Yorùbá 3 rd person pro-forms are pro-dps - PDF

Description
Yorùbá 3 rd person pro-forms are pro-dps Oládiípò Ajíbóyè Solveiga Armoskaite University of Brisith Columbia 1. Introduction Many languages formally distinguish between reflexives and reciprocals. For

Please download to get full document.

View again

of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Government & Nonprofit

Publish on:

Views: 18 | Pages: 10

Extension: PDF | Download: 0

Share
Transcript
Yorùbá 3 rd person pro-forms are pro-dps Oládiípò Ajíbóyè Solveiga Armoskaite University of Brisith Columbia 1. Introduction Many languages formally distinguish between reflexives and reciprocals. For example, in English the reflexive construal is triggered by a so called reflexive pronoun as in (1) whereas the reciprocal construal is triggered by a so called reciprocal pronoun as in (2). Reciprocal (1) They saw each other in the mirror. Reflexive (2) They saw themselves in the mirror. Thus, in the English data in (1-2) there is a 1: 1 correspondence between form and meaning as illustrated in (3): (3) Form 1 Meaning 1 Form 2 Meaning 2 Such a 1:1 correspondence between form and meaning in the realm of reflexive and reciprocal construals is however not universally attested. In Yorùbá (like in the other languages of the Kwa family) we do not find a formal distinction between a reflexive and a reciprocal construal as indicated in (4) (Awóyalé 1986:5). Reciprocal & reflexive (4) a rí ara a won we saw body Gen us/our We saw ourselves/each other/one another. (Awóyalé 1986) Literally the phrase which triggers the reflexive and reciprocal construal means our body and in fact, this literal meaning is available as well. (5) ara a won body Gen they/their their bodies / themselves / one another / each other (Awóyalé 1986) Thus, in Yorùbá, one and the same phrase is associated with three different interpretations. Confronted with behavior of ara a won, there are at least two hypotheses available to account for the 3 way ambiguity. We could postulate 2 that we are dealing with three different forms (which just happen to sound the same) (6a). Alternatively, we could assume that there is just one form and that there is a principled reason as to why there are three meanings associated with this form as in (6b): (6) a. Form 1 Meaning 1 b. Meaning 1 Form 2 Meaning 2 Form 1 Meaning 2 Form 3 Meaning 3 Meaning 1 An approach, roughly along the lines of (6a) is suggested by Awóyalé (1986:13). He argues for the insertion of four features: [±Reflexive], [+Poss], [+Pronominal] and [±Singular] which would set apart the reflexive form from the reciprocal form. This amounts to saying that we are dealing with a difference in formal properties associated with the difference in meaning. We consider this solution undesirable because it is stipulative. Firstly, it does not explain what calls for such a feature. Secondly, it does not resolve the ambiguity between the reflexive versus reciprocal, i.e. we will also need a [± Reciprocal] feature and since the surface form for the two features remains the same the motivation for insertion of either feature becomes unclear. In this paper, we adopt the alternative in (6b) according to which we are dealing with one and the same form which allows for different interpretations. In particular, we follow the view of Déchaine & Wiltschko (2002, 2004) according to which pro-forms (including reflexives) are not primitives of natural language. Accordingly, dedicated canonical reflexives are not expected. They argue that the syntactic properties as well as their binding-theoretic behavior is derived from their categorical identity as D, φ, or N. In this paper we argue that the behavior of ara a won can be captured with the assumption that they are of category D. 2. The data in detail Coming from a perspective of a language with (apparently) dedicated reflexives, the Yorùbá 3 rd reflexives strike one as multifunctional: they can be interpreted as either reflexives or reciprocals. (7) Jímò àti Kúnlé féràn ara a won. Jimo and Kunle like body Gen 3pl Possible interpretations i) More salient 1 : [Jimo and Kunle] x,y like [themselves] x,y. Context: Jimo and Kunle are lawyers and partners in a company. They split the workload and the rewards. Overall, they run smoothly as a unit and are satisfied with what they do. 1 Preferred by the speaker without any particular context. 3 ii) [Jimo] x and [Kunle] y like [each other] x, y. Context: Jimo and Kunle are lawyers in rival companies and have never worked together as a team. However, they follow each other s career with interest and really like each other on the professional level. (8) Wón se ara a won. 3pl do body Gen 3pl Possible interpretations i) More salient: [They] x,y hurt [themselves] x,y. Context: Two kids played with fireworks. One firecracker exploded right in their arms. ii) They hurt each other. Context: Two kids played with wooden swords and managed to inflict serious injuries on each other. Furthermore, Yorùbá 3 rd reflexives appear in environments were reflexives are not expected, namely as possessors. (9a) is unacceptable in English, while the equivalent in Yorùbá is grammatical. (9) a. *John likes himself s parents. (Lebeaux 1983) b. Jímò féràn òbí ara a rè. Jimo like parent body Gen 3sg Lit. John likes