16). And the description of the verbal system in Part V includes an indication of the meanings associated with the three voices (Chapter 45), and a fairly full account of the meanings both of the various tenses (Chapters 46-56) and of the individual radical extensions (Chapters 57-8).
The Appendices contain a certain amount of relevant but supplementary material which would be out of place in the body of the work. Some deal with certain phonetic, phonological, and semantic details which are of indirect rather than direct relevance to the main theme, while others contain lists of other forms referred to—such as adverbials, and prepositions—or a more detailed breakdown of certain categories of nominals and some types of nominal stems than is necessary in the main study. Others again consist of tables summarizing features discussed in several different chapters, or containing representative examples of verbal complexes and verbal radicals of various types.
Finally, I would like to express my appreciation of all the help and encouragement I have received from a great many people—particularly my colleagues in the Department of Africa at the School of Oriental and African Studies, with whom many points have been discussed as they arose, and the various Fulani who have so patiently and understandingly collaborated with me at various stages, especially Alhaji Ibrahim Jallo, Wazirin Gombe, Malam Ibrahim Abubakar, Alhaji Muhammadu el-Nafaty, and the late Malam Salihu Kumo. They have not only provided the essential raw material on which this book is based, but have helped me to an appreciation of the beauties of a wonderfully rich language.

Notes
1. L. Homburger. “Les représentants de quelques hiéroglyphes égyptiens en peul” Bull. de la Société linguistique de Paris, 1930
“Eléments dravidiens en peul”, Journ. Soc. Afric. 18, 2 (1948),135-43
M. D. W. Jeffreys, “Speculative origins of the Fula language” Africa 17, 1947, 47-54
F. W. Taylor, A first grammar of the Adamawa dialect of the Fulani language, O.U.P., 1921
H. G. Mukarovsky, Die Grundlagen des Ful und das Mauretanische, Vienna, Herder, 1963.
2. See n. 3 on p. 3.
3. Fuuta-Jaloo is the authentic local form of the name, so Alfâ Ibrahîm Sow confirms, although the French spelling is normally Fouta-Diallon or Dialon, anglicized as Fuuta-Jalon or in various other ways. Since long vowels are not normally marked in English the spelling Fuuta-Jalon has here been used in the English text.
4. Under the reorganization of 1967, Katsina and Zaria Provinces together form North-Central State, Kano Province is now Kano State, Bauchi, Bornu, and Adamawa Provinces together form North-Eastern State, while Plateau Province is part of Benue-Plateau State.
5. This tentative classification is suggested on the basis of a pilot survey carried out in 1955-6 during a year's tour of Fulophone areas from Nigeria, through Niger, Dahomey, Upper Volta, and Mali, to Senegal, Gambia, and Guinea, as well as on the basis of the available literature. Each of the dialects could well be further subdivided.
6. See also Leslie H. Stennes's A reference grammar of Adamawa Fulani, African Language Monograph, No. 8, African Studies Center, Michigan State University, 1967, which appeared when the present work was in the press.

Part II. — Grammatical Introduction

2. — Syntactic categories and form-classes

2.1. The structure of Fula sentences can be analysed in terms of Subject, Predicator, Object, Complement, and Adjunct, as described in Chapter 7. The various forms which are exponents of these categories are of differing lengths, and may be classified as follows : words, components, and elements {these two being sub-forms normally occurring only within complexes, or other longer forms), complexes and groups, and particles. Of these, words, components, and elements will sometimes be referred to jointly as primary syntactic categories:

A word is the smallest stable form, stable here meaning that it can stand alone in response, e.g. Bello (a man's name), deptere / deftere (FJ) ‘book’, suudu ‘hut’, hanko / kanko (FJ) ‘he’, ton ‘there’, binnduɗo / winnduɗo (FJ) ‘one who wrote’.

Components are unstable—they cannot stand alone even in response, but occur only in complexes, e.g. nder ‘in’, maako / makko (FJ) ‘his’, maaru / mayru (FJ) ‘its, of it.’ 1

Elements are also unstable forms, but mostly of simpler structure than components (viz. monosyllabic forms with structure CV or CVC), e.g. nde, 'o in the examples below 2.

Complexes, which are stable, are ‘close’ combinations (see below) of various kinds, such as :

Two words, e.g., deptere Bello / deftere Bello (FJ) : book of Bello, Bello's book

Word + component, or component +word, e.g.
deptere maako / deftere makko (FJ) book of-him, his book
nder suudu in the-hut
Word+ element, or element+ word, e.g.
binnduɗo / winnduɗo nde (FJ) one-who-wrote it
nde Bello that-of Bello, Bello's one
Element + component, or component + element, e.g.
nde maako / makko (FJ) that-of his, his one
'o-wartii / o 'artii (FJ) he/she has-come-back
waddu-ɗum / 'addu ɗum (FJ) bring it!
Two components, e.g.
nder maaru / nder mayru (FJ) inside of-it, in it

(There are also longer complexes in which one of the items is itself a complex, or a group of some kind, cf. § 6.4, Type A.)

The structure of each kind of complex is fixed, and normally there is close nexus, or cohesion between the constituent parts, such that nothing can intervene which is not included in the structure of such a complex.

Group is an inclusive term used to cover a variety of forms, all of which occupy the same position in a sentence; in any given instance a group may be represented by a word, a complex, or certain specific combinations of words, complexes, components, andfor elements. Groups are of two kinds—nominal groups (see §§ 6.1-11) and adverbial groups (see §§ 6.11-15).

A particle is an unstable form which occurs only in combination with other forms, but is not part of the basic structure of a complex, group, or sentence, as are components and elements, and is syntactically less restricted. The main types include certain sentence particles, pre-clause particles (conjunctive particles) and post-clause particles, linking particles, pre-group particles, and intensifying particles (ideophones). Particles are relevant here in so far as there is syntactic correlation between some particles and the range of tense suffixes which can be used with them (see § 7.20 (B. iii), Chapters 47-56 passim and Appendix 16).

2.2. Cutting across the primary syntactic categories given abovewords, components, and elements-are certain other categories, or form-classes, the main distinction being between nominals, verbals, verbo-nominals, adverbials, and prepositions. Parallel to this is the sub-categorization of particles referred to in the last paragraph.

2.3. The various types of word, component, and element in each form-class are listed in Table 1 (see also § 3.2 for the various types of nominal). The majority of these types—in fact all except nouns, adjectives, verbals, and verbo-nominals—comprise limited systems, which consist of a relatively small number of items; these can be defined by listing, although in the case of some types of nominal (e.g. demonstratives) it would be possible to define them by reference to their structure. Lists of such forms are given in Appendices 3 (nominals), 17 (adverbials), and 18 (prepositions), and the structure of various types of nominals is described in §§ 22.2, 23.2- 5, 24.2-8, 25.2-5, 25.9. Full nominals (nouns and adjectives), verbals, and verbo-nominals (both infinitives and participles) belong to open systems, and cannot be defined by listing, but can all be defined by their structure, and by reference to one or more of the series of nominal andfor verbal suffixes, viz.

Nouns and adjectives consist of a stem 3 + a suffix belonging to one of the series of nominal suffixes listed in Table 6, p. 88. For the distinction between nouns and adjectives, see Chapter 15.

Verbals consist of a radical 3 + a suffix belonging to one of the series of tense suffixes or imperative suffixes listed in Appendix 10, with or without subject, object, or preterite elements as described in Chapters 37 and 38.

Verbo-nominals consist of a verbal radical together with both a verbal suffix belonging to one of the series listed at the foot of Appendix 10 and a nominal suffix belonging to one of the series listed in Table 6, p. 88.

Arnott. Nominal and Verbal systems of Fula. Syntactic categories and form-classes

2.4. Two minor systems consist of the quasi-verbal words hiin, ndaa, and ndikka, and the stabilizing elements ɗon and ton, referred to in § 7.8 (d), (e), and (b).

Notes
1. The term ‘component’ is also used occasionally in a looser sense to refer to the constituent parts of verbal forms.
2. The term ‘element’ is also used occasionally in a looser sense, e.g. to refer to the constituent parts of nominals.
3. ‘Stem’ is used throughout this book to refer to the fundamental part of nominals, ‘radical’ to refer to the fundamental part of verbals. Verbo-nominals, being hybrid forms, have a stem which itself includes a radical, cf. § 60.1.

