DOCUMENT RESUME FL 023 040 ED 382 037 Alidou, Ousseina AUTHOR Evidence for Foot Structure in Hausa. TITLE PUB DATE 95 25p.; For complete volume, see FL 023 039. NOTE Journal Research/Technical (143) PUB TYPE Reports Articles (080) Kansas Working Papers in Linguistics; v20 p1-24 JOURNAL CIT 1995 EDRS PRICE MFO1 /PCO1 Plus Postage. *Hausa; Language Research; *Linguistic Theory; DESCRIPTORS *Morphology (Languages); *Nouns; Oral Language; Structural Analysis (Linguistics); *Syllables; Uncommonly Taught Languages *Reduplication (Phonology) IDENTIFIERS ABSTRACT McCarthy and Prince (1986, 1990) have put forward the prosodic Morphology Hypothesis to account for morphological processes, such as reduplication and truncation, that typically require that their output conform to a particular shape of template. This hypothesis claims that morphological templates are analyzable in terms of prosodic units. This paper shows that Hausa nominal reduplication and nickname formation are best analyzed as involving the specification of a foot template. Thus these two processes provide supporting evidence for McCarthy and Prince's Prosodic Morphology Hypothesis. Contains 11 references. (Author/JL) *********************************************************************** * Reproductions supplied by EDRS are the best that can be made from the original document. *********************************************************************** U.S. DEPARTMENT OF EDUCATION of Educabonai Research and Improvement ATIONAL RESOURCES INFORMATION ED .EVIDENCE FOR FOOT STRUCTURE IN HAUSA CENTER (ERIC) "PERMISSION TO REPRODUCE THIS his document has been reproduced as MATERIAL HAS BEEN GRANTED BY received from the person or organization 0 originating it. 04,.(1._1'\ Minor changes have been made to improve reproduction quality. -Th-t OUSSEINA ALIDOU Points of view or opinions stated in this document do not necessarily represent official OERI position or policy. TO THE EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC)." Indiana University / Ohio State University A bs tract: McCarthy and Prince (1986, 1990) have put forward Prosodic Morphology Hypothesis the to account for morphological processes (such as reduplication and truncation) that typically their require that output conform to a particular shape of template. This hypothesis claims that morphological templates are analyzable in terms of prosodic units. will show this paper that Hausa nominal In I reduplication and nickname formation best analyzed as are involving the specification of a foot template. Thus, these two processes provide supporting evidence for McCarthy and Prince's Prosodic Morphology Hypothesis. 1. Introduction McCarthy and Prince (1986, 1988 and 1990) have observed that a variety of morphological processes in different languages require that output conform to their a particular shape of template. They refer to such phenomena as shape invariant morphology: The types of morphological processes that of a typically require the specification particular template include reduplication and truncation. In reduplication, what reduplicates normally is specifiable by a specific template, and, in many truncation processes the output also conforms to a specific shape. In order to account for such morphological processes McCarthy and Prince (1986, 1988 and 1990) have put forward the Prosodic Morphology Hypothesis. This claims hypothesis that morphological templates are analyzable in terms of prosodic units which can be defined as a syllable (and various types of syllables, such as light syllable or hez-1, syllable). Kansas Working Papers in Linguistics, Volume 20, 1995, pp. 1-24 r 2 In this paper I show that Hausa has morphological processes that make use of an invariant foot template. specially I consider nominal reduplication and nickname formation and show that these are best analyzed as involving the specification of foot a template. these two processes Thus, provide supporting evidence for McCarthy and Prosodic Prince's Morphology Hypothesis. This paper is organized as follows: Section 2 reviews previous account of Hausa nominal reduplication; section 3 presents the analysis of the same process within McCarthy and Prince's prosodic morphology framework; section 4 discusses the limitation of a CV-template account of the Hausa nominal reduplication; section looks at other 5 evidence besides reduplication for the role of foot in templatic morphology in Hausa, focusing on nickname formation; and section 6 presents the conclusion summarizing the results of my analysis. 