‘ “WW I IHIHIHI! 1 llJllHlH THE. EDEQPHQNE £24 HAUSA (0—3 _x_s '—I :1: moooo “L“fwests ‘3» ”'10 Degree of M» A. Ei’EECfilGI-W STATE UN EVERSiTY Mary Jo Moore 1988 mm H “Wu '1 \ UJEL 10714 mi JUL 0 7 $828 1 1??? IiJUL 3»: PO 055‘ 320: Emmga éa. Human . THE IDEDP HONE 1N HAUSA By Mary Jo Moore A THESIS Submitted to Michigan State University in partial mlfillment of the requiremmts for the degree of MASTER OF ARTS Department of Linguistics and Oriental and African Languages 1968 ACKNOWLEDGEEQ TS I am indebted to the following individuals for their assistance in the preparation of this thesis: «Joseph H. Greenberg. whose course in “African Languages and Linguistics" at the 1967 Linguistics Instimte inspired me to unden- take an [analysis of Hausa ideophones.‘ «William J. Samaria. who. during an interview at the ISA meetings in Ann moz- last sunnner. encouraged me to pursue the present tepic and also gave me some insights into the problems involved in describing ideophones in African languages. «Benjamin G. Ishaku. who acted as the primary consultant for this project and whose excellent course in Advanced Hausa at Michigan State University last year helped to prepare me for such a stucb'. «my thesis committee. Meyer L. Wolf. 31th M. Brand. and Charles H. Kraft. for their useful suggestions. knowledgeable criticism. and encouragement during the course of my work. «Ibrahim Wada and Steven Lucas. for their assistance as native speakers of Hausa. «those fellow graduate students who read an earlier draft of this thesis and offered their connnents and suggestions. in particular. Ernest 1“. mm and Earl Herrick. ii TABLE OF CDNTEN TS IN T170111 CTION PHONOIDGY The Phonanes Syllable Distribution and Phonological Shape Vowel metribution and Tone Consonant Distribution Intonation and Juncture MORPHOPHONEMICS Vowel Alternation Consonant Alternation Altemation Between Phonological Shapes Rednplication Recurring Partials Metathesis SYNTAX Ideophones in Obligatory Positions Ideophones in Optional Positions Grammatical Functions I The Enphasis Transformation PROBLEMS IN DEEEIMINING MEANING The Problan of Context The Problem of Semantic Categorization The Problem of Determining Relationships Between Morphane Structure and Meaning iii \OOO-P-F’l-J 12 15 15 17 22 APPENDICES I. The Afroasiatic and Niger-Kordofanian Languages Families II. Phonological Types of IdeOphones III. Sound/Meaning Gmups IV. Examples of Ideophones in Context V. Alphabetical Listing of Ideophones in the Selected mta BIBLIOGRAPHY iv INTHDIII CI'ION The phenomenon of sound symbolisn and non-arbitrary morphane structure is widespread in the languages of the world. although often restricted to a minor role within a particular language. For example. in English there are recurrent pattems within some words which could be assigned some kind of semantic value apart from the meaning of the whole word. as in lump. hump. bump. clump. dump. and splash. spatter. splatter. and splat. That is. one could conceivably assign morpheme stains to m. and 30;: or _s_o;-_. But in English as in many other languages. this phenomenon. though interesting. is not prominent. In the languages of Africa. the phenomenon is indeed prominent. and is particularly characteristic of a class of words nomally referred to. in the literature as ideoPhones.l This thesis will attempt a description of ideophones as they occur in the Hausa language of West Africa. Hausa is the first language of approximately 15 million people of Northem Nigeria and the second or trading language of perhaps another 10-15 million. It is classified by Greenberg as a member of the Chadic branch of the Afroasiatic family.2 1"Ideophones by count constitute. next to nouns and verbs. a major part of the total lexicon of African languages...." William J. Samarin. "Perspectives on African Ideophones.“ African Studies. XXIV (1965). 121.. 2Joseph Greenberg. The Languages of Africa (Bloomington. 1966). 1 impre /' ‘ . ; x r 4 f \4 {on 3" I. mi K I / ,. r ,. j N w 2 Hausa seems to participate in a Sprachbund uniting - it to African languages to which it is not related genetically. The ideo- phone pWs an important role in this Sprachbund in that it is a feature of Hausa miniscent of the Niger-Congo languages of sub- Saharan Africa. Hausa is in contact with some of the Niger-Congo languages. for example. Fulani. Yoruba. Nupe and Tiv. The precise family relationships of Hausa. its contact languages within the Niger-Congo group. and the other languages cited in this paper. are diagrammed in Appendix I. ! Although ideOphones in African languages are often classified with adverbs. they do not always function as such. mile it is true that ideophones often function in Hausa in a manner that Mt be labeled I'adeerbial." they also function in other ways} Consequently. (it seems best to define them primarily as a lexical category and only “fit: . 'a 42‘ o" the ideophone may be taken to be an expressive unit of phonaesthetic speech which is most commonly used for intensification or big-powered description. and which is highly dependent on context for semantic value. The corpus for this study includes examples of ideophones from texts (stories and some proverbs). from tapes (extanporaneous render- ings of stories and life histories). from native speakers here at Michigan State University. and from Abraham' 3 dictionary.2 —IIt is only as a point of reference that the tem “adverbial" is employed here. It is important to note this inaanuch as Hausa exhibits no mozphological category which one would want to label “adverb. " Abraham. motio- of the Hausa Len London 1 62 Mama? the dictiok . we'mu . --' .. o: -- * d( It seeing. )mm native speaker reaction. that many ideOphones in the dictionary are either archaic or are dialectal or idiolectal variants. f secondarily as a grammatical category. As a working definition. then. , a f I 1 v‘ I 3 The procedure for collecting the data was as follows: All of the ideophones in the texts and tapes were listed with as much context as seemed necessary. These were then checked with Mr. Ibrahim Wade. a native speaker of Kano Hausa. for accuracy in translation and pronun- ciation. This yielded approximately 100 items. Other emples. perhaps about fifty. were collected from Mr. Wade. from Mr. Benjamin Ishaku. a native speaker of Zaria Hausa. and from Mr. Steve: Lucas. a nears-native speaker of Niger Hausa. But the difficulty of obtaining a large mugh corpus by this means made it necessary to resort to the dictionary for further examples. In this way. it was possible to collect from the dictionary some one thousand emnples of ideophones. and so obtain a franc of reference for subsequent work with Mr. Wade and Mr. Ishaku. The majority of these examples were discarded for reasons mentioned in a previous footnote. reducing the corpus to l#05 ideophones. with from one to ten contexts for each. This mailer corpus is henceforth referred to as the "selected data." PHONOLOGY Although the definitive description of Hausa phonolog has yet to be written. the subject has been treated by a number of linguists and useful reference grammars are available. However. there is justi- fication for giving special treatment to the phonologr of the Hausa ideophone. inaanuch as various discussions of the phenomenon of ideo- phonic speech in African languages have made statements to the effect that ideophones manifest a subsystem (or by-system) within the phonology of the language under analysis.1 The Phonemes The consonant phonemes are charted as they occur in the ideo- phones in the selected data. yielding a total of thirty phonemes. This differs from Kraft's description2 and also from Greenberg's 1"The special features of ideophones in Shona lead us to recogiize in this language a subsystem of syntactical. morphological. and phono- logical units and structures peculiar to ideophones." George Fortune. Ideophones in Shona (London. 