3 — Outline of the nominal system

3.1. The basis of the nominal system of Fula (the essential features of which are set out in Appendix 3) is a class system similar in kind to class systems found in many West African languages as well as in Bantu languages. Nouns, and nominals of most other types, fall (in Gombe Fula) into 25 different classes on the basis of agreement between nominals having the same referent. Thus wudere ‘cloth’ and sawru ‘stick’ are assigned to different classes on the basis of the different patterns of agreement illustrated below:

(a) wudere mawnde nde'e wudere ndeye? sheede maare / wudere mawnde nde'e wudere ndeye? coggu mayre (FJ)
cloth big this cloth which? price of-it

(b) sawru mawndu ndu'u sawru nduye? sheede maaru / sawru mawndu ndu'u sawru nduya? coggu mayru (FJ)
stick big this stick which? price of-it

The agreement is operated by concord markers which

(i) vary from class to class (e.g. -re, -nde, nde'e, nde- for the class to which wudere belongs, as against -ru, -ndu, ndu'u, ndu- for the class to which sawru belongs), and

(ii) are of a different kind in different types of nominal. Thus in some (e.g. wude-re and maw-nde, saw-ru and maw-ndu) the concord marker is a suffix;

in others (e.g. nde-ye, ndu-ye / nde-ya, ndu-ya (FJ)) it is the first of two constituent elements 1;

in others (e.g. maa-re, maa-ru / mayru, mayre (FJ) it is the second of two constituent elements 1;

While in the case of others (e.g. nde'e, ndu'u) the whole form marks the concord.

Since there are 25 different classes, the concord markers of any given type of nominal constitute a substitution-series of 25 different forms. (In most cases the actual shapes are different, but there are a few pairs of concord markers having the same shape but distinguished by other formal differences (cf. § 17.6). Numerals, however, constitute a separate sub-system within the nominal system, a sub-system of four classes with a regular pattern of correspondence to the classes of the main system, and a distinct series of concord markers; (see § 27.10.)

3.2. The nominal system includes nominals of all the three primary syntactic categories referred to in § 2.1 — some words, some components, and some elements, as well as a special type of suffix which it is convenient to include in the general term ‘nominal’:

Words Components
Nouns ‘Full nominals’ Possessive pronouns +
Adjectives Referentials Sp.
  Elements
Independent pronouns + Subject elements +
Near demonstratives Sp. Object elements +
Far demonstratives Sp. Genitive elements
Interrogatives Sp. Relative elements
  Suffixes
Simple numerals 2 Possessive suffixes +

3.3. Some of these types of nominals may be grouped together in other ways on different bases. Thus for certain purposes it is useful to refer jointly to the following as ‘Specifiers’ (as indicated by the entry Sp. in § 3.2): near and far demonstratives, referentials, and interrogatives. Again, the nominals marked + in the list in § 3.2 consist of two sub-series-{a) a set of concordant forms which enter into the system of agreement described in § 3.1, and a smaller set of non-concordant forms which do not contain concord markers, viz. xst and 2nd person forms, and the non-concordant 3rd person forms described in §§ 22.6, 26.1 ff. The latter are here treated as being outside the agreement system, although they are included in the nominal system as a whole, because their behaviour in general resembles that of the corresponding concordant forms.

3.4. Of the concordant nominals, full nominals (such as sawru, mawndu above) and numerals consist of a stem (which is subject to alternation 3 in the initial consonant 4, and in some cases to other morphophonemic variations ) 3 and a concord suffix, with morphophonemic variations in certain cases. The concord suffixes, or nominal suffixes, fall into four series, to be called ‘grades’, as shown in columns 4-7 of Appendix 3. Nominals of other types have a variety of different structures, as set out in § 22.2, again with morphophonemic variations in some cases.

Table. Fula nominal system

3.5. The ensuing description of the morphology of the nominal system and its operation within the sentence (Part IV) will therefore involve an account of

  1. the class system in general, the relation of the various classes to one another, and the extent to which generalized functions and meanings may be assigned to individual classes or groups of classes (Chapters 12-16)
  2. the morphology of full nominals and numerals, including the patterns of initial-consonant alternation (Chapters 18 and 27 §§ 8- 13).
    other features of the stem, including morphophonemic variations (Chapter 19), the various series of concord suffixes and their morphophonemic variations (Chapters 17, 21, and 27, §§ 8-13), and patterns of correlation between stem and suffix (Chapters 17, 20)
  3. the shapes of other nominals and the morphological analysis of such as are capable of analysis, and morphophonemic variations where these occur (Chapters 22-5)
  4. certain features of the syntactical behaviour of the various types of nominals, in so far as it is correlated with morphological features, or with morphological categories (Part IV passim).

3.6. Nominals of all kinds will be written as separate words 5. The constituent parts will sometimes be separated by a hyphen, where stem and suffix or other parts are under special consideration; otherwise they will be written as single units without a hyphen.

Notes
1. These forms are not analysed in terms of stem and prefix or suffix, because of the brevity of the second element in ndeye and other comparable forms, and because, in contrast with the stems of nouns and adjectives, no specific meaning can be assigned to -ye or maa-.
2. See §§ 27.2-3
3. ‘Variation’ and ‘variant’ are here used as general terms, and also in the special phrases ‘free variation’ and ‘free variant’. ‘Alternation’ is used for systematic variation, where the altemants are in complementary distribution correlated with morphological or phonological features, or with certain individual forms. ‘Alternances’ are specific sets of alternants, e.g. the w/g/ng altemance referred to in note 4 below.
4. wur-o: a compound (class 9); gur-e: compounds (class 24); ngur-oy small compounds (class 6) where the initial consonant is variously a continuant (w), a plosive (g), and a nasal compound (ng). Cf. Chapters 8 and 18, esp. §§ 8.4-10, 18.3.
5. Except when subject and object elements occur within verbal complexes. Cf.§ 4-10.

4 — Outline of the verbal system

4.1. While a few verbal forms (referred to in the next paragraph) rank as words, the majority of them are here analysed as complexes, because of the inseparability of their constituent parts and their frequent morpho-phonemic interdependence.

4.2. The core of every verbal form is a verbal base (hereafter referred to simply as a base) consisting of a radical 1 + a tense suffix-the various series oftense suffixes (listed in Appendix 10) being the typical morphological elements of the verbal system. Some forms-such as the imperative forms of intransitive verbs—consist of a base only, and may be analysed as words, e.g. wartu / 'artu (FJ) 'come back!', jooɗa / jooɗo (FJ) 'sit down!'; the remainder are verbal complexes consisting of a base (here a component) and one or more of three types of element—subject element, object element, and preterite element.

4.3. Imperative complexes consist of base + one or two object elements, e.g. waddu-ɗum / 'addu-ɗum (FJ) 'bring it!', hokku-mo-ɗum /'okku-mo-ɗum (FJ) 'give him it!' Other verbal complexes consist of subject element and base (e.g. 'o-warii / 'o 'arii (FJ) 'he has come', 'o waray / 'o 'aray (FJ) 'he will come'), or subject element and base together with (i) a preterite element, or (ii) one or two object elements, or (iii) both preterite and object element or elements. Examples :

(i) 'o-warii-no / 'o 'arii-no (FJ) he had come
'o-waray-no / 'o 'aray-no (FJ) he was about to come
(ii) 'o-waddii-ɗum / 'o 'addii ɗun (FJ) he brought it
'o-hokkii-mo-ɗum / 'o-'okkii-mo-ɗun(FJ) he gave him it
(iii) 'o-waddii-no-ɗum / 'o-addii-no-ɗun (FJ) he had brought it
'o-hokkii-no-mo-ɗum / 'o-okkii-no-mo-ɗun (FJ) he had given him it

4.4. Of the two constituents of the base, the tense suffixes are the principal morphological features distinguishing one tense from another (e.g. -ii as against -ay in (i) above). They fall into three distinct series —the Active, Middle, and Passive voices—with different syntactical behaviour and to some extent contrastive meaning; and, cutting across the voice-distinction are other sub-groupings—based on common phonological, morphological, syntactic, or semantic features-including ‘relative’ and ‘non-relative’ tenses, positive and negative, past and nonpast, etc. (cf.§§ 29. 4-5).

4.5. Radicals are of various types, including simple radicals (e.g. war- / 'ar- ‘come’) and extended radicals, analysed as basic radical+ radical extension (e.g. wart- / 'art- (FJ) ‘come back’ < war- / 'ar- (FJ) ‘come’ + the reversive extension -t-, janggin- ‘teach’ < janng- ‘read, learn’ + the causative extension -n-/-in-).
The radical extensions form a closed system of nineteen different extensions with distinct meanings and to some extent differing syntactic behaviour (cf. Chapters 57-9). Radicals are subject to certain morphophonemic variations, particularly:

  1. alternation in the initial consonant, as in 3rd person singular 'o-warii / 'o-arii (FJ) ‘he came’ as against 3rd person plural ɓe-ngarii / ɓe 'arii (FJ) ‘they came’ (cf §§ 8.5-10 and 35.43-4)
  2. variation in the shape of radical extensions, as in 'o-maɓɓitii-ɗum / o-udditii-ɗun (FJ) ‘he opened it’ as against maɓɓutu-ɗum / udditu-ɗun (FJ) ‘open it!’ (cf. §§ 9.19 and 57.4-9).