2. previous Studies on the Hausa Nominal Reduplication This section deals with the analysis of a large class of a reduplicative nouns in Hausa, a Chadic language spoken in West Africa. this In of the language reduplication is inherent structure of the monomorphemic nouns presented in Newman (1986) as well as the class 2 nominal plurals discussed in Davis (1988). The most intensive and insightful study of the process of reduplication has been provided by Newman (1986) and his subsequent research on the subject. But prior to Newman's account of reduplicative nouns in Hausa was Gouffe's (1975) analysis of the reduplication process in the language. Davis (1988) also provided an analysis of Hausa reduplicative nouns within Marantz (1982) framework. Three proposals quite divergent from one another emerged from the tree studies. 3 3 Gouffe suggested that the Hausa nominal reduplication involves in some cases prefixation of the CVC-sequence --a copy of the initial string of the base -- to the base and in other case reduplication of the last consonant of the base and the initial vowel and consonant of a plural In (1) an suffix. example of the stem initial CVC reduplication for deriving nominal plural is illustrated with the word KaRfii, whose plural is KaRKaRfaa, and in (2) the stem final consonant /k/ and -un- from the plural suffix are reduplicated between the nominal stem and the plural suffix for deriving jakunkunaa from jakaa. (1) 'strength' KaRf-ii --> *KAR-KaRf-aa --> KaK-KaRf-aa 'strengthening (2) jak-aa 'bag' jak-un-k-unaa --> 'bags' hak-ii 'grass' *hak-uw-k-uwaa --> --> hak-uu-k-uwaa 'grasses' Though Newman has given credit to Gouffe for having provided an extensive and accurate description of the general process of reduplication in Hausa, he also criticized Gouffe's work for having failed to make some generalization about the reduplication process involved. Furthermore, Newman argues against CVC-prefix of reduplication postulated by Gouffe's for nominals, and suggests that synchronically what reduplicates in Hausa reduplicated nouns is disyllabic suffix which is a copy of the two rightmost syllables of the root. Following the affixation directionality postulated in Wilbur (1973) and Marantz (1982) Newman argues that the suffixation of the reduplicative affix to the stem directionality is from right-to-left, illustrated as below: 4 (3) *biri + Copy 2$ --> biri-biri birbirii --> 'fruit pigeon' daaguraa + Copy 2$ --> daaguraa-guraa --> daagurguraa 'pl. act. gnawing' In (3) above, the vowel of birbirii 'fruit pigeon' is lengthened because synchronically Hausa has the tendency to lengthen nominal final vowels. In Newman's proposal proceeds by reduplication reduplicating the last two full syllables of the stem and after that a regular rule of syncope applies to delete the stem-final vowel when it is in final position in the word. Davis (1988) proposed an analysis of nominal plurals reduplication in Hausa in Marantz's CV-template approach to Davis argues reduplication. that Hausa reduplicative the an interfix --a type of empty morpheme: usually template is beginning with a vowel followed by one or two consonant (i.e, a rhyme plus an onset) which does not add new meaning to the word and which is inserted between the (nominal) stem and (plural) the In Marantz framework reduplication suffix. is viewed as a normal affixation process whereby a phonemic empty reduplicative CV-template is specified as an affix and it acquires its melodic content from the stem through a phoneme copying process. The phonemes then associate one-to-one to the CV-template. The following example illustrates the account by Davis of the reduplication in which there is autosegmental spreading in addition to the phoneme copying process: (4) b a k u n a b a k u n a Interfixing I A 1 A I 1 A I I A 1 I cvv c v c vv c vvc + vcc + v c vv --> 5 b a k u n a spreading Autosegmental --> 11 A 1 A 1%%, + vcvv cvv c + u n a b a k Phoneme Copying 1 A IN ---> 11 A cvv c + vec + vcvv bakuna u n a b a k Association ., Right-to-left I A I 11 A --> cvv C + VCC + V C VV bald . (4) shows that [bakunkunaa] derived from [baakunaa) by is interfix -unk- between the stem the inserting reduplicative and the plural suffix -unaa. The interfix acquires its phonemic content autosegmental by spreading of the final stem consonant and copying of the phonemes of the plural suffix before the final vowel with a right-to-left association. 3. Hausa Nominal Reduplication: Analysis within McCarthy and Prince's Framework In the following sections I analyze the set of data provided (1986) by Newman regarding the monomorphemic reduplicated nouns well as the as data on nominal reduplicative plural given in Davis (1988) within McCarthy and Prince's (1986, 1988 and 1990) theory of prosodic morphology. The main claim of McCarthy and Prince's theory regarding the reduplication process that is reduplicative a template is analyzable as some sort of invariant prosodic unit which must be structurally licit prosodic constituent of the language a under consideration. They posit that a prosodic constituent can either be a prosodic word (Wd) which is the minimal word in the language, a foot (F) which consists of up to two syllables; a light (monomoraic) syllable (p.) which is a syllable with a short U 6 vowel; a heavy bimoraic OW that is a syllable with a long vowel or ending with a consonant ; and a core syllable which is a light syllable in which only one consonant can precede the According to McCarthy and prince the reduplicative- vowel. template gains its melodic content by a process which will first copy the phonemic content of the stem and then associate one- copied phonemes to the prosodic template. The to-one P.