1962). p. 3. "Here is...a class of morphemes where is revealed a phonemic 'by-system' in the language.“ Samarin. pp. cit.. p. 119. in reference to Nguni ideophones. "The existence of an important phonological subsystem covering phon- aesthetic adverbs. or ideOphones...“ F. W. Parsons. "Some Observations on the Contact Between Hausa and Ehglish.’I mposium on Mtflm (Brazzavile. 1962). p. 201. in reference to Hausa. "Their phonologr forms a sub-system in relation to the phonolog of other parts of speech..." Leslie H. Stennes. A Reference grammar of Adamawa Egg; (East Lansing. 1967). p. 17. in reference to Fulani ideophones. zxraft describes twenty-five consonant phonemes for Hausa. excluding the labialized and the palatelized series of velar stOps. and including /"y/ and two 313. a trill and a flap. Charles H. Kraft. A Sing of z. 5 description1 of the consonant phonemes of Hausa. but the discrepancies are due more to different analytical approaches than to differences between the phoneme inventory of the whole language and the phoneme inventory of the ideophones alone. For emple. the decision to include a labialized and palatalized series facilitates the analysis ' of syllable types. since labialized and palatalized consonants are the only contoid clusters that occur word initially in Hausa. Treating them as single consonants yields two major syllable types (CV. CVC) instead of three (if CCVC were included). The vowel phonemes are charted separately for long and short vowels. yielding a ten-vowel system in accord with both Kraftz and Greenberg. Although the mu spectrum of five long vowels and five short vowels occurs in ideophones. long vowels have a very low fre- quency and possibly a predictable pattern of distribution.3 This is in contrast with a more evenly balanced frequency of long and short vowels in the rest of the language and also an unpredictable pattern of distribution for vowel length. The present analysis of tone in Hausa ideophones is canpatible with the analyses of Kraft and Grealberg for the whole language. Hausa figtax (Hartford. 1963). p. 19. fi/ and /I'/ do not contrast in the data. and /"y/ doesn't occur at all. lGreenberg describes thirty-one consonant phonanes for Hausa. includ- ing the labialized and the palatalized series of velar stops. and both /‘i"/ and /i'~/. He also includes /py/ and excludes /d/. Joseph Greenberg. "Some Problems in Hausa Phonolog.“ Lanaage. XVII (1941). 316-323. zAlthough Kraft analyzes an ei t-vowel system in the above-mentioned reference (excluding /e/ and /o/ . he describes a ten-vowel sistem later. 1 . in 3 Introduction to §poken Hausa (East Lansing. 1965). 1). 3It is interesting to note by way of comparison that vowel length. though phonemic for Fulani in general. is not distinctive for Fulani ide0phones. See Stennes. Op, cit.. p. 19. Chart 1. The Phonemes of Hang Mhones Fronted Alveo- Bilabial Alveolar Alveolar Palatal Velar Glottal STOPS checked 6 :1 ts k ’ flat Rw sharp1 Ry voiced b d g flat 8" sharp a voiceless p t k flat kw sharp klr FRICATIVES voiceless f 8 sh h voiced z AFFIECATEB voiceless c voiced j RESONANTS nasal m n liquid 1 r semi-vowel w Y front central back front central back high 1 u high ii uu mid e 0 mid ee 00 low a low a tones: high (unmarked) low C) falling (If) 1The terms g3 and M mean @ializgd and palataliged respectively. They are preferred here for convenience. and also because they will be used later to isolate a natural class. The tam checked should be inter- preted as meaning gl_ottalize . and is preferred here for these same reasons. These terms are used with the understanding that they can be labels for articulatory features as well as acoustic features. See Morris Halls. "On the Bases of Phonolog“ in Fodor and Hats. E9. Structure of Lanfigg (Diglewood Cliffs. 1961+). p. 326. 6 7 The "transcription“ anployed in Chart 1 and in this thesis is the standard Hausa orthograpmr but with tone and vowel length indicated. High tone is unmarked and low tone is marked with a grave accent. The falling tone is marked with a combination of acute and grave accents. Length is represented by doubling the vowels. One deviation fmm the standard orthography is /'/ instead of /'_/ for glottal stop. The diph- thongs a and a; are analyzed as /aw/ and /ay/ respectively. since the pressure of admissible syllable types suggests treating the final vocoid as a consonant. However. the standard orthograqahical represen- tation is preferred for the written Hales in this paper. Several additional explanations are in order for the consonant phonemes in Chart 1. /f/ and /p/ are normally analyzed as allophones of one phonane. Kraft describes than as freely alternating variantsl and Greenberg as dialectal variants.2 The decision here is to treat them as separate phonemes which. like mam other consonant phonemes in ideophones. may altemate freely. There is a near-minimal pair in the language3 and possibly also in the ideophones. But regardless of this. it seems to be the case that certain words alms are pronounced with /p/ and others always with /f/ and certainly it is common for than both to occur within one idiolect. Although /g/ occurs in only one ideophone in the larger (pre- selected) corpus. it has been included in the chart of consonant phonemes 1Emit. 5 Stag: of Hans; mtax. p. 20. 2"We m divide Kano weakers into f-speakers and p-speakersJ' Greenberg. uSome Problems in Hausa Phonoloar.“ p. 322. 3/paa/ '.ledge'. /fa/ 'indeed' 8 because it completes the labialization and palatalization of velar 1 Also. /dw/ occurs steps that Greenberg posited in his description. in the data. but only in one example. and in that example is freely variant with /d./.2 For this reason the phoneme /dw/ was not posited. /fy/ also occurs but not in the selected data. and thus does not appear on the chart. Mb g stribution and Phonologcal Shae There is more than one possible analysis of Hausa syllable structure. A common analysis is that Hausa consists of long and short syllables. and that long syllables have tw0 variants (cvc and cw) resulting in the types CV. CW. and CVC. This my well. be the best analysis for the language as a whole. but it may not be the best analysis for ideophones. The long CW syllable. though cannon in the rest of the language. is uncommon in ideophones. This plus the fact that vowel length may be non-distinctive in ideophones accounts f3}. decision here to represent syllable structure in Hausa ideophones by closed and Open syllables. i.e. the types CVC and CV. The two syllable types CVC and CV may both occur in isolation. or word initially. word medially. or word finally. The distribution of syllable types in relation to tone and vowel length is charted below. big: tone low tone high tone low tone 3311”“ m” long vowel long vowel short vowel short vowel aerVC X I WW I X X X T/yaa yi as at/ 'he relapsed into silence' fitagiaiytafiyaaM“tarlaatafiyaaM/'sheswalking ea ' 9 That is to say. long vowels may not occur in a closed syllable.l Ideophones may assume a variety of phonologcal shapes. These various shapes are charted in Appendix II in terms of syllable dis- tribution. tone pattern. and fmquency. The most common shape for an ideOphone to assume is CVC with a high tone. There are 68 examples of this Wpe in the selected data. The second most common shape is CVCVC with a high-high tone pattern. There are 39 instances of this type in the selected data. The third most common shape is cvccvc (with a low-high tone pattern). There are 27 instances of this type. It is interesting and indeed signi- ficant. considering the subsystemic nature of the ideophone. that the exception in phonological word shapes for the rest of the language is the rule in the phonological structure of ideophones. Most Hausa words (excluding ideophones) aid in an opal syllable. The majority of ideophones do not.2 Iowel Estribution and Tone The vast majority of ideophones employed in this stutw manifest perfect vowel identity. In ideophones of the type CVCCVC with a high- higl tone pattern (file second most common phonological shape. the first being monosyllabic) there is no exception to perfect vowel identity. and the vowels manifested are a-a. 1.1. or u-u. These vowels have the higrest frequency of occurrence in ideOphones. while /e/ and /o/ have a comparatively low frequency of occurrence.3 1"This is not subsystanic but is true of Hausa in general. 