There are also morphophonemic variations in many tense suffixes variations correlated sometimes with features of the radical, and sometimes with features of other elements in the complex.

4.6. The arrangement of the constituents of the complex (i.e. their order of occurrence) is not constant. The most common arrangement is that shown in the above examples, viz. subject element + base + preterite element (if any) + object element(s) (if any); but different arrangements are found in certain definable circumstances, e.g. nde ngartu-ɗaa? / nde 'artu-ɗaa? (FJ) ‘when returned-you?’ (Relative Past tense, Active, in which the second singular subject element follows the base).

4.7. Subject and object elements (listed in Tables 16 and 19, pp. 194, 212) are series of nominal elements, with two sub-series, one consisting of concordant 3rd person forms which are part of the nominal agreement system, and one consisting of non-concordant 1st, 2nd, and 3rd person forms, as described in §§ 22.4-6, 25.2-3 and Chapter 26.

4.8. The preterite element has two alternant forms -no and -noo, their incidence being statable partly in grammatical terms and partly in phonological terms (cf. §§ 38.8-11).

4.9. The ensuing description of the morphology of the verbal system and its operation within the sentence (Part V) will therefore include an account of:

  1. the tense system in general, including the various cross-categories, such as the three voices, positive and negative, etc;
  2. the base, including
    1. the radical—the various types and their phonological structure and morphophonemic variations (including initial-consonant alternation), the various radical extensions, their shape and morphophonemic variations, and their syntactic behaviour
    2. the tense suffixes, their shape and morphophonemic variations
  3. the series of subject and object elements and their distribution and morphophonemic variations
  4. the preterite element, its uses, its two forms and their distribution
  5. the arrangement of the various components of the verbal complex

It will also include an account of the relevance to the tense system of the presence or absence of final glottality (as described in Chapter 11) and systematic variation in the position of the salient syllable — the syllable which can coincide with a peak in the sentence intonation pattern (cf. Chapter 10)—and, finally, an indication of the main uses of the various tenses, described in terms of meaning and of syntactical behaviour.

4.10. The components of verbal complexes will normally be separated by hyphens; the base, however, will be written as a single unit, unhyphenated, except in some cases where radical and/or suffix are under special consideration.

Note
1. Cf. n. 3, p. 11.

5 — Verbo-nominals

5.1. Verbo-nominals-which are of two main types, infinitives and participles-are hybrid words consisting of a verbal radical+ a tense suffix + a nominal suffix, with the potentiality of a preterite element between the two suffixes. Examples:

Infinitives
loot-u-ki / loot-u-de to wash
loot-aa-ki / loot-itaa-de to wash oneself, to get washed
loot-ee-ki / loot-ee-de to be washed
Participles
loot-u-ɗo (one) who has washed (something)
loot-u-ɓe (those) who have washed
loot-u-noo-ɗo (one) who had washed
loot-ii-ɗo (one) who has washed himself
loot-otoo-ɗo / loot-oytoo-ɗo (FJ) (one) who will wash himself
loot-aa-ɗo (one) who has been washed

5 .2. Fundamentally, a verbo-nominal may be regarded as a special type of nominal, with a stem consisting primarily of verbal forms. The nominal suffixes belong to two of the regular series of concord suffixes found in full nominals (cf. §§ 60.7, 14), and alternation of the initial consonant of the stem follows the nominal rather than the verbal pattern; but the tense suffixes form a small system corresponding to the tense suffixes of verbals and reflecting differences of both tense and voice, and morphophonemic variations within the stem follow, in most cases, the pattern found in verbals.

5.3. Syntactically, verbo-nominals resemble verbals in some respects (e.g. the relation between a verbo-nominal and a nominal dependent on it, cf.§ 60.20), but in other respects they resemble nominals-infinitives occupying the same position in sentence structure as nouns, participles the same position as adjectives.

5.4. The description of verbo-nominals in Part VI will therefore include an account of all these features — radical, tense suffixes, preterite element, nominal suffixes, and the morphophonemic variations involved, together with an account of relevant syntactic behaviour.

6 — Nominal and adverbial groups

Nominal groups

6.1. Whereas the role of verbals in sentence structure is relatively easy to describe (cf. § 7·3), the role of norninals is more complicated, and can be most satisfactorily and concisely described by means of intermediate categories, nominal and adverbial groups, which are described in this chapter.

6.2. A nominal group may consist of any of the following:

  1. a sole nominal
  2. a genitival complex, of which there are two types
  3. a concordant nominal chain (hereafter called simply a nominal chain)
  4. a relative clause
  5. a verbo-nominal word or phrase
  6. an appositional group multiple nominal groups
  7. a serial nominal group multiple nominal groups

Of these, a ,f, and g are of little further relevance to the present study, and will simply be exemplified without further discussion, but b, c, d, and e require more detailed treatment.

6.3. A sole nominal may be a noun or any other nominal or verbo-nominal word. Examples:

noun Bello laamɗo chief na'i cattle
adjective daneeji white ones (e.g. cattle)
independent pronoun miin / min (FJ) I 'aan / an (FJ) you hanko / kanko (FJ) he/she
simple numeral 1 ɗiɗi two sappo ten
interrogative moye? / hombo (FJ) who? ndeye? / honnde? (FJ) which one?
near demonstrative 'o'o this one (person)
nde'e this one (e.g. book)
far demonstrative 'oya that one
infinitive jannguki / janngude (FJ) to read, reading
participle binnduɗo / winnduɗo (FJ) (one) who wrote
kiiɗɗi / kiɗɗi old (ones) (hiiɗ- / hiɗ- = be old)

6.4. A genitival complex consists of two items in close nexus, i.e. nothing can intervene between them. Examples:

laamɗo Dukku Chief of Dukku na'i maako / makko (FJ) his cattle
daneeji maako / makko (FJ) his white ones
mo Dukku that of Dukku ɗi maako / makko (FJ) his ones

Genitival complexes are of two kinds, Type A, such as the first three examples above, in which the first item may be a nominal of several different types, and Type B, in which the first item is always a genitive element, such as mo, ɗi in the last two examples above.

In Type A complexes the first item may be:

A noun laamɗo Dukku na'i maako / makko (FJ) (as above)
na'i moye? / hombo (FJ) whose cattle?
An adjective daneeji maako / makko (FJ) his white-ones (e.g. cattle)
A numeral ɗiɗi maɓɓe twice their number (lit. two-of-them)
A verbo-nominal
Infinitive hatuki Bello Bello's courting
Participle kiiɗɗi maako / kiɗɗi makko (FJ) his old-ones
jeyaaɓe maako / makko (FJ) his slaves (lit. possessed-ones)

The second item may be (i) any of the nominal words which can occur as sole nominal, except an independent pronoun, (ii) a possessive pronoun (e.g. daneeji maako, etc. above), or (iii) any of the other types of nominal group, including another genitival complex (e.g. na'i laamɗo Dukku ‘the cattle of the Chief of Dukku’) 2.

In Type B complexes, the first item, as stated, is always a genitive element, this being the only circumstance in which this element occurs; the second item is as in Type A. Examples are mo Dukku ‘that of Dukku’, ɗi maako ‘his ones’, ɗi laamɗo Dukku ‘those of the Chief of Dukku’, ɗe wartuki / artuɗi (FJ) ‘(money) for returning’, ngol 'oya ‘that man's one’ (e.g. gown).

6.5. A nominal chain is a chain of nominals in agreement with one another, the agreement being operated by concord markers, italicized in the following examples:

I II III
wudere raneere nde njogii-mi nde'e
cloth white which I'm-holding this
na'i boɗeeji (ɗiɗi) ɗin
cattle red (two) the the = the (two) red cattle
duuduki belki / welki (FJ) ki'i
piping sweet this = this sweet piping
I III
boɗeeji ɗi'i
red (ones) these = these red ones

What is of primary relevance here is the agreement between the various nominals in the chain, this being the main distinction between a nominal chain and other sequences of nominals, such as genitival complexes and appositional groups, where there is no agreement. Of secondary importance is the restricted distribution of the various types of nominal within the chain, as set out in Table 2. Briefly, three positions may be distinguished, I Head, II Qualifier(s), Ill Specifier, as indicated in the above examples, and it is noteworthy that some morphological distinctions are paralleled by differences in the potentiality of occurrence within a nominal chain.