parents of his body.? John likes the parents of his. In order to express the intended meaning, English makes use of a possessive pronoun in combination with own as in (10): (10) John likes his own parents. On the other hand, Yorùbá 3 rd reflexives/reciprocals sometimes fail to appear where reciprocals are expected. The use of a reciprocal in English (11a) is possible, yet its counterpart in Yorùbá (11b) is ungrammatical. (11) a. John and Mary like those pictures of each other s friends. (Lebeaux 1983) b. * Jímò àti Mary féràn fótò òré ara a won. Jimo and Mary like picture friend body Gen 3pl The reciprocal meaning intended in (11a) is encoded by means of the distributive pronominal in (12). (12) Jímò àti Mary féràn fótò òré eni kòòkan won. 4 Jimo and Mary like picture friend person distributive 3pl i) More salient: Jimo and Mary like the pictures of each other s friends (more salient) ii) Jimo and Mary like the pictures of some third party. In the remainder of the paper, we provide a formal analysis for the behavior of Yorùbá ara a won. We start by discussing the theoretical background for our analysis (section 3) and then provide the necessary evidence (section 4). 3. Theoretical assumptions According to the second hypothesis introduced in section 1, Yorùbá ara a won is not a dedicated reflexive form. This is in line with a proposal by Déchaine & Wiltschko (2002, 2004) according to which there there are no dedicated reflexives. In particular, Déchaine & Wiltschko (2002, 2004) recognize at least three proform types: pro-dp, pro-φp, pro-np as illustrated bellow. (13) a. D b. c. 2 D φ φ N My 2 2 φ N φ N self him self self The syntactic properties as well as their binding theoretic properties are a consequence of their categorical identity, i.e. each proform type is associated with a set of characteristics in the environments of predication, local and nonlocal binding as laid out in (12). (14) Reflexive proform typology D-reflexive φ-reflexive N-reflexive local binding possessor syntax reflexive=reciprocal co-argument binding predication emphatic emphatic reflexive -- non-local binding reflexive reflexive logophor long distance reflexive -- (Déchaine & Wiltschko 2004) Though Déchaine & Wiltschko 2002, 2004 show that typically proforms that are of category φ display an ambiguity between a reflexive and a reciprocal 5 construal (e.g. French se) we argue that a in Yorùbá it is the possessor syntax of ara a won which is responsible for this behavior. Thus, we claim that we are dealing with an instance of category D. In what follows, we provide evidence for this claim. 4. Yorùbá 3 rd reflexives are pro- Ds. The evidence Based on the characteristics in (14), this section provides four pieces of evidence that support the pro-d analysis of Yorùbá reflexives. i) they have DP-possessor syntax ii) they appear as epithets which are inherently definite iii) they cannot function as bound variables We go through the complex structure of Yorùbá reflexives in (4.1), possessor syntax and epithets (4.1), emphatic pronoun use (4.2), non-local binding (4.3) and evidence for an alternative form of reciprocals (5). 4.1 The elements of Yorùbá reflexive structure Yorùbá has two series of pronouns: weak and strong. Only the weak pronouns series are used to form reflexive/reciprocal forms. (15) Yorùbá pronominal paradigm Strong Pronouns Weak pronouns Case Nom/Acc Gen Nom Acc Gen 1 st sg èmi mi mo mi M+ mi 2 nd sg ìwo re o é M+ é 3 rd sg òun rè ó -- M+-- 1 st pl àwa wa a wá/wa M+ wa 2 nd pl èyin yín e yín M+ yín 3 rd pl àwon won wón won M + won (Adesola 2004) We first observe that Yorùbá reflexives are complex. 3 rd person reflexive is constructed of three elements : a noun ara body combined with the pronominal clitic rè, gives the reflexive meaning of self (Yorùbá does not mark gender), a is the genitive mid-tone vowel which copies the last vowel of ara. (16) ara a rè body Gen 3sg himself/herself (Ajíbóyè 2005) The morphological complexity of ara a rè is of course consistent with the proposed DP structure. Evidence for the complexity of this phrase stems from 6 the fact that the NP ara body can occur independently, outside of the reflexive construction. (17) Mo rí ara obìnrin arewà 1sg see body woman beautiful I saw a body of a beautiful woman. A similar example with English self is not acceptable unless restricted to very specific context as in the example bellow. (18) * I saw a self of a beautiful woman. (19) Yoga teachers often witness how students discover the forgotten inner selves. (Strang Burton, p.c.) Thus, the complex structure of Yorùbá reflexives is the first piece of evidence of their DP structure. Furthermore, we argue that we are dealing with a possessive (i.e., Genitive construction). Yorùbá 3 rd reflexives have a possessive structure, as shown below. We assume (following Ajíbóyè 2005) that genitive case in Yoruba is signaled by a copy of the preceding vowel. The same copy is found in reflexive/reciprocal phrases as well as in possessor phrases: 2 Reflexive: Possesum Gen Possessee (20) ara a rè body Gen 3sg himself/herself (Ajíbóyè 2005) Possesive: Possesum Gen Possessee (21) ìwé e rè book Gen 3sg his/her book (Ajíbóyè 2005) Note that the reflexive possessor phrases form a part of epithet in Yorùbá. reflexive [NP Gen 3sg] epithet [NP NP Gen 3sg] (22) Ó se ara a rè òdè ara a rè 3sg do body Gen 3sg idiot body Gen 3sg He hurt himself. his idiot self (Ajíbóyè 2005) 2 Note that we have to assume that the possessum moves to SpecD. This might be due to case reasons but must be left for future research. 