2There are 301+ ideophones in the selected data which end in a closed wllable and 101 which end in an open syllable. 3The infrequency of /e/ and /o/ is true for the whole language. C7. The relationship between vowel identity and tone can be stated 10 in terms of the following generalization: tone identity implies vowel identity The reverse is not true: that is. vowel identity may occur without tone identity. Alternatively. the generalization may be stated: vowel disparity implies tone disparity Again. of course. the reverse is not true. Tone disparity may occur without vowel disparity. There are seven exceptions to the generali- zation out of 37 examples of vowel disparity in the data (and 368 emples of vowel identity): lukwi-lukwi riya-riya sako-sako § § S carkeoleeo s § § buuyaayaa § Q kiikam § ~ mkwii It has been mentioned in this paper that long vowels are less common in Hausa ideophones than in other lexical elements in the language. One suspects accordingly that vowel length is non-phonetic in the phonological subsystem of ideOphones. The following rules. with w some exceptions. attempt to predict length in ideophones; /ee/ and /oo/ occur in any CV syllablel 1An exception to this is the ideophone sake-sako which ends in short /o/ - possibly explained by the fact that identity of tone pattern in a rednplicated form tends to produce a levelling of vowel quantity. 11 /ii/ occurs in open word-final syllables with high tone if the tone on the previous syllable is low1 /aa,/ a. occurs in ideophones of the type CHCVC on the low syllable2 ‘ b. occurs in CVCV-CTC'; on the final syllables: also occurs in some cv(c)cvcv word types /uu/ 41. occurs in ctcv on the high syllable b. occurs in cvcvlctct on the final syllables: also occurs in some CV(C)CVCV word types The vowels /i.i/. /aa,/. and /mr/ also occur as v in CV word types (that is. cv in isolation) but arch a length is quite unrelated to Iinherent vowel length. Rather. it can be dragged on interminably.‘ depending on how extensive a sound is being mimicked and how long the speaker? s breath holds out. For example: ‘ s s s o. e 3 maciidii ya naa tafiyaa §_I.I_u_.... snake it in-state-of going 'the make is slithegigg along‘, Occasionally this type of prolonged length occurs in ideophones other than the type CV. For emple: itaacee yaa faadii $2.... tree it + compl. fall 'a tree fell W3 Like length in CV word types. this largth would not be enacted to conform to patterns of predictability. 1m exception to this is the ideophone rumcwii. although the reduplicated form of the same ideophone follows the rule. 2'I'hese word types line up semantically with CHCCVC in the sound symbolian categories. time the lengthening of the vowel in the CV- syllable is not unexpected. In the emples in context throughout this paper the ideophones and their free translations are underlined. 12 Consonant mstflbution Only the following consonants may occur word finally in ideophones: p. t. k. f. s. sh..m. n. l. r. w. y. This group of consonants foms two classes according to the divisions in Chart 1‘.' The first class is — _ most elegantly de- + some picted in terns of - checked distinctive features. _ voiced p.t.k. f. 3.311 - sharp , - flat The second class is l i Gr 313/7" i) all the resonants in leroWeonnon . I“, Chart 1: V” W 1- " J V The first consonant in a cluster occarrring word medially is invariably one of the set m. n. r. l. w. y. unless the cluster consists of doubled consonants. The restriction isolates the class of cones: nants labelled as resonants in Chart 1.1 In three syllable words ending in an open syllable and in some three and four syllable words ending in a closed syllable. there is a pattern of identity of the second and third consonants. This is dis- cussed in the section on Moghgphonemics under the heading Recurring Partials. Intonation and Juncture The normal declarative intonation pattern of the Hausa sentence is a downward drift which maintains tonal distinctions but with the absolute pitch of each tone progressively lower on a five-pitch scale relative to its position in the ntterance.2 To illustrate: lThis seems to be true for the language as a whole. thus should not be considered subsystemic for ideophones. zThe idea of a five-pitch scale is from Kraft's An Introduction to 30km Hausa. pp. 17-21 (as originally described Carl ton s ausa° Basic Course (Washington. 1963). by e Hodge in 13 E 233- dau yaa—faa ‘ A shi Ra k 3 ree igi (£11 2 da 1 yaa sa _ \ s e I e o yea dauree shi da igiyaa yaa faadii kasa , 'he tied it with rope' '_it fell to the ground'_ It is normal for the ideophone to violate this pattern. The ideophone occurring utterance finally is likely to have an extra hig pitch which igores the dovmward drift of the rest of the utterance. or it may have an extra low pitch. To illustrate: malala til _ j mmLaLkm yaa 1? taa gu $1 an bu 2 1 taa ruwaa yaa kwantaaw yea 135ng 'water is lying all over the place' 'he's guy j ta gaa llv fa yaa tsa j naa ta shinsa _2. 1 saa Ea Sal Sal Sal fl baatsaatsaa Q Q Q Q Q Q Q Q Q“ . Q‘II‘Q“ “*Q ”m.” ta naa tafasaa 5g 6g; Sal gaashinsa yaa tsayaa W 'it's boiling brig? 'his hair is standing fl end‘ It is common for the ideophone to be set off from the rest of the sentence by a pronounced juncture. This is equally true of utterance final position and utterance initial (emphatic) position.1 and seems to lFivaz associates pronounced juncture with an aberration of the tonal pattern in his discussion of Zulu ideophones. "The junctures surrounding ideophones have the effect of isolating the ideophone tonologically Iran the rest of the utterance. Across these junctlres no tonal down-step takes place." Derek Fivaz. Some ASpects of the Idepphone in 912 (Hartford. 1963). p. 17. 11+ be directly related to the "cause and effect" notion or |'sentence alooiness" of some ideophones.1 _: lSee page Z? of this paper. MDIPHOPHONEMICS The sanantic value of ideophones is derived from internal and external factors. The internal factors are overall phonological shape and the recurrence of certain phonemes. both of which contribute to the non-arbitrary nature of the ideophone. that is. its sound symbolisml handle the internal factors morphophonemically. Sane reference to The external factor is context. It scans appropriate to context will be made later. Eowel Egggtion Altemation among full phonemes is common in the phonological sub-system of Hausa ideOphones. Two kinds of alternation are appa- rent: one which seems unrelated to a change in semantic value. and another kind which affects semantic value minimally so that two words can be very similar but yet not identical. The first kind is sensu- tically irrelevant (SI) alternation and the second. semantically relevant (SR) alternation. In the vowel systan. alternation is found among short vowels only. SR alternation is restricted to alternation between /a/ and /i./. and there seems to be no SI alternation between these two vowels. SI alternation is found between /a/ and /o/ (only one example occurs in the data). between /a/ and /u/ and between /i/ and /u/. The most Sometimes referred to in linguistic literature as honetic symbolisn or phonaesthesis. . p 15 16 prevalent form of SI alternation is of /i./ with /u/. Vowel alterna- tion can be diagrammed as follows: SI 1 """" I u I, so N ~ e ‘9 ‘7/ ,o It, / x a, 9 I "I +grave The “attraction" between /a/ and /u/ and the “repulsion" between /a/ and /i/ would be evidalce in the subsystem for categorizing /a/ as [+ grave] in a distinctive feature analysis. as shown in the secbnd diagram above. The slight difference in ”meaning" betwem SE-differentiated pairs is often brought out by differences in contextual distribution. For emple. in the pair of ideophones /farat/ and /firit/ which mani— fest /a/ and /i/ in SR alternation. both words mean something like “suddenly" but /farat/“i;s commonly used in the context of a ridden entrance into a conversation. while /firit/ is commonly used in the context of a sudden entrance into or exit from a place. For ample: Q yaa yi gm yaa amsaa he 4- compl. do he + compl. answer 'he answered Mi yaa rite firit he + compl. go-out 'he left Mi Apparently. /farat/ is not necessarily restricted to the context of conversation: /firit/. however. is not used in such a context at all. Other examples of SR alternation between /a/ and /i/ are: 17 car 'it balances - referring to money'_ cir 'it balances. is sturdy - referring to things. mmiture. etc. caf "an admit catch‘ cif '_a sudden stop', cak '_the rain has stopped M? cik 'fa man. train. or car has stopped completgz', Ehcamples of SI alternation between /i/ and /u/ are: air. zur '22-! red' rimis. mmus 'meat. leaves dried up' birjik. mix-31k 'people. grain in grammes! An example of SI altemation betwem /a/ and /u/ is: lagaf. luguf 'limp. overcooked'. W The problan of detemining the limits of altemation with conso- nants is far greater than with vowels. if only because there are more consonants. Yet there my be an explanation other than this obvious one. for there is a sense in which there is greater phonetic similarity among vowels than among consonants. That is. when the whole set of vowels is considered for possible pairings. one against the other. one would not be as air-prised to find SR or SI alternation between any two long vowels. or any two short vowels. as one would be to find such altemation between any two diffuse consonants. for example. or any two voiced consonants. Theifact is that SI and SR altemation can be“ observed between the most unexpected pairs of consonants as well as between expected pairs. The question is whether or not it can be usemJJy described. One soon 18 reaches a point of no return whereby all words that rhyme and have something in common semantically are candidates for SR alternation. if not SI alternation. It is only reasonable. then. to restrict attention to SI alternation among consonant phonemes. One nomally expects alternation between mu phonemes that share distinguishing features. But among ideophones. where seemingly unlike phonemes alternate freely. new criteria must be sought for isolating alternating pairs. In other words. it might prove reveal- ing in the areas of sound symbolisn and expressive speech to search for phonemic alternation between consonants previously believed too unrelated phonetically to be paired in this W} It seens useful here to consider the notion of primary opposi- tions. that is. oppositions which cut the whole set of vowels or the whole set of consonants more or less directly in half. In the Hausa vowel system. this would give us one primary opposition. i.e. long vs. short. but among the consonants four primary oppositions: voiced vs. voiceless. stapped vs. continued.2 diffuse vs. compact. grave vs. acute.3 Two consonants might conceivably alternate even if they share only one primary apposition.“ lThe writer has in mind the types of relationships between consonants set forth in Kenneth L. Pike‘ 3 Phonemics. p. 70. 2There are enough steps in flames to warrant this a primary opposition. 3These last two oppositions are used here with the understanding that they can be defined in articulatory terms as well as acoustic terms. LPIn the phonologr of Hausa ideOphcnes. the consonant pairs which are totally unrelated in terms of the primary oppositions are: /p/ with /y/. /i/: /5/ with /8h/. /C/: /b/ with /sh/. /<=/: /h/ with /d/. /d/: /R/ with /2/. /n/. /l/. /r/: /Rw/ with /2/. /n/. /l/. /r/: /Fressly. for the sole pulpose of', The altemation is semanticaJJy relevant in the first pair and semantically irrelevant in the second and third. These are the only emples of metathesis in the selected data. SYNTAX The grammatical distribution of ideophones can be described in terms of one list of words with members that function in various positions within the major clause types;L A discussion of these clause types is not necessary in the present analysis. What is sign- ficant for Hausa ideophones is that they do occur in all major clause types. although this does not seem the case with ideophones in some other African languages.2 The question arises whether ideophones are constituents of clause level constmcfions or whether they are in a sense superimposed on the clause stmctnre. There are three reasons for raising such a question. First. ideophones have been classified with exclamations and interjections in some previous analyses of African languages: second. there is often a pronounced juncture between the ideophone and the rest of the sentence:3 third. there is no overt syntactic linkage between ideophones and other words in the sentence. whereas most other grammatical units in Hausa normally manifest syntactic linkage of some j'Krafl: describes five major clause types in his Stag of Hausa max. It is obvious from the data for this thesis that ideophones may occur in any of the five types or sub-types within the five types. 2‘For example. Fortune' 5 description of Shona ideOphones restricts their occurrence to four construction types only. These construction types would represent finer divisions than the clause types of Kraft. and thus represent a significant contrast to the wide distribution of ide0phones in Hausa clause types. Fortnna. op, cit.. pp. 6-17. 35cc page 13. 1 27 sort.l Although the possibility of a special by-systanic relationship between the ideophone and the rest of the clause might be suggested for some ideophones.2 one could not posit such a relationship for ideOphones in general since ideOphones can occur as obligatory constituents within the clause as well as. and more characteristically as. Optional constitumts. In no sense could an obligatory ideophone be said to be "superimposed" on the clause structure. In describing the possible positions of the ideoPhone within the clause. the most useful division of syntactic constructions for this description of ide0phones does not coincide mth a typological division of clauses. probably because clause types in Hausa are defined primarily in terms of differences between predicate nucleii. whereas ideOphones typically occur outside the predicate nucleus. The references. then. will be to'kinds of‘constluctions. rather than clause types. The labels that appear with the examples below are meant to isigli‘iy positions within the“: constructions. The type of element which fills a particular position is indicated only when the descrip- tion danands it. Thus the label :3 would read "obligatoly nominal position” and fill; would read "obligatory nominal position filled by a noun.