I II III
Head Qualifier(s) Specifier
Noun or   Near demonstrative
Infinitive or   or
Infinitival phrase or Adjective or Far demonstrative
Adjective or Participle or or
Participle or Participial phrase or Interrogative
Participial phrase or Relative clause or or
Relative clause or Numeral or Retrieval
Numeral or    
Genitival complex A or    
Genitival complex B Genitival complex B  
(Mutually exclusive) (Potentially exclusive) (Mutually exclusive)

Thus nouns and adjectives, distinguished on morphological grounds in § 1 5.2, differ syntactically in that while adjectives (such as raneere, boɗeeji) occur in either I Head or II Qualifier position, nouns (such as wudere, na'i) can occur only in Head position. Again, the four types of specifier (near and far demonstratives, referential, and interrogative) are confined to III Specifier position (where they are mutually exclusive), and nothing else can occur there; and while genitival complexes of type A occur only in Head position, complexes of type B can occur in either Head or Qualifier position. Examples:

I II
(Type A) na'i maako / makko (FJ) boɗeeji
red
  cattle his = his red cattle
(Type B) ɗi maako / makko (FJ) boɗeeji
red
  his ones = his red ones
  na'i boɗeeji ɗi maako / makko (FJ)
  cattle red of his = the red cattle of his, his red cattle

6.6. A relative clause is a clause (§ 7.18) introduced by a relative element (cf §§ 25.11-13), this being a nominal element marking agreement. Examples:

nde njogii-mi which I am-holding
ɗe Bello waddani / 'addani (FJ) ɓiɓɓe 'am which Bello brought-for my children
ndu / ko ɓe nyaamata hannde which they will-eat today
mo ngii-ɗaa / yii-ɗaa keenya / hanki whom you saw yesterday

Like other clauses, a relative clause may include an object or objects, such as ɓiɓɓe 'am, and/or adverbial groups such as hannde, keenya / hanki (FJ). On the other hand, there are certain restrictions on the verbal tenses occuring in relative clauses (§§ 25.17, 55.1-2), and also certain general restrictions on their form and use (§§ 25.13, 15-16).

6.7. Verbo-nominal phrase. Parallel to the morphological distinction between a nominal and a verbo-nominal (§§ 5.1- 2) is a syntactical distinction. A verbo-nominal, like a verbal but unlike a nominal, can have one or more objects and/or adverbial groups dependent on it. Exemples:

(participial) shooduɗo / sooduɗo (FJ) na'i (hannde)
one-who-bought cattle (today)
gadd(an)ɗo / 'addanɗo (FJ) (ɓiɓɓe 'am) nyaamdu
one-who-brought (for my children) food
(infinitival) resuki kare maɓɓe 'e lesdi / rowtiri donngal maɓɓe 'e leydi (FJ)
to-put their loads on the ground

The term ‘verbo-nominal phrase’ is used to refer to the whole of such a phrase, consisting of a verbo-nominal together with any dependent object{s) and/or adverbial groups. Note that whereas a verbo-nominal phrase can occur in position I or II in a nominal chain, or in 2nd position in a genitival complex, only a sole verbo-nominal, without dependent forms, can occur in 1st position in a genitival complex.

6.8. An appositional group consists of two or more items, not necessarily in the same nominal class, and without any other grammatical connection, e.g.

soobaajo 'am / giɗo an (FJ) | moodibbo | 'Umaru my friend Malam Umaru
Biɓɓe maako / makko (FJ) | Bello 'e Towɗo 'e 'Umaru his children Bello and Towɗo and 'Umaru
kuuje / piiji (FJ) tati, | 'ƴiiƴam 'e haayre 'e haako three things blood and stone and leafage
(i.e. animal and mineral and vegetable.)

(| indicates the boundary between the different items.)

6.9. A serial nominal group is a series of items linked by 'e ’and‘ or koo ‘or’ placed between each pair of items, and potentially before the first item, e.g.

Bello 'e Towɗo 'e 'Umaru (§ 6.8 above) 'ƴiiƴam 'e haayre 'e haako ('e) janguki 'e winnduki / (no) jangude e windude (FJ) (both) reading and writing ɓiɓɓe 'am 'e ɓe maaɗa 'e ɓe maako / makko (FJ) my children and yours and his

6.10. The constituent items of appositional and serial nominal groups may be almost any of the kinds of nominal group listed in § 6.2 above. They contain no additional features of relevance to the morphology over and above the features of the individual items of which they are composed, but are included here for completeness.

6.11. The positions which nominals and nominal groups of various kinds can occupy within longer nominal groups are summarized in Appendix 8 3.

Adverbial groups

6.12. An adverbial group may consist of:

  • v. a sole adverbial 4
  • w. an adverbial complex
  • x. a relative adverbial clause
  • y. an appositional adverbial group
  • z. a serial adverbial group

Of these, sole adverbials (such as the adverbial words yaasi ‘outside’, ton ‘there’, nyalawma / nyalorma (FJ) ‘in the daytime’, yeeso ‘in front’) need no further comment here, nor do the appositional and serial adverbial groups (e.g. ton yaasi ‘there outside’, and dow 'e les / dow e ley (FJ) ‘above and below’), since their composition is similar to that of appositional and serial nominal groups (§§ 6.8, 6.9 above).

6.13. Adverbial complexes consist of a preposition (as listed in Appendix 18) + a nominal group of any kind (as listed in§ 6.2), or an adverbial (as listed in Appendix 17), e.g.

preposition + a. nder suudu in the hut
+ b. nder suudu maaɗa in your hut
+ c. nder suudu ndu'u in this hut
+ d. nder ko 'o-wi'i in what he-said
+ e. nder warɓe keenya / nder 'arɓe hanki (FJ) among those-who-came yesterday
+ adverbial bako jango / ado jango (FJ) before tomorrow

Features of nominal groups are therefore repeated within adverbial complexes. In addition, however, an adverbial complex may consist of a preposition + a possessive pronoun-the only place in sentence structure where a possessive pronoun occurs, apart from genitival complexes, e.g. nder maɓɓe ‘among them’, nder maaru / mayru (FJ) ‘in it’, hakkunde mooɗon ‘between you’, etc.

6.14. A relative adverbial clause consists of a clause introduced by a relative adverbial element to, haro / hari (FJ) or 'e ‘where’, nde ‘when’, no ‘how’, which correspond, morphologically and in other ways, to the relative elements which introduce (nominal) relative clauses (cf. § 6.6), e.g. mi-'anndaa to 'o-yahata ‘I-don't-know where he-is-going’, nde 'o-yahi, … ‘when he-went, …’

6.15. The foregoing descriptions of nominal and adverbial groups show the various positions within such groups that may be occupied by the various types of nominal, the morphological features of which are discussed in Part IV. The next chapter gives a systematic account of the structures of various types of sentence, and shows the role of nominal and adverbial groups, and of verbals, within these structures.

Notes.
1. See §§ 272.2-3.
2. A further distinctive feature of genitival complexes is that not more than one syllable is accented (§ 10.3), even where, as in the case of two words, both parts of the complex are individually capable of accentuation. For instance, baaba ‘father’ and bumɗo ‘blind’ are both capable of accentuation, and in a nominal chain (§ 6.5.) baaba bumɗo ‘a blind father’ bumɗo will always be accented and baaba will also be accented in certain circumstances; but in a genitival complex baaba bumɗo ‘the father of a blind man’ the second item bumɗo will not be accented, whether or not baaba is.
3. See also Arnott, ‘Nominal groups in Fula’, Neue afrikanistische Studien, Hamburg, 1966, 40-60.
4. The term ‘adverbial’ is used in preference to ‘adverb’ to emphasize the difference between their behaviour and that of adverbs in some other languages, in particular the ability of many of them (as one type of adverbial group) to function as Subject of a sentence; cf. § 7.3.

7 — Sentence structure

A. — Simple sentences

7.1. A full description of the structure of all types of sentence is unnecessary, and it is sufficient first to describe the basic structure of simple sentences of various types, and the position of nominal and adverbial groups and of verbal forms within them, and then to summarize the structure of other less simple types of sentence.

7.2. Simple sentences are of two main types-verbal and non-verbal sentences-i.e. those which have a verbal form as the nucleus of the sentence, and those which do not. Verbal sentences may be further subdivided into neutral sentences and what will here be called emphatic sentences. The basic structure of sentences of all these types can be described in terms of Subject (S), Predicator (P), Object (O), Complement (C), and Adjunct (A), the letters in brackets being used for the formulaic representation of the sentence structure.

7.3. The possible exponents of the above categories are to some extent the same in all types of sentence, though there is some variation, particularly in non-verbal sentences, and certain restrictions which are of minor importance here.
In general the exponent of Subject, of Object, and of Complement, in any sentence in which they occur, may be a nominal group or an adverbial group, though there are some restrictions on the kind of adverbial group which can function in these capacities. Subject and object elements also can function as Subject and Object respectively in certain non-verbal sentences (cf. § 7.8). For more complicated exponents of these categories, see section B below (§§ 7.18-20).
The exponent of Predicator in verbal sentences is a verbal (word or complex); in non-verbal sentences it is one of a small number of quasi-verbal forms and stabilizing elements (cf. § 7.8 (d), (e)).
The exponent of Adjunct in both verbal and non-verbal sentences is normally an adverbial group, but certain types of nominal group are also possible, and in non-verbal sentences of some types the preterite element can function as Adjunct (cf. § 7.8 (a)).