- 3 direction of the mapping is left-to-right for prefixation, right- An example to-left for suffixation, and variable for infixation. illustrating how the theory works is shown below for Mokilese in which progressive aspect reduplication involves prefixation Consider the sample of an invariant heavy syllable template. data below from the Mokilese language taken from McCarthy and Prince (1988: 21): 'plant' (5) pOd-pOdok pOdok 'weave' paa-pa p a 'find' dii-di.ar di.ar 'ben' caa-caak caak a a a+ a+ a a a pOdok --> a+ a (6) /I /I\ /I\ /I\ /I\ I\ I\ pOdok p0 dok pOdok p0 dok p0 dok d-Association c-phoneme b-prefixation a-/pOdok/ copying =[pOdpOdok] a+ a a+ pa --> a+ a a (7) /\ / I \ I\ I\ p a pa paa pa pa d-Association b-prefixation c-phoneme a-/pa/ copying [paapa] 7 7 a+ 00 a+ 00 di.ar --> a + as (8) \ /I I\ I\ /I I\ diar diar di a r d is r di i b-prefixation a-/di.ar/ d-Association c-phoneme copying a+ 00 a+ 00 caak --> a + a a (9) /1 \ /I I\ I\ I\ ca ak caak caak ca ak caa a-/caak/ b-prefixation d-Association c-phoneme copying [caacak] McCarthy and (1986) show Prince that the of template reduplication in Molikese is a bimoraic syllable (apt). In (7-8) they argue that in order for the reduplicative target to be satisfied --to be bimoraic -- the vowel of the first syllable of the word must be spread to link with the second mora of the prosodic affix as shown in (7d) and (8d). In addition, there are some requirements that the heavy syllable affix be either a CVC or a CV1V2 (where V1=V2). Thus the second vowel of the word di.ar cannot be a constituent of the syllabic prefix in that Mokilese. (9) shows superheavy syllable cannot a represent the syllabic prefix. Thus, the coda consonant of the stem, /k/, is not a constituent of the template affix. Now that I briefly summarized McCarthy and Prince's approach the reduplication process to turn I next its to application to reduplicative processes in Hausa. I argue here of the that the target in Hausa prosodic invariant affix reduplicated nouns must be a foot. Furthermore, agreeing with Newman (1986) I claim that the reduplication process involves suffixation rather than prefixation or infixation of a foot-affix to the nominal stem. Phonemic melody association from the base to the affix is directionally from right-to-left (Marantz in my analysis the nominal stem to which Finally, 1982). reduplication applies includes the stem final vowel. 3 8. The sample of data below for nominal reduplication in Data: Hausa comes from Gouffe and Davis Newman (1975), (1986) (10) represents the set of data which Newman refers (1988). to as the synchronically frozen reduplicated nouns, and (11) the nominal plural reduplication discussed in Goufft and Davis. (10) reduplicated nouns (Newman 1986) Synchronically frozen 'thin metal bracelet' warwar6o 'butterfly' bilbfleo zanzanda 'small pock marks' ttinttipli 'sacre while ibis' 'membrane' tantaanfi balbeelaa egret' 'cattle 'lame excuse' KyaKICeegaa 'mange of goat' gaRgadaa kwaRkwaasaa 'drive ant' 'investigation' dfddigii ktikkaida 'cracked cooking pot' Thdindinda grasshopper' 'female (11) Class 2 plurals and Davis 1988) (Gouffe 1975 baktinkdnaa 'months' ttdtindlinaa 'high ground' jakunktinaa 'bags' battitttikaa 'matter' gartiurdwia 'towns The above set of data seem to show two kinds of reduplication While in (10) it seems that the pattern in Hausa. patterns involves for the most part, the reduplication of a prefix which syllable of the initial (nominal) is stem, (11) the in the reduplicative affix looks like a heavy syllable infix. Analysis: A Prosodic Morphology, Account: I argue that the apparent varied patterns of reduplication in Hausa shown in these two sets of data above can be captured by a single 9 in McCarthy and Prince's prosodic morphology generalization is claim reduplicates what The in nominal theory. in Hausa is a suffix whose prosodic target is a foot reduplication The copied phonemes are mapped right -to -left to the 0). (5 exemplified below: nominal stem as b- suffixation (12) a- base c- phoneme copying F F A A a > a a+ as a-4- a a a a /I\ II\ I\ I\ I\ IR wa roo waroo wa roo waroo d-Stem final vowel e- coda rule deletion F F A a a+ A a a a +a a /I\ I\ I wa r waroo war waroo f- Association F A a a a + /I\ /I\ A war wa roo = [warwaroo] (13) a- Base b- Suffixation c- Phoneme copying F F aaa>aaa+aa coo+ as A A /I\ /I\ I\ I\ I\ II\ I\ I\ I\ ?adimaa ?adimaa ?adimaa ?adimaa 13