“ The following labels appear in the description that follows: 1Such as gender agreement. a genitival link between possessor and possessed. agreement of verb forms with their objects. agreement of verbal auxiliary (person-aspect indicator) with the subject noun. etc. 2As observed by Mr. Ishaku. the relationship between certain sentences and certain ideophones seems to be one of cause and effect. The sentence states the cause “the stone fell into the wateru and the ideoPhone states the effect I'a sound like tsundum.u + obligatory I+ Optional N nominal position V velbal position H modifier position PA person-aspect position All adj ectival nominal position transitive verbal position intransitive verbal position E'Pf: of: locative phrase position n noun ideo ideophone Ideophones in Obligatog Positions In the first constmction type to be considered. the subject nominal is obligatory in the absence of a nomally obligatory person- aspect indicator. and the predicate nucleus is filled by the obligatory velbal _n§._a_ followed by an obligatory nominal: + N + V + N mm as. tafasaa water in-state-of boiling 'the water is boiling' If modified by an ideophone the obligatory nominal in the predicate can be deleted. in which case the ideophone itself becomes the obliga- tory element: 30 + N + v + Mn 1’ m: ideo Q Q Q Q Q Q ruwaa naa tafasaa Sal-Bal-fi water in-state-of boiling '_the water is boiling brislfiz' + N + V + N2 ideo ruwaa naa E‘z.':tl-t‘>::ll-63l~ water in-state-of '_the water is boiling brifl', An obligatory ideOphone can also occur with the verb 2.1 in which case (and this is the ordinary situation in Hausa clause structure) the subject nominal is optional and the first obligatory element is the person-aspect indicator. The obligatory verbal elanent is followed by an obligatory nominal (the filler of) which is deletable if modified by an ideophone. the ide0phone itself becoming the obligatory element in the instance of such a deletion: _._—.—— H r... .- + PA 4- V + N am yi nauyii they do heaviness ‘. they are heavy' +PA +V +Ntn sun yi nauyii they do heaviness '.they are ram heavy', ol+ M: ideo Q E + PA + v + H: idea sun vi iiééb they do {they are very heayz', 1it is interesting to note that the commonest ideophonic construction in Shona is also the verb "do" (-_t_i_ in Shona. E in Hausa) followed by an ideophone. Fortune. 92, cit.. pp. 6-7. 31 Every instance of a potentially (structurally) obligatory ideophone will not be realized as such. as there are sanantic restrictions in some cases. Just when the obligatory nominal modified by an ideophone can be deleted and when it cannot is not clear. but it is clear that the conditioning is semantic. not structural. It“ the situation calls for it. an ideophone can function as an obligatory nominal without a structural deletion: s s s 1 abin nan Regan-R azau nee thing this is 'this is cruncm' Ideo bones in tional Positions In one construction type ideophones occur with adjectival nominals2 as optional intensifiers: + PA + v + N I AN naa gaa tsuntsuu farii I + compl. see bird white 'I saw a bird that was white' lThe ide0phone here occupies an obligatory nominal position. The construction (in terms of Kraft's clause types) is + NP + NP + 992. 2idJectival nominals comprise a semantic and distributional sub-class of nominals such as far}; 'white'. bakii 'black'. zaafii 'hot‘. dayaa one. etc. 32 +PA +v +N f(+AN imzideo) naa gaa tsuntsuu farii fat I + compl. see bird white {I saw a bird that was pure white', If the adjectival nominal were to precede the noun it qualifies. it would manifest a genitival link: naa gaa farin tsuntsuu I + compl. see white-of bird '_I saw a white bird' It is interesting to note that no linkage occurs when an acmectival nominal precedes an ideophone. as in the first example on this page. It is also interesting that the ideOphone must follow the adjectival nominal directly and thus could not occur in the example immediately above where the necessary syntactic linkage prohibits such a position- ing. In another construction type ideophones occur with transitive or intransitive verbs as optional verb modifiers: 1' N + PA , + v + N 3 1-1: ideo mutaanee sun cika daakinsa cunkug people they 4- compl. fill room-his 'peoPle filled his room to overflowing', +PA ‘+ Vt +N in: ideo mun kashe garaa tururu we + compl. kill termites 'we killed throngs of termites‘ + PA + vi 1' LP 1' H: ideo yaa faadii .5 mm tslndum it +compl. fan to water 'it fell into the water alash', 33 i N + PA + vi i tuulun yaa cika maaRil jug it + compl. _ fills 'the jug is full to the brim' If one were to make a case for the “aloofhess” of ideophones in the clause structure. it is this use of the ideophone as optional verb modifier that would provide the best evidence. Here is where a case could be made for "topic and comment" or ”cause and effect" relationships between the ideophone and the rest of the sentence. We It seans beside the point to categorize the ideophone grammati- cally} The ideophone is primarily a modifier. and if it modifies a noun. then it functions adjectivally. men it replaces the noun it modifies. then it functions nominally. Or it may function nominally without a structural deletion. When it modifies a verb. then it functions adverbially.2 The following examples illustrate ideOphones fulfilling various grammatical functions within the constrictions. l. Ryan-Ryan functioning (a) adverbially and (b) nominally: (a) almiyaa ta n£a sin karmaamii kgyaufigyau she-goat she in-state-of eating leaves {the goat is munching on dried leaves' lIt is interesting to note here Westermann's ambiguous grammatical classification of Ewe ideophones: ”From their function they must be classed as adverbs. for they all describe the action of a verb: but most of them may be used as substantives or adjectives.” Westennann. mitts: P- 187- 2The terms "afiectivally" and "adverbially" are employed merely as points of reference. The morphological categories of adjective and adverb do not exist in Hausa. See previous footnote on page 2.. 34 (b) abin nan kayau-Rayau nee thing this is '_this is crunghy', 2. big'ik functioning (a) adjectivally and (b) adverbially: (a) almai hatsii birjik .3 kzasuwaa there-is grain at market 'there is plenm of grain at the market'_ (b) mayaakaa sun cika fiilii m warriors they + compl. fill field 'warriors W covered the field', 3. Bul Bul mnctioning (a) adjectivally and (b) nominally: (a) sun yi aiba‘ta 6u1 Bul they 4- compl. do fatness {they are My. (b) yea. yi Bul 6ul kamar giiwaa he + compl. do like elephant 'he's as fat as an elephant'. The h s T sfo tion The ideophone normally follows the element that it modifies although not necessarily contiguously. That is to say. an ideophone functioning adverbially will follow the verb and also the noun object if there is one. The result is that the most common position for an ideophone to occupy as modifier of the predicate is at the end of a clause or sentmce. However. there are permutation rules which move the ideophone (or other modifiers) to a position preceding the predicate for emphasis. Ebcample (b) below is the result of applying the emphasis transformation to (a). Ebcaunples (c) and (d) are not related in this way. but merely illustrate the ideOphone tag in normal 35 position (c) and emphatic position (d). Q (3.) ya naa taflyaa tinkis tinkis (b) (c) (d) he in-state-of walking 'he is walking slowly and tiregz'f s o s 1 s . e tinkis tinkis ya kee tafnyaa he in-state-oi‘ walking 'slowlz and tireglx is how he is walking'_ mun ci abinci mun yi _tgn_ we + compl. eat food we + compl. do 'we've eaten and are stuffed'. s _ s s s s s mmu kee bazaanmiyacanwaniabuba we in-state-of neg. fut. we can eat certain thing neg. ‘stszed as we are. we couldn't eat another thing', 11333 is in complanentary syntactic distribution to fig. p¥finf§ PmBLEMS IN DETERMINING MEANING It is characteristic of ideophones that they are not pure symbols. but require a context or situation in order