Fuller details of the structure of the various types of simple sentence and the exponents of the various categories are given in §§ 7 4-9 7.I 8- 22 below.

Verbal sentences

Neutral sentences

7.4. The most common type of neutral sentence has a structure which may be formulated as follows:

(A) (S) P (O) (O) (A)

that is, Predicator, potentially preceded by an Adjunct, or a Subject, or Adjunct followed by Subject, and potentially followed by one or two Objects 1, and by any number of Adjuncts (the point over A in the formula indicates the potentiality at this point of a succession of any number of Adjuncts of various kinds) 2. In this type of sentence the exponent of Predicator is always a verbal form (word or complex). The exponent of Subject and Object is normally a nominal group of some kind, but it may be an adverbial group of certain kinds (e.g. a locative adverbial group, as in the last examples in § 7.5). The exponent of Adjunct is normally an adverbial group of some kind, but certain types of nominal group can function as Adjunct in specific circumstances; for instance a temporal noun or nominal group (e.g. nyannde 'alarba ‘day (of) Wednesday’), or a locative noun or corresponding nominal group (e.g. wuro (maako) ‘(his) compound, home’) after a verb of motion. Simple examples, with one-word nominal and adverbial groups, would be:

P wartu / artu (FJ) come back!
'o-wartii / 'o-artii (FJ) he has come back
'o-waddii-ɗum / 'o addii-ɗun he brought it
AP hannde 'o-wartii / hannde 'o-artii (FJ) today he came back
PA 'o-wartii hannde / 'o-artii hannde (FJ) he came back today
SP Bello wartii / Bello 'artii Bello has come back
ASP hannde Bello wartii / hannde Bello 'artii (FJ) today Bello came back
SPO Bello waddii sheede / Bello 'addii ceede (FJ) Bello has brought money
SPOA Bello waddii sheede hannde / Bello 'addii ceede hannde (FJ) Bello brought money today

In the present analysis, expressions such as 'o-wartii / o-artii (FJ) and 'o-waddii-ɗum are both analysed simply as Predicator (P) without Subject or Object, since 'o- 'he' and -ɗum 'it' are here analysed as elements within the verbal complex, and the whole complex functions as Predicator. (From a syntactical point of view the different treatment of, e.g., Bello wartii / artii (FJ) and 'o-wartii might seem open to question, though even syntactically it has some advantages 3 ; but morphologically there are good reasons for including both subject and object elements within the unity of the complex (cf. § 28.3).

7.5. Other examples, in which each nominal and adverbial group consists of a single word or a two-item complex, are as follows:

SPA Bello wartii wuro / Bello 'artii wuro (FJ) Bello has returned home
SPAA Bello wartii wuro hannde / … 'artii wuro hannde Bello returned home today
SPAAA Bello wartii wuro hannde daga yaadu / … 'artii wuro hannde iwri yaadu (FJ) Bello returned home today from a journey
(A)SPOO (hannde) Bello hokkii Mamman sheede ɗen / … okkii Mamman ceede (FJ) (today) Bello gave Mamman the money (lit. money the)
SPOAA Bello waddii sheede ɗen daga luumo hannde / … waddii ceede ɗen iwdi luumo hannde (FJ) Bello brought the money from the market today

A slightly more involved sentence, with longer pieces functioning as Subject, Object, and Adjuncts, would be:

S P
soobaajo 'am 'o'o, Moodibbo Bello Dukku,
friend my this, Malam Bello (from) Dukku,
waddii / addii (FJ)
brought
O A1
gude ɗiɗi booɗe ɗe'e
cloths two fine these
daga luumo Gommbe / iwde luumo (FJ)
from market (of) Gombe
A2  
nyannde 'alarba saallinde ndeya
day Wednesday past that
 

i.e. this friend of mine, Malam Bello Dukku, brought these two fine cloths from Gombe market the Wednesday before last.

SP janngo ɗon-wara / janngo no arde (FJ) tomorrow is-coming
*(With adverbial group functioning as S or O)
SP non he 'yii / non o y'ii (FJ) so-much is-enough
SP dow wuuwaama / dow fittaama (FJ) on-top has-been-swept
PO mi-wuuwii dow ɗu'um / mi-fitii dow ɗu'un (FJ) I've-swept this upper-part

7.6. A less common type of sentence has a Complement 4 in place of Object in one of the two Object positions, i.e.

  • (A) (S) P (C) (O) (A)
  • or (A) (S) P (O) (C) (A)

The exponent of Complement is usually a nominal group, but it may be an adverbial group of certain kinds. The exponents of the other categories are the same as stated in § 7.4 above, e.g.

(A)SPC (jooni) Bello laatake bumɗo / jooni Bello laatike bumɗo (FJ)
(now) Bello has become blind
SPOC Fulɓe ngaɗii Bello laamɗo / Fulɓe waɗii Bello laamɗo (FJ)
the Fulani made Bello chief
ginnawol lakkit-an-ake Bello mboodi / jinnaaru waylit-an-ii Bello mboddi (FJ)
the djinn turned into a snake for Bello (-an- = for)
SPCO ginnawol lakkitorake mboodi hakkiilo / jinnaaru waylii mboddi hakkille (FJ)
the djinn turned into a snake slowly (lit. with (-or-) care)
PC ɗum-laatoto janngo (FJ idem)
it-will-be tomorrow
*(With adverbial group functioning as C)

Emphatic sentences

7.7. For the present purpose four types of emphatic sentence will be distinguished 5, and will be labelled S-emphatic, O-emphatic, C-emphatic, and A-emphatic (indicating that at the semantic level there is ‘exclusive emphasis’ 6 on the exponent of Subject, Object, Complement, and Adjunct respectively, cf. § 55.6). The structure of S-emphatic sentences is similar to that of the neutral sentences considered above, except that S is not optional, but an essential part of the structure, e.g.

  • S P (O) (O) (A)

In the other types, Object, Complement or Adjunct is in initial position, before the optional Subject, e.g.

  • O (S) P (O) (A) (O-emphatic)
  • C (S) P (O) (A) (C-emphatic)
  • A (S) P (O) (O) (A) (A-emphatic)

As will be seen later, this type of sentence involves a variation in the sentence intonation pattern 7, and also certain restrictions on the tense of the verbal 8. Examples of the four types are:

SPOA Bello waddi sheede hannde / Bello 'addi ceede hannde (FJ) Bello brought money today
OSPA sheede Bello waddi hannde / ceede Bello 'addi hannde Bello brought money today
CP mboodi ngol-lakkitii / mboddi ngol-laati (FJ) it turned into a snake
CSP Waziiri Bello laatii / Bello laatii Waziiri (FJ) Bello became Waziri
ASPO hannde Bello waddi sheede / hannde Bello 'addi ceede (FJ) Bello brought money today

‘Special questions’ are a specialized type of emphatic sentence, in which an interrogative(§§ 23.1 ff.) functions as Subject or as Object, or an interrogative adverbial functions as Adjunct (cf. § 55.5).

Non-verbal sentences

7.8. Non-verbal sentences are of five types:

(a) The first type may be analysed as S C (A), i.e. Subject + Complement with or without Adjunct(s). The exponent of Subject may be a nominal group or an adverbial group of certain kinds (such as an adverbial of time), or a subject element 9. The exponent of Complement may be a nominal group or an adverbial group of certain kinds (such as non ‘thus’, feere ‘different’, daga toye? ‘from where?’). The exponent of Adjunct may be an adverbial group or the preterite element no. (In this type of sentence, which constitutes the ‘identification’ construction, the intonation peak coincides with the salient syllable of the Complement, as indicated by the superscript symbol u in the examples below. The vertical line marks off the Subject from the Complement. The English ‘is’, ‘am’, etc. are placed in brackets in the translations to underline the absence of an overt equivalent in the Fula sentence.)