to have semantic value. This semantically elusive nature of the ideophone creates problems in analysis. some of which are discussed below. The Problem 01.: Context It has been suggested that a sentence acquires meaning by virtue of a kindfiof near}; away of possible meanings'through con- textual clues until it comes to mean whatever possible meaning is 1 If this is true left after the subtraction process is complete. for sentences. it is true to a far greater degree for words. and especially words such as ideophones which are virtually meaningless (or multiply meaningful. which amounts to the same thing) apart from a context. Harv words in language are meaningml with or without a context: that is. they symbolize more or less equally well with or without a referent. The ideophone. however. requires a referent. although to be sure this referent may be either linguistic or non- linguistic. That is to say. the particular context which peels away all other possible contexts may be a noun. a verb. a phrase. or a clause. or it may be an event or situation.2 But some type of l'Sentences mean what they do. so to say. BY DEFAULT. the competing meanings having been eliminated in one way or another." Martin Joos. 2§?5§)°g5§-78 Linguistic Theory of Meaning.“ Studies in Linguistics. XIII 9 c 2 As if the event or situation were the “topic" and the ideophone the "cement. " QR 37 referent. be it linguistic or non-linguistic. is essential if an ideophone is to I'mean" something. An ideophone has a broad semantic field in which it operates. which is why we are able to approadmate a definition in a list of ideOphones out of context. But this is also why a literal translation has been consistently avoided in examples in context throughout this paper. The ide0phone acquires meaning within the context of the whole sentence. hence its meaning is most accurately glossed within a free translation. The Problem of Semantic Qggggzation The various semantic fields of Hausa ideophones fall into categories which are worth listing inasmuch as they reveal the areas of meaning in the language "covered" by ideophones. But there is the danger of imposing artificial semantic categories on the language. rather than allowing the natural categories to reveal themselves. Perhaps a valid method of delimiting semantic categories would be to present several native speakers with ideOphones listed without reference to context or translation. and ask that they arrange into groups those items which seen to ”fit“ together. The resulting cate- gories might have greater validity than an externally-conceived listing in which there would necessarily be much overlap due to non-native judgment. . A thorough attanpt at semantic categorization is not possible at this time. A broad approximation of the semantic categories pertinent to ideophones in Hausa is as follows: movements. sounds. appearances and states of mind. and intensity or measure. An attempt was made to 38 use Samarin' s fifteen categories for African ideOphones.l but it did not seen possible to describe accurately the semantic categories of Hausa ideophones in this way. For one thing. the senses are repre- sented by two verbs (instead of five as in migrish). i.e. 11.1 'hear. taste. snell. feel' and Lil-é '_see'. This would seen to indicate patterns of perception that an external list such as Samarin’ 3 might likely distort. e P blem of terminin tionshi 5 Between Mo heme Structure gd Meaning To come to terms with the phenomenon of sound symbolism is to attempt to isolate the elenents within an ideophone which make it ”mean" one thing rather than another. that is. to isolate the predict- able elanents within a morphane. This has been attempted in Appendix III by grouping tagether ide0phones with similar sounds and related meanings. It should perhaps be made clear that these sound/meaning groups were arrived at by inspecting data. In other words. at no time was a native speaker of Hausa asked to list all the ideophones meaning “splash in deep water" or “a dragging sound." It was strictly a matter of matching sound with meaning after the data was collected. It was not apparent at the start of the project that arch groups could be isolated. Determining the exact nature and extent of sound/meaning groups would of course require further research. Appendix III is an approxi- ISamarin has suggested the following semantic categories for classify- ing ideophones in African languages: appearance. arrangement. emotion. meamre. motion. odor. quality. shape. sound. state. taste. temperature. time. touch“ and weight. William J. Samarin. ”Perspective on African IdeOphones. p. 119. 39 mation of what some of these groups might be. At this point in the research one can only suggest the nature of the non-arbitrary elements of morpheme structure in the sound/ meaning groups. The following list is an attempt to suggest non- arbitrary morpheme structure in some of the sound/meaning groups. Reference should be made to the groups themselves in Appendix III. Camel“ 'the sound of a splash in deep water‘. 03w 'the sound of a blow’, can 'a bad feeling or smell‘, cli(c2)03is or 011(02)03ii 'a dragging sound', (tone patterns variable) clvlczvlczvl 'spread about. spread out' (either all high or all low tones) C]_V(Cz)03000300 'unbalanced. loose' cilcx‘rls '_ cruncmr. coarse' | I cvlcvli‘ limp Even when the non-arbitrary elements of morpheme structure are ‘ determinable. it is not to say that these are necessarily the onomato- poeic elements within the ideophone. but rather that these are the elanents with which the native speaker associates certain noises or movements or states in real life. For there is a relationship between sound and meaning that is well established in the mind of the native speaker. established. that is. psychologically. in the manner of phoneme recog‘rition. It is perhaps one reason wlw the language learner renains an outsider for so long. For until he masters the sound/meaning relationships. i.e. until he can predict what an ideophone will probably mean., the foreigier remains well outside of the Hausa speaker's frame of reference. APPENDIX I THE AFTDASIATIC AND NIGER-KORJDFANIAN LANGUAGE FAMILIES Niger-Kordofanian Nige ngo Kordofanian West Atlantic Mande Voltaic wa Benue-Congo damawa I-ULANI ENE NUPE YORJBA B id TIV B 17.1 SHONA H] Rm Greenberg's The Languages of Africa. 1966. The names of , languages are in capital letters. The names of language families and sub-families are in small letters. 40 APPENDIX II PHONOIDGICAL TYPE OF IDEDPHONES Monosyllabic and disyllabic forms are listed according to the distribution of syllable types within the word. Tone is indicated. Polysyllabic forms are listed according to the number of syllables within the word. plus the recurring syllables. Tone is not indicated except in citing the most common types. The frequency with which each phonological shape occurs in the selected data is indicated in parentheses. Variations of mpes are listed in descending order of frequency. The symbol etc, means that a particular form can continue to be rednplicated depending on the erdcent of movement or sound of the referent. Monogzllabic Ideoghones cvc (68) cvc-cvc (13) etc 69c cr‘rc etc. (8) mic (16) 0170.090 ( 8) cvc cvc cvc etc. (5) 0370 (8) ‘ cfrc-cvc (1) ‘ cvc c§c cm (2) 0% (11) air 01? G1} etc. (4) cv ( 2) cv cv cv etc. (2) Maggie Ideophones cvcvc (39) stoic-civic (9) cvcvc cvcvc cvcvc (1) cvcirc (17) cficvc- sieve (a) circle claim 6170176 (1) crime (7) cvoirc- cvcx‘rc (4) cvm‘rc (3) cvcvc-cvcvc (3) § ~ cvcvc-cvcvc (3) cicim-cvcvc (1) aficcvc (27) . cvccvc-cs‘tccfic (3) circcvc ciccvc circcvc (1) cvccvc (4) cfiocvc-cficcvc (3) cvccx‘rc ( 3) cvccfrc-cvccvc (1) 1+1 42 cx‘rcv (4) cvcv-cvcv (10) cvcv (1) cvcv-cx‘rcx‘r (9) circx‘r (1) circv—cx‘rcv (6) over? (1) cvcx‘r-cvcir ( 3) S s cvcv-cvcv ( 2) cvcv-cvcv etc. (1) circcir (1) cficcv-cficcv (2) Po ic hones closed syllable final: 515253 (22) 815253-513283 (5) 31 32 52 ( 9) 5‘1 S2 S3 54 (3) s1 52 53 53 (2) The most common type is ‘31 52 S3 (12). open syllable final: 5"1 S2 32 (38) 31 S2 S2 ‘ S1 32 S2 (1) s1 52 s3 ( 2) The most common types are $1 S2 $2 (19) and §1 52 §2 (17). m The majority of polysyllabic ideophones ending in a closed syllable are of the type 81 S2 33. The majority of polysyllabic ideophones ending in an open syllable are of the type 31 $2 $2. Altogether there are 1148 monosyllabic. 