Ali | shooho / Ali | ko baaso (FJ) Ali (is) poor
Bello | bumɗo / Bello | ko bumɗo (FJ) Bello (is) blind
ɓe | nyamneteeɓe / ɓe | ko nyamminteeɓe (FJ) they (are) (ones-who-get) fed, cf. § 60.23
baaba maako | moodibbo / baaba makko | ko jannoowo (FJ) his father (is) a teacher
duuɓi 'am | sappo jooni / duuɓi 'an | sappo jooni (FJ) I am ten years old now (lit. years my ten now)
ɗu'um | feere / ɗu'un | no seedi (FJ) this (is) different
yaadu maako | meere /yaadu makko | ko meere (FJ) his journey (was) in vain
na'i | jam / na'i ɗin | no e jam (FJ) the cattle (are) well
ɗum | deptere / ɗun | ko deftere (FJ) it (is) a book
mi | bajjo / ko mi | bajjo (FJ) I (am) an only child
hannde | Alarba / hannde | ko Alarba (FJ) today (is) Wednesday
'aan | daga toye ? / 'an | ko honto iwru-ɗaa? (FJ) where do you come from? (lit. you from where?)
nga'al | ba ngala / nga'al | ko nii this (is) like that
'o | Dii'o no / Dii'oojo nun (FJ) he (was) formerly a District Officer (D.O.)

A sub-division of this type is the one-word sentence non ‘it is so’, consisting of Complement only without Subject, and the corresponding negative sentence naa non 'it is not so' (cf. the common inquiry koo naa non? ‘or not so?’, i.e. ‘isn't that so?’).

(b) The second type of non-verbal sentence may be analysed as S P (A), i.e. Subject + Predicator with or without one or more Adjuncts 10. The Subject is 11 a nominal group or subject element, the Predicator ts one of the stabilizing elements ɗon 12 ‘exists, is present’, ton ‘is present yonder’, with or without a preterite element (which in this type of sentence forms a complex with the stabilizing element as in the case of verbal complexes). The Adjuncts are adverbial groups, and since this is the locative construction, a locative adverbial group is very common, e.g.

deptere ɗon / deftere ɗon (FJ) there is a book
Bello ɗon / Bello no ɗoo (FJ) Bello is present, is here
mi ɗon-no / miɗo ɗon nun (FJ) I was present
'o ton hannde / himo ton hannde (FJ) he is there today
Ali ɗon (-no) ɗo'o / Ali hari no ɗon (FJ) Ali is (was) here
Bello ton ton / Bello no ton (FJ) Bello is over there
'o ɗon nder gelle jooni / himo nder gelle jooni (FJ) he is in the town now

(c) The third type may be analysed as A S (A), i.e. Adjunct + Subject with or without further Adjuncts. The initial Adjunct is an interrogative adverbial (no ‘how? how many?’, to which here means ‘how?’), the Subject is a nominal group—very often a genitival complex—and any further Adjuncts are adverbial groups 13, e.g.

no ɓanndu? how (is your) body? i.e. how do you feel?
no sheede ngaari? / coggu ngaari? (FJ) how-much (is) the-price-of a-bull?
no 'innde maa? / no wi'ete-ɗaa (FJ) what (is) your name?
no duuɓi maɓɓe? / duuɓi maɓɓe (FJ) what (are) their ages? (lit. how years their?)
to kuuɗe? / no golle ɗen wa'i (FJ) how (are) works? ‘how's works?’ ‘how's things?’

(d) The fourth type may be analysed as P (O) (A), i.e. Predicator with or without Object and with or without Adjunct or Adjuncts. The Predicator is one of the quasi-verbal 14 interjections ndaa / hino (FJ) ‘here is, there is, this is, that is’ (used in pointing something out) and hiin (or hii) ‘here you are, here is’ (used in proffering something), each of which can be used alone; the Object is a nominal group or object element the Adjuncts are adverbial groups, e.g.

ndaa / hino (FJ) here it (etc.) is!
ndaa baaba maako / hino baaba makko (FJ) there is his father
ndaa mo (ton) / hino mo (ton), hino mo too (FJ) here he is, there he is (over there)
hiin / e hin (FJ) here you are
hiin huunde maaɗa here is your thing
hiin nde / hino nde (FJ) here it (e.g. book) is

(e) The structure of the fifth type resembles that of the fourth type, but the Object is always present. It has the structure P O (A), i.e. Predicator + Object with or without an Adjunct or Adjuncts. Here the Predicator is the quasi-verbal form ndikka (or 'igga) ‘rather, better’ 15. The Object is a nominal or adverbial group (not an object element, as in the fourth type), Adjuncts are adverbial groups, particularly one consisting of dow ‘over, (rather than)’ + a nominal or adverbial group, e.g.

ndikka ɗu'um / ndikka ɗun (FJ) this is better (lit. better this)
ndikka ɗu'um dow ɗuma / ndikka ɗun edii ɗuma (FJ) rather this than that

7.9. Apart from the absence of a verbal form in these types of sentence, the most significant point about them for the present study is the occurrence of subject and object elements as Subject and Object, respectively, whereas where they occur in verbal sentences they function as part of the verbal complex.

Sentences containing a ‘prelude’

7.10. The basic structure of simple sentences of various types is as described above. A very common type of sentence, however, is one in which the main part of the sentence is preceded by what is here called a ‘prelude’—a preliminary word or phrase marked off in some way from the remainder. Any item occurring in the prelude may be said to be ‘pre-posed’.

7.11. This type of sentence has received little or no attention in existing grammars and the account given here is a tentative one which will, no doubt, require amendment in the light of further study.

7.12. In some cases the arrangement is the same as that given above in describing the basic structure of sentences, but the prelude is marked (i) by the addition, after an initial Subject or Adjunct, of an emphasizing particle such as kam, boo, or (ii) by the intonational feature here called ‘suspension’ (§ 10.7. ) and indicated by in the examples, or by both particle and suspension. (In the ensuing examples the prelude is marked off from the main part of the sentence by a vertical line; the symbol ⎸⎸ indicates the apex of the intonation pattern-cf. § 10.2.)

Neutral sentences

S P O A Bello kam | waddii sheede hannde / addii mbuuɗi hannde (FJ)
as for Bello, he brought money today
suka 'o'o˜ | waddi sheede hannde / addii mbuuɗi hannde (FJ)
as for this youth, he brought money today
S P hanko kam | 'o-wartii / kanko kam | 'o artii (FJ)
as for him, he has come back
miin kam | mi-anndaa / min kan | mi-anndaa (FJ)
as for me, I don't know
A S P O hannde kam | Bello waddii sheede / hannde kan | Bello addii mbuuɗi (FJ)
hannde kam | Bello waddii sheede / hannde kan | Bello 'addii mbuuɗi (FJ)
today, Bello did bring money

Emphatic sentences

S O P Bello kam | sheede waddii / Bello kan | mbuuɗi addii (FJ)
as for Bello, he brought money today
hanko kam | sheede o- waddi /hanko kam | mbuuɗi o-addii (FJ)
as for him, he brought money
A S O P hannde kam | sheede Bello waddi / hande kan | 'mbuuɗi Bello addi (FJ)
today, Bello brought money
SC Bello kam | bumɗo Bello / Bello kan | ko bumɗo (FJ) Bello is blind
A SC jooni kam | Bello bumɗo / jooni kan | Bello ko bumɗo (FJ) Now, Bello is blind

Non-verbal sentences

SC Bello kam | bumɗo Bello / Bello kan | ko bumɗo (FJ) Bello is blind
A SC jooni kam | Bello bumɗo / jooni kan | Bello ko bumɗo (FJ) Now, Bello is blind

7.13. In other cases there is a different order as compared with a corresponding sentence with basic structure, some item being preposed instead of occupying its normal position in the sentence. There may also be an emphasizing particle, or suspension, or both, but the abnormal order is itself, sufficient to mark the prelude, e.g.

Neutral sentences

(basic) A S P O hannde | Bello waddii sheede / hannde | Bello addii mbuuɗi (FJ)
today Bello brought money
S A P O Bello (kam) | hannde waddii sheede
as for Bello, today he brought money
suka 'o'o- | hannde waddii sheede
as for this youth, today he brought money
suka 'o'o kam | hannde waddii sheede
the same
O A S P sheede (kam) | hannde Bello waddii
sheede (kam) | hannde Bello waddii-ɗe
as for money, today Bello brought it/some

Emphatic sentence

S A O PBello kam˜ | hannde sheede waddi
--as for Bello, today he brought money
O A S Psheede ɗen kam˜ | hannde Bello waddi-ɗe
--as for the money, it was today Bello brought it

7.14. In certain circumstances a pre-posed item is recapitulated by a pronominal form in the appropriate place in the main part of the sentence (the appropriate place for subject and object elements normally being within the verbal complex which functions as Predicator), e.g. 'o-, mi- in some examples in § 7.12 above, -ɗe in some examples in § 713, muuɗum and maako in § 7.15 below. See also § 26.6 {end).

7.15. In most cases the pre-posed item corresponds to the whole of the exponent of a given category in the corresponding basic sentenceSubject, Object, Adjunct, etc., as in the examples in §§ 7.12-13 above. Sometimes, however, the pre-posed item corresponds to a part only of the exponent of the category in question, particularly to the second item in a genitival or adverbial complex. In such cases the pre-posed item is recapitulated by the appropriate possessive pronoun, e.g.