175 disyllabic. and 82 poly- syllabic forms in the corpus of 405 ideOphones. APPENDIX III SOUND/MEANING GROUPS Splash in deep water 2£&&M&_ raham tsdndm tsindnm zundum bundum bunjum kindim mnjum bunjdm bundum-bundum glash in shallow water facal Q Q face-face up-and-down motion as in bubbling or waving Q Q Q dul-dal-chl Q Q Q Bal fial 6al bail-bah fil Q Q Q pil pil pil dripping £fifififi Q Q Q Q Q ra ta ta ta ta 43 flutter of wings fir fur fir-fir-fir 3‘irif foots s hooves 1.;er kararap sakalat sakalat newsman. mmmqmuwe tinids-tinkis talkie-tins tikis tiSis biris air nikikii niRi-niki ninkwri-rdmmn bikini-bikini ciBi-cifii tiRis-tikis W falling: the sound upon O dim or dim . a bum Q Q Q tim tim tim Q Q gem-gem Q Q Q fam-fam-fam faram-farsm burum Q Q rugum Q Q Q ruguzum katsgm kwatsam Q kwaram mas Rus Rus Rus mukus karas Raras (sound of eating nuts. raw potato. etc.) bgmd regch ken katakau i‘intinkau kintdnkan fan fanfau sonnets: cik cif caf shaf kaf Rat Rakat cancak «Sneak dungum 19335191111 RScSokan dookacakam kaca-kaca dried 11 well- cooked rimis rumus W bulus lalas 11118 11613 didis lumus luk wl-lnkwi coarse texture. cruncgy Q Q Q Q Bans-Barns Q Q Bares Q Q Q Q garas-garas Q Q Q Q giris-giris Q Q Q Q (cums-Rams Q Q Q fantsartsar preponderance of things or peggle birjik bdrjrk farjak bdrduk name oinkis ainkus timis firmis «learn oincim cinema Q mm 6338.0 8 1‘68 #6 all over. flailing intensitz of color malala zir lamama zur yamama wur samama wul ratata sul raBaBa caBaBa da5a6a ti ht st ht rashashaa near. 2:: shoemeemEe kir hululu das rututu dos furtu’or tam mdudu kam mruru an; 3:" gem-gem zam-zam m kyam wuf sham caf aankarai carat Barbadai karaf simindai intensig of taste or feeling buttocks shaking c1311 bdsrr-bdsur clan tuBur-wtfiur lau _ mmfiur-tumsur kalau katafiur-katafiur 47 W esoksokso cark30n30 cirkoo-cirkoo oirkéokéo dagoogoo ' cSokSokEo sake-sake moo-131m 1350045500 katoo-Ratoo on fter the other rii rii rii rii rival-ma suction (as when walking in.mud2 ' Q 'Q easel-oafial case-cise t still tsai tsit mum-marvmarlnar karhkarhkarbkar Q Q darwdar 1 11 t t luguf lugub-lugub summsum moist fresh ahar sharaf sharaf- shin}: Shaka! shatai‘ char- sham thick. hem QQ kitdb e e e kitib‘bfl) 10151115111 e s s katafiarfiar jagalgal noiselesilil snoot_hl_l,z Q amen. QQ Q 1+8 shining fa]. Ryel-Ryal swans. 911W budu-budu butu-butu buzuu-buzuu Q Q Q bsuzuuzm Q Q Q buram suu Q Q Q aayaa beatsaatsaa mass movement Q m.... (m‘uOOOO bm‘ O... Q tum... Q We... the..." Q shes..." Q 083.... restless. discontented tunicur Q Q inns Wu tsuruu APPENDIX IV EXAMPLES OF IDEDPHONES IN mNTEXT lgcookan Q Q Q Q Q Q yaa, dauki sulensa kacookan yaa baa yaaroo he+compl. take shilling-his he-I-compl. give boy "he took his lgst shilling and gave it to the boy'_ an daukee su g‘cookan one-i-compl. carry them '_they have g; been removed' Q Q Q sun taashii kacooggn they+compl. get-up {they moved out lock. stock. ang barrel' gag: sai zakaraa yaa cafke faaraa yaa hadiiyee then rooster he-I-compl. snatch locust he+compl. swallow 'all of a sudden the master snatched the locust and gulped it down‘ yaa cafe RwallSO _c_a_i_‘ he+compl. catch ball 'he caught the ball adroi_t._'|i' riigar yaa yi gar; da shii gown-the itI-compl. do with him 'the gown suits him', 1:29; IE taa shaare gidanta fig; she+compl. sweep house-her 'she swept her house veg clean'_ mun shiryaa tsaf we-I-compl. prepare {we're g; ready' taa saunas tsaf cikin gidan mijint; she-I’compl. stay in home-of husband-her 'she lived peacem in her husband's house' yaa wanku _t_a_s_ it‘I-compl. is-washed 'it's washed mtless' Q Q Q Q i naa do. sulee goomaa _t_a_s_ I in-state-of with shilling ten '1 have 9g}; ten shillings'_ sun rashes _t_a_§ they+compl. disperse 'they have g; dispersed‘_ mwaa yaa daukee Lag water it+compl. cease 'the rain has co_mplete;y ceased' 51 i naa d; baawaa gudaa dayaa _t_ak__ I in- state-of with slave unit one 'I have 993.: one slave' koowaa ya cee 39$ sai. mutuwaa anyone he says than death ‘it is death to amone who makes even g single comment' QQ Saray'ii sun kwashe kaavans; E thieves they+compl. move stuff-his {thieves stole all his stuff'f sun yi mule; zoobEe 15g they+compl. do to-you circle 'they W surrounded yon? Q sake jikee shi lg; they+compl. soak him 'they attacked him tooth 2d fight? 1 an} d3 shoekaraa bekmh 9;; I in-state-of with year seven 'I'm exact]; seven years old', yaa tsayaa 9;; he-l-compl. stop 'he came to a complete stop', Q Main yaa dauru gig; box-the fill-compl. is-tied 'the box is tied tighgx't 52 2.1.: teebur yaa tsayaa 93:; table it-I-compl. stand 'the table is stum', sheekaraa ashirin g; sai naa ' gan k5 year twenty then I+compl. see you 'it's been a full twenty years since I've seen you' 135111511 yea oaks. pip jug it+compl. £111 'the ing is no.1 W‘ Q Q 922... Q Q Q Q Q yaa fadi maganaa cafipl he-I-compl. speak word ‘he spoke uncouflyi Q Q Q hanyaa tan yi 3939; road its-compl. do 'the road is mugg' Q Q Q Q taa sub da miyaa pp_S__ she-I-compl. pour with soup 'she made 5 bg mess when she threw away the soup‘. i.emul. yaa fits simul he-I-compl. go-out 'he ”got off" without wrong (i.e. he was acquitted)‘ 53 saman teebur yaa yi zumul top-of table ital-compl. do 'the top of the table is veg; anooth' yaa warkee sumul he-l-compl. heal 'he healed comletely‘. suBul zooben yaa fits sufiul ring-the it+compl. gO-out ‘the ring slipped off essay! Baraawdn yaa fits gig; thief-the he-I-compl. go-out 'the thief escaped eggs and sound'_ gaashin kargn bakii suBul nee hair-of dog-the black is 'the dog's hair is a glpgfl black? Mam mat‘fiyaa sun tsayga dznkam travellers they+compl. stop 'the travellers stopped short (undecided what to do nert)'_ Audi m as. Audu is 'Audu is tacitugp' ruwaa ya naa nan dankam water it in-state-of here 'this emanse of water is stamt' dungup zzn baa k5 shii dringg Iewill give you it 'I will give you all of it' sun bar kasarsd ddnggp they+compl. leave land-their "they completglg migrated' an yi mas; ding 'he's been stgpped of his possessions' Q Q QM Q Q Q Q Q Q baakon yaa tafi mlungdum guest-the he+compl. go-off 'the guest went off without a word', soojaa yaa tdfi salunédum soldier he-I-compl. go-off 'the soldier went off W'. jirgin Rasaa ya 133.3 tafiyaa m train it in-state-of going 'the train is approaching ggieLly (i.e. with its engines off)‘ APPENDIX V ALPHABETICAL LISTING 0F IDEOPHONFS IN THE SELECTED DATA The following ideOphones are listed without glosses in accordance with the decision discussed on page 37 to avoid literal translations of ideophones. The list is alphabetized according to the traditional orthograpiw. Q Q badoo Sal cancak bal-bal Sal-Gal car Q Q Q Q Q Q barkatai. Sal Sal Sal caraf Q Q Q Q Q Q Q Q Q Q baatsaatsaa Balak-Balak oarkooleoo bazar-bdzdr Sam-Sam cas bikikii-bikikii this sale biris assess. cawket ciwket céwket birjik Bingilgil cisi-cisi bdu. . . . 512.1 cif budu-budu Bul Bul cik bulus Buntur cincim bum Butlk cinkis bundum cir mam-bandh 0:3. 0 o e aanO-d‘lkoo Q bunjum caBaBa cookookoo Q . e e bunaum cafia-caba cukuu-cukuu bdrdnk 03.631 cunkus bdrjik sisal-easel cus Q Q Q Q Q bursuusuu case-case minim cEacim daBaBa bdsur-bdsur caf daBas butu-butu cai dafa'_~;n Q Q Q buuyaayaa cak d; l; 1; la la bumu-buzuu oaks-oaks (131:! Q Q Q Q Q Q Q Q Q buuzuuzuu cakar cakar cakar dandan 55 dinkmm Q Q das-das dirshmn dukuupdahuu Q Q Q dul-dnl-dul dum Q dnm QQ QQ dumus-dumus QQ durum dwds-dwds daf dagOOgoo dindak Q Q darbdar das dau Q dau deereeree dii. . . . didis dim dir Q Q Q dookacakam Q Q Q dookookoo dds Q dun.... Q dum ddngum £330.00 Q Q faca-faca facdpfaca raca1 Q Q Q faafaratan 56 faak 1.1 fal Q fal fam fan-fan fan-fam-fam Q Q Q fan-fam-fam Q Q Q fantsartsar farém-farsm farat farj ak fat Q fau Q fan fanfau fes Q fil Q Q fil-fil fintinkau fir fir-fir-fir jabs-jab; 35a51g31 jdngwam iisibsib 35mm ringinsim afraid-surge jirif stigma 313nm kacapkaca kaci. . . . k5e8okan kaf kakaf kakau kal kalmadaipkalmadai kann kan-kan karaf . . ~p-karardp karbkarbkarbkar an. ac kitasirsér katasdr-katasdr katakau katanas katsam kau kii. . . . midi-1:101 kici kici kici kici kindim kintinkau kitib kitibtib kueulsul kurum QQ kusukum kwal kwaram kwarlmatsan kwatsam kyam Raf Rakat Ralau kangaran Rarara kararau Q Q Q Q karts karas Rarau Rat Q Q katoo-katoo 57 A Ran was Rayau-kayan Q Q Q keereeree kii. . . . Q Q kiikam kirin Rundnm (cumus Q Q Rurungus Ruranus Rania-Runs kus Rus kus Rwal anal sear QQ Rmrangwam Ryal—Ryal Wesloeree Q Q Q lasasa 15500-13600 ldkoo-lakoo lamama lan lifiis lilis 11? lip 11p lugub-lugub luguf lukwi-lukwi lum lungum Q maskil malala mznkas marmarmarmar maze. mass-maze. minimini mukus Q Q muzuu-muzuu nikikii niki-niki pan Q Q Q Q P8 P8 PI PI pas pash 0 part Q Q Q p11 pil pil Q Q Q rabajaa rafiafia Q Q Q raf-raf-raf rak ramau mshaa-rashza rashashaa ratata Q Q Q Q ratatata rayan reeran rimis Q Q Q Mei-118311 rii rii rii riya-riya fins Q Q Q ruguzum rumus runkwii fimmnfimni rututu sak QQQQQQQQ sakakakakakakeka sako- sako Q Q smhushha sal 383131318 Q Q sans-sans Q Q Q sarBadai émku émmni émnémi sintsir smdnmwh §m&6o mm“. suBul Q Q mmmm éfimmmfimm shakaf sham shar Q shar sharaf elk sharaf sharaf Q Q Q Q sharef- SW shataf whafiu Q Q Q sheaneemee shirim- shirim shum Q Q shuree- smree &. Q 0 O taf- taf-taf tak tikis tikis-tikis deeds. dm thanfimn ththtm dam tinkis tinkis tinnis an”. tuBur-tzfidr tududu tuk tuk tuk tuk mmBur-bimsur tunkur tinkus-tunkus tamis 59 tsaf 25251 tsaf tsaf 251511 tsagaga zam-zam Q Q Q Q Q tsagal-tsagal zankalkal Q Q Q tsai zankarai tsam zaraa-zsraa Q tsindum zir Q tsirit . 2001:3050 tsit sunbur tsaclsolsc mndum tsdhuu 2111‘ Q tall!“ zumu-zurdn tsaruu tsumt Q Q Q “€388 Q Q Q ml-wsl-wal warai-waraipwarai washar Q Q Q Q watsal-watsal waayam wuf wul wr Q mr Q Q wurjanjan yama Q Q Q yam- am-yam yfirér yauo o o o BIBLIOGRAPHY exts for the Co 5 Fletcher. Roland S. flea Spyings and Folk-102p. Iondon: Oxford University Press. 1912. Fimtuwa. Rania: Yadudu. Sgfli Kunnenka Napp? Zaria: Gaskiya Corporation. 1960. Q-gg-QkNan; igpp_ru_n Nan 5p Qro Dagg Jakaiiyg. Zaria: Gaskiya Corporation. 1953. North Regional Literature Agency. Karamin Sani' Littafi Na Farko. Zaria: Gaskiya Corporation. I957. . Karamin Sani" Lit 1 Na B . Zaria: Gaskiya Corporation. 9 . T es for the Co us The tapes for this corpus wore recorded by Charles H. Kraft in Nigeria at various times during the last nine years. They are listed below by date. Recorded 1959-1960: 1. Hausa fables told by Andu Mwada 2. Stories told offhand and from Littafin Kgon @131 and Ka Kara Karatu by Musa Mwada. Audu Manda. and Sha ,aibu Mubi 3. Stories told by mdu Manda and 8113' aibu Mubi 4. Riddles told by Sha‘aibu Mubi and Ishnaila Mubi Recorded 1966: 5. man Mutum da Sana'arsa as read by Ibrahim Nada Recorded 1967: I ' 6. Hausa fables told by Umaru Mohamed 7. Hausa customs discussed by Umaru Mohamed and Ibrahim Mohamed 8. Life and courtship of Umaru Mohamed. Umaru Mohamed and Ibrahim Mohammad discussing Nigerian history. Umaru interviewing Ibrahim regarding the latter's life history 60 61 9. Life story of a Zaria woman 10. Hausa riddles told by a Zaria woman che; References Abraham. R. C. Dictionagz of the Hausa Langagg. Iondon: University of Iondon Press Ltd.. 19 2. . The Langgagg of the 'Hausg PeOple. Iondon: University of Iondon Press Ltd.. 1959. Bloomfield. Leonard. M. New York: Holt. Rinehart and Winston. 1933. Bolinger. Dwight L. 'Rime. Assonance. and Morpheme Analysis." word. VI (1950). 117-136. Brown. Roger W. Review of "Untersuchungen zur Onomatopoiiee“ by Heinz Wiesemann. Langggge. xxx: (1955). 84.91. Firth. J. R. Papers in Linafistics. ”£422. Iondon: Oxford University Press. 1957. Fivaz. Derek. Some Agects of the Ideophone in Zulu. Hartford: The Hartford Seminary Foundation. 19 3. Greenberg. Joseph H. "Concerning Inferences from Linguistic to Non-linguistic Data." no Merican Apthpppologis . LVI (December. 1954). 3.19. . Esme in Linggistics. Chicago: The University of Chicago Press. 1957. . The Langgages of Agricg. Bloomington: Indiana univerSiWO 19 o . "Some Problems in Hausa Phonolog.” Langggge. XVII Halls. Morris. "On the Bases of Phonolog.‘ in The Strucmg of Lanflge. Ed. by Jerry A. Fodor and Jerrold J. Katz. Ehglewood Cliffs: Prentice-Hall. Inc.. 1964'. Hockett. Charles F. A @Lse in Modern Idnaistic . New York: Macmillan Company. 1958. Hodge. Carleton T.. and Umaru. Ibrahim. Hang; Basic Course. Washington: Foreign Service Institute. 19 3. Jakobson. Roman. Fant. C. Gunnar. and Halls. Morris. P_reliminaries to Speech Analysis. Cambridge: The M.I.T. Press. 1965. 62 Joos. Martin. “Semology: A Linguistic Theory of Meaning." Studies in Linggistics. XIII (1958). 53-70. Kraft. Charles H. A Stug of Hausa mtax. Hartford: The Hartford Seminary Foundation. l9 3. Kraft. Charles H.. and Abubakar. Salisu. g Introduction to Smken Hausa. East Lansing: African Studies Center of Michigan State University. 196 5. Kunene. D. P. "The Ideophone in Southern Sotho. Part 1." Journal of Afpican @flages. IV (1965). 19-39. Longacre. Robert E. Grammar Discover; Procedures. The Hague: Mouton and Compamr. l9 . Ogden. C. K.. and Richards. I. A. The Meaning of Mpggpg. New York: Harcourt. Brace. and World. Inc.. 1923. Parsons. F. W. "An Introduction to Gender in Hausa.“ Afrifl Langgage Studies. I (1960). 117-136. . Letter dated November 22. 1967. from Mr. Parsons to Charles H. Kraft concerning the problem of defining Hausa ideophones. "The Operation of Gender in Hausa: The Personal Pronouns and Genitive Capula.“ African Langggge Studies. II (1961). 100.124. . "The Operation of Gender in Hausa: Stabilizer. Dependmt Nominals. and Qlalifiers.“ Afrifl Langgage Studies. IV (1963). 166.207. . "Some Observations on the Contact Between Hausa and Erglish.” Sypposium on Multilinggalian. Brazzaville: Commission for Technical Co-operation. l9 2. Pike. Kenneth L. Phonemics. Ann Arbor: The University of Michigan Press. 1947. Read. Allen Walker. "English Words with Constituent manents Having Independent Semantic Value.“ in Philolo Egg The Malone Appiversagz Studies. Ed. by Thomas Kirby and Harry Woolf. Baltimore: Johns Hepkins Press. 1949. Samarin. William J. "Determining the Meanings of Ideophones.” Egg Journal of West Agrican Langgages. IV (1965). 35-41. . The Gbpya Lanmage: Grammar. Texts. and Vocabularies. Beérlgeley and los Angeles: University of California Press. 19 . . "Perspe tives on African Ideo hones." African Studies mv (1965 . 117-121. p r ' 63 . Review of “Reduplicative Words in English. A Study of the Types Tick-tick. Hurly-burly and W.“ (Nils Thun. Uppsala. 1963). Journal of African Languages. V (1966). 161-163. Stennes. Leslie H. A Reference Grammar of Adamawa Fulan . East Largsing: African Studies Center of Michigan State University. 19 7. Westermann. Diedrich. A Stugy of the Fire Langgage. London: Oxford University Press. 1930. HICHIGQN STQTE UNIV. LIBRRRIES ll|l3|ll|1|||l|llllllllllllllHIIIHIVIIIl lllllllllllllllllllHl 293107144044