Neutral sentences

Bello kam˜ | na'i muuɗum fuu mbaatii / maayii (FJ)
as for Bello, all his cattle have died
(corresponding to na'i Bello fuu mbaatii
all Bello's cattle have died)
'o'o boo˜ / oo ɗoo (FJ) | mi-'anndaa 'innde muuɗum
and this one, I don't know his name
hanko kam˜ | mi- 'anndaa 'innde maako / makko (FJ)
as for him, I don't know his name
ndu'u kam˜ | mi-waalataa nder maaru / mayru (FJ)
as for this, I won't spend the night in it
(corresponding to mi-waalataa nder ndu'u
I won't spend the night in this one

Non-verbal sentences

debbo gooto˜ | 'innde muuɗum Yaawuro
one woman, her name (was) Yawuro
hanko kam˜ | 'innde maako Bello
as for him, his name (is) Bello

7.16. In some cases the items in the prelude cannot be related to the structure of the main part of the sentence, as in the above cases, but must be regarded as in grammatical hiatus with the rest of the sentence, e.g.

  • 'aan ni, | saa 'a tetetee as for you, it's luck that's wanted

7.17. The occurrence of items in the prelude is of particular significance in relation to the grammatical behaviour of independent and possessive pronouns, see §§ 24.15- 17, 26.6.

B. Simple sentences containing one or more clauses

7.18. Whereas the simple sentences discussed so far may be considered as consisting of one part only, the remaining types of sentence consist of more than one part, some or all of which have a structure identical with or similar to that of simple sentences. Each such sentence like part of a sentence may be referred to as a clause.

7.19. Clauses with a structure identical with that of simple sentences are illustrated in § 7.20. In other types of clause there is an extra element or particle not found in simple sentences, e.g. a relative or relative adverbial element in relative or relative adverbial clauses 16, a clause-initial conjunctive particle in some other types of clause, also illustrated below.

7.20. A very common type of sentence is structurally a simple sentence, but the exponent of one or more of the categories Subject, Object, Complement, and Adjunct is a clause, as described above.

(B. i) Clause as Subject

A relative clause, which has already been described as one kind of nominal group (§§ 6.2, 6.6) often occurs as Subject, e.g.

S P
nde shoodu-mi keenya nden | majjii one-which I-brought yesterday the | has-got-lost

Otherwise subject clauses are rare except with a verb of causing or preventing 17, and the radical haan- ‘be appropriate’(§ 54.19), e.g.

  • 'o-somu waɗi 'o-jooɗii — he-was-tired caused he-sat-down

(B. ii) Clause as Object

The type of subordinate clause, and the tense of the verbal in this clause, is correlated to some extent with the radical. For instance with some verbs (e.g. verbs of commanding, wishing, requesting, permitting) 17 the verbal in the object clause is in the Subjunctive, e.g.

'o-wi 'ii mi-wara / mi 'ara (FJ)he-said I-should-come 'o-torake-yam mi-hokka-mo nyaamdu seɗɗa / 'o-torii-lam yo mi-okkor-mo-nyaametee seeɗA (FJ)he-begged-me (that) !-give-him food a-little, i.e. he begged me to give him a little food mi-yerdaaki Bello jooɗoo ɗo'o / mi jaɓaa yo Bello jooɗoo ɗo'o (FJ)I-don't-agree (that) Bello should-sit here

Others (e.g. verbs of relating, knowing, believing, perceiving, showing) 17 may be followed by a relative clause, a relative adverbial clause, or a simple clause (i.e. clause without introductory element or particle) without restriction of tense, or in some circumstances a clause introduced by koo, e.g.

'o-wi'ii-yam ko Bello waɗata / 'o-haalanin-mi ko Bello waɗata (FJ)he-told-me what Bello is-doing 'o-wi-'ii-yam to Bello yahi,where Bello went 'o-wi'ii-yam Bello wartii / 'artii (FJ)Bello has-returned mi-'anndi/mi-nanii / mi-yi'ii ko Bello waɗataI-know/I-hear/I-see what Bello is-doing mi-'anndi/mi-nanii/mi-yi'ii Bello wartii / 'artii (FJ), etc.I-know/1-hear/I-see (that) Bello has-returned, etc. mi-'anndaa koo Bello wartii / mi-'anndaa si Bello 'artiiI-don't-know whether Bello has-come-back

(B. iii) Clause as Adjunct. This may be

(a) a relative adverbial clause, introduced by one of the relative adverbial elements nde ‘when’, to, haro ‘where‘, no ‘how’, bano ‘as’, e.g.

mi-suuɗii-ɗum haro ɗum-yi'ataako / mi-suuɗii-ɗum ka ɗum-yi'ataako (FJ)
I've hidden it where it can't be seen

nde 'o-yi'i-mo, 'o-noddi-mo when he saw him, he called him

(b) a temporal clause introduced by the conjunctive particles bako, ɗooko 'before', ko 'since', to 'when', ɓaawo 'after', sey 'until', e.g.

mi-'ummoto bako Bello wara / mi-immoto ado Bello arde (FJ)
I'll start out before Bello comes

(c) a conditional clause introduced by to 'if', koo 'even if', dow 'on condition that', e.g.

to 'o-warii, mi-wi 'ay-'on / si o 'arii, mi haalanay-on (FJ)
if he comes, I'll tell you

(d) a causal clause introduced by :ngam 'because', nde 'since, seeing that', e.g.

nde mi mawɗo, sey ɗalaa-yam / ɓaa ko mi mawɗo, accan (FJ)
since I (am) an important person, leave me

'a-yi'ataa-mo, ngam 'oɗon-suuɗii / a yi'ataa-mo, ɓaa himo suuɗii (FJ)
you won't see him, because he's in hiding

(e) a purpose clause introduced by ngam 'in order that, so that', or a simple clause (without introductory particle) with a verbal in the Subjunctive, e.g.

mi-yaaru-no-ɓe (ngam) ɓe-pemmboo / mi naɓu-no-ɓe fii yo ɓe fembo (FJ)
I had taken them so that they might be shaved

(f) a modal clause, with the verbal in the Stative or Continuous tense, e.g.

'o-wartii 'emo-hiimoo / o-artii himo hiimoo (FJ)
he came back (he was) pondering

Clauses functioning as Adjunct can—like other items functioning as Adjunct—either follow or precede the Predicator. For most types of clause the normal position is after the Predicator (and Object, if any); but conditional clauses and Adjunct clauses introduced by nde 'when', to 'when', most often precede the Subject (if any) and the Predicator, as in the examples in (a) and (c) above.

(B. iv) Clause as Complement

There is a less common type of sentence which may be analysed as having the same basic structure as a non-verbal sentence of type (a) (§ 7.8(a)), but with a clause functioning as Complement cf. § 54.20, e.g.

doole mi-dilla / bee mi-yaha (FJ) I must go
doole maɓɓe ɓe-koota / bee ɓe hoota (FJ) they must go back

Part III. — Outline of Phonological Features

  1. The consonant system
  2. The vowel system
  3. Sentence intonation patterns
  4. Final glottality

Part IV. — The Nominal System

  1. The class system-general outline
  2. Meanings associated with the various classes — some generalizations

Full nominals

  1. Full nominals — general
  2. Nouns and adjectives-range of classes
  3. The singular/plural and other relationships
  4. Suffix grades
  5. Initial-consonant alternation
  6. Other features of nominal stems
  7. Stems combining with suffixes of different grades
  8. Suffix variations

Other nominals

  1. Nominals other than full nominals — general
  2. Specifiers
  3. Noun substitutes
  4. Nominal elements
  5. The ɗum series of pronominal forms
  6. Numerals

Part V. — The verbal system

  1. Introduction
  2. The tense system — general
  3. The categories of subject and object elements
  4. The categories of verbal radicals

Minimal complexes

  1. The arrangement of the components
  2. The subject elements
  3. The tense suffixes of minimal complexes
  4. The radical

Enlarged complexes

  1. Enlarged complexes — general
  2. 1-object and 2-object complexes
  3. Preterite complexes
  4. The tense suffixes of enlarged complexes

Other features of verbal complexes

  1. Intonation pattern
  2. Final glottality

Further grammatical features of the verbal system

  1. Summary of the characteristics of individual tenses
  2. Anomalous verbal forms
  3. Imperative forms
  4. Voice

The uses of the tenses

  1. The factors involved
  2. The General Past
  3. The Emphatic Past
  4. The General Future
  5. The Vague Future
  6. The Desiderative
  7. The Stative and Continuous tenses
  8. The Negative tenses
  9. The Subjunctive
  10. The Relative tenses
  11. Tense sequences in serial sentences

Radical extensions

  1. Radical extensions-general
  2. The individual extensions
  3. Extensions in combination

Part VI. — Verbo-nominals

  1. Verbo-nominals — morphology and syntax

Appendices
Bibliography

Review

Le professeur Arnott présente en fait sous ce titre une grammaire peule complète qui, outre les amples développements exigés par le traitement des systèmes nominal et verbal consacre aussi nombre de pages aux autres points de la morpho-syntaxe ainsi qu'à la phonologie de la langue étudiée.
Parmi les nombreux “dialectes” peuls, l'auteur a choisi celui de Gombe (Bauchi Province, Northern Nigeria) sur lequel il travaille depuis de nombreuses années. Ce choix est d'ailleurs heureux en ce que ce “dialecte”, très représentatif de l'ensemble des formes du peul oriental, demeure grammaticalement assez proche aussi des “dialectes” occidentaux.
Après une brève introduction générale, le lecteur aborde une introduction grammaticale qui constitue à la fois une sorte de répertoire terminologique, un premier aperçu sur les sytèms nominal et verbal et un développement sur les structures syntaxiques, allant des séquences nominales à l'analyse de l'énoncé complexe. Viennent ensuite la description du système phonologique et des faits morpho-phonologiques puis une étude très complète des nominaux, de leurs substituts et de leurs déterminants déictiques.
Le long traitement du verbe qui la suit reprend et complète les travaux publiés antérieurement par D.W. Arnott, et notamment son article en 1966 dans Afrika und Übersee. Le chapitre suivant sur les extensions radicales aurait pu être étoffé et surtout aurait gagné à occuper une autre place, compte tenu de l''importance de ce phénomène dans la langue.Il précède celui sur les nomino-verbaux (infinitifs et participes) qui clôt le corps de l'ouvrage, lequel est toutefois suivi de 18 appendices, d'une bibliographie et de plusieurs tableaux.
Il ne saurait être question ici de formuler les remarques et les critiques d'ordre linguistique que peut appeler dans ses détails une étude qui par sa richesse et sa profondeur surpasse nettement les grammaires peules antérieurement publiées et en font un ouvrage indispensable pour quiconque s'intéresse à cette langue.
Cependant nous noterons que la rançon peut-être inévitable des qualités nous a semblé être une certaine difficulté de lecture que ressentira encore plus un lecteur profane. Certes D.W. Arnott n'utilise pas un des métalangages  — pour ne pas dire jargons — chers à certains auteurs, mais le très large usage qu'il fait des abbréviations et conventions chiffrées paraît en bien des cas compliquer inutilement la lecture de son oeuvre.

P.F. Lacroix

Obituary

D.W. Arnott (1915-2004)

D.W. Arnott was a distinguished scholar and teacher of West African languages, principally Fulani (also known as Fula, Fulfulde and Pulaar) and Tiv

David Whitehorn Arnott, Africanist: born London 23 June 1915; Lecturer, then Reader, Africa Department, School of Oriental and African Studies 1951-66, Professor of West African Languages 1966-77 (Emeritus); married 1942 Kathleen Coulson (two daughters); died Bedale, North Yorkshire 10 March 2004.

D.W. Arnott was a distinguished scholar and teacher of West African languages, principally Fulani (also known as Fula, Fulfulde and Pulaar) and Tiv. He was one of the last members of a generation of internationally renowned British Africanists/linguists whose early and formative experience of Africa, with its immense and complex variety of peoples and languages, derived from the late colonial era.

Born in London in 1915, the elder son of a Scottish father, Robert, and mother, Nora, David Whitehorn Arnott was educated at Sheringham House School and St Paul's School in London, before going on to Pembroke College, Cambridge, where he read Classics and won a “half-blue” for water polo. He received his PhD from London University in 1961, writing his dissertation on “The Tense System in Gombe Fula”.

Following graduation in 1939 Arnott joined the Colonial Administrative Service as a district officer in northern Nigeria, where he was posted to Bauchi, Benue and Zaria Provinces, often touring rural areas on a horse or by push bike. His (classical) language background helped him to learn some of the major languages in the area - Fulani, Tiv, and Hausa - and the first two in particular were to become his languages of published scientific investigation.

It was on board ship in a wartime convoy to Cape Town that Arnott met his wife-to-be, Kathleen Coulson, who was at the time a Methodist missionary in Ibadan, Nigeria. They married in Ibadan in 1942, and Kathleen became his constant companion on most of his subsequent postings in Benue and Zaria provinces, together with their two small daughters, Margaret and Rosemary.

From 1951 to 1977, David Arnott was a member of the Africa Department at the School of Oriental and African Studies (Soas), London University, as Lecturer, then Reader, and was appointed Professor of West African Languages in 1966. He spent 1955-56 on research leave in West Africa, conducting a detailed linguistic survey of the many diverse dialects of Fulani, travelling from Nigeria across the southern Saharan edges of Niger, Dahomey (now Benin), Upper Volta, French Sudan (Burkina Faso and Mali), and eventually to Senegal, Gambia, and Guinea. Many of his research notes from this period are deposited in the Soas library (along with other notes, documents and teaching materials relating mainly to Tiv and Hausa poetry and songs).

He was Visiting Professor at University College, Ibadan (1961) and the University of California, Los Angeles (1963), and attended various African language and Unesco congresses in Africa, Europe, and the United States. Between 1970 and 1972 he made a number of visits to Kano, Nigeria, to teach at Abdullahi Bayero College (now Bayero University, Kano), where he also supervised (as Acting Director) the setting up of the Centre for the Study of Nigerian Languages, and I remember a mutual colleague once expressing genuine astonishment that “David never seemed to have made any real enemies”. This was a measure of his integrity, patience and even-handed professionalism, and the high regard in which he was held.

Arnott established his international reputation with his research on Fula(ni), a widely used language of the massive Niger-Congo family which is spoken (as a first language) by an estimated eight million people scattered throughout much of West and Central Africa, from Mauritania and Senegal to Niger, Nigeria, Cameroon, Central African Republic and Chad (as well as the Sudan), many of them nomadic cattle herders.

Between 1956 and 1998 he produced almost 30 (mainly linguistic) publications on Fulani and in 1970 published his magnum opus, The Nominal and Verbal Systems of Fula (an expansion of his PhD dissertation), supplementing earlier works by his predecessors, the leading British and German scholars F.W. Taylor and August Klingenheben. In this major study of the Gombe (north-east Nigeria) dialect, he described, in clear and succinct terms, the complex system of 20 or more so-called “noun classes” (a classificatory system widespread throughout the Niger-Congo family which marks singular/plural pairs, often distinguishing humans, animals, plants, mass nouns and liquids). The book also advanced our understanding of the (verbal) tense-aspect and conjugational system of Fulani. His published research encompassed, too, Fulani literature and music.

In addition to Fulani, Arnott also worked on Tiv, another Niger-Congo language mainly spoken in east/central Nigeria, and from the late 1950s onwards he wrote more than 10 articles, including several innovative treatments of Tiv tone and verbal conjugations, in addition to a paper comparing the noun-class systems of Fulani and Tiv (“Some Reflections on the Content of Individual Classes in Fula and Tiv”, La Classification Nominale dans les Langues Négro-Africaines, 1967). Some of his carefully transcribed Tiv data and insightful analyses were subsequently used by theoretical linguists following the generative (“autosegmental”) approach to sound systems. (His colleague at Soas the renowned Africanist R.C. Abraham had already published grammars and a dictionary of Tiv in the 1930s and 1940s.)

In addition to Fulani and Tiv, Arnott taught undergraduate Hausa-language classes at Soas for many years, together with F.W. (“Freddie”) Parsons, the pre-eminent Hausa scholar of his era, and Jack Carnochan and Courtenay Gidley. He also pioneered the academic study of Hausa poetry at Soas, publishing several articles on the subject, and encouraged the establishment of an academic pathway in African oral literature.

The early 1960s were a time when the available language-teaching materials were relatively sparse (we had basically to make do with cyclostyled handouts), but he overcame these resource problems by organising class lessons with great care and attention, displaying a welcome ability to synthesise and explain language facts and patterns in a simple and coherent manner. He supervised a number of PhD dissertations on West African languages (and literature), including the first linguistic study of the Hausa language written by a native Hausa speaker, M.K.M. Galadanci (1969). He was genuinely liked and admired by his students.

David Arnott was a quiet man of deep faith who was devoted to his family. Following his retirement he and Kathleen moved to Moffat in Dumfriesshire (his father had been born in the county). In 1992 they moved again, to Bedale in North Yorkshire (where he joined the local church and golf club), in order to be nearer to their two daughters, and grandchildren.

Philip J. Jaggar