7 An epithet (Greek and Latin epitheton; literally meaning 'imposed') is a descriptive word or phrase, e.g. in italics, Richard the Lionheart, Paul the doughnut lover and so on. Epithets are invariably definite, and so are the possessor phrases. The two facts together give us the second piece of evidence for treating Yorùbá 3 rd reflexives as pro-ds. 4.2 Yorùbá reflexives as emphatic pronouns Another fact supporting the pro-d treatment of Yorùbá reflexives is their emphatic use. Used emphatically, they induce an emphasis on the denoted individual. On the other hand, they do not depend on the verb and remain optional as any independent full nominal phrase. Thus, the combination of the two qualities being independent and optional gives another reason to group Yorùbá reflexive pronouns with pro-ds. (23) a. Èmi fún ara a mi rí Méri. 1sg prep body Gen 1sg see Mary I myself saw Mary. b. Mo rí Méri fún ara a mi. 1sg see Mary prep body Gen 1sg I saw Mary myself. 4.3 Non local interpretation of Yorùbá reflexives According to Déchaine & Wiltschko, pro-dps cannot function as bound variables. Such an interepretation is restricted to pro-φp. Thus, we predict that Yorùbá reflexives cannot be interpreted as bound variables. This prediction is borne out as indicated by data involving VP ellipsis. Yorùbá reflexives are compatible only with the strict identity reading indicating that they cannot be construed as bound variables: (24) Adé àti Kúnlé so pé fótò ara a won wà ní títà béè ni Jímò A and K say Comp photo body Gen 3pl be on selling so J àti Bósè náà so béè and B deic say so Ade and Kunle say that pictures of themselves/each other are on sale and Jimo and Bose say so too. i. *A and K say that pictures of themselves are on sale, J and B also say that pictures of A/K are on sale 8 ii. Most salient: A and K say that pictures of themselves are on sale, J and B also say that pictures of A/K are on sale too iii. A and K say that pictures of each other are on sale, J and B also say that pictures of A/K of each other are on sale iv. *A and K say that pictures of each other are on sale, J and B also say that pictures of J/B of each other are on sale This fact can be explained as follows. The interpretation of VP-ellipsis involves copying of the VP in the left conjunct to the gap in the right conjunct. Copying a VP containing a free variable, as in (25), results in a sloppy reading. (25) John hates himself and Bill does too. John λx (x hates x-self) and Bill λx (x hates x-self) sloppy reading By contrast, copying a VP with a constant in it can only yield a strict reading. (26) John hates Mary and Bill does too. John λx (x hates Mary) and Bill λx (x hates Mary) strict reading Since the sloppy reading is not acceptable in Yorùbá, we conclude that the reflexive pronominal expression can not be a variable and is a pro-d. 5. An unambiguous reciprocal form As noted in section 1, Yorùbá reflexives lack the formal distinction between reflexives and reciprocals. The same form (example (7) repeated bellow as (27) for convenience) may have both interpretations. (27) Jímò àti Kúnlé féràn ara a won. Jimo and Kunle like body Gen 3pl Possible interpretations i) More salient: [ Jimo and Kunle] x,y like [themselves] x,y. Context: Jimo and Kunle are lawyers and partners in a company. They split the workload and the rewards. Overall, they run smoothly as a unit and are satisfied with what they do. ii) [Jimo] x and [Kunle] y like [each other] x, y. 9 Context: Jimo and Kunle are lawyers in rival companies and have never worked together as a team. However, they follow each other s career with interest and really like each other on the professional level. Having said that the two interpretations are available, we observe that the reflexive interpretation is more salient, i.e. it is the default one. The reciprocal interpretation can be coerced by the context because the pronominal form ara a won body Gen 3pl is not a primitive dedicated reflexive. There is an alternative, unambiguous way to express reciprocal relationship in Yorùbá. (28) Jímò àti Mary féràn fótò òré eni kòòkan won. Jimo and Mary like picture friend person distributive 3pl i) Jimo and Mary like the pictures of each other s friends. ii) Jimo and Mary like the pictures of some third party The form of the unambiguous reciprocal is consistent with the prediction by Déchaine & Wiltschko (2004) that reciprocal forms are invariantly of category D. Crucially, the D position in the unambiguously reciprocal D phrase eni kòòkan won person distr. 3sg is occupied by a distributor rather than by the Genitive in the ara a won. (29) Reciprocal (Déchaine & Wiltschko 2004) Yorùbá D D 2 2 D φ D φ distributor 2 kòòkan 2 φ N won N variable reciprocator 3sg eni person Interestingly, in some cases where the unambiguous reciprocal eni kòòkan won is used, e.g. in (28) above, the ambiguous reflexive/reciprocal pronominal ara a won is disallowed. (30) a. John and Mary like those pictures of each other s friends. (Lebeaux 1983) b. * Jímò àti Mary féràn fótò òré ara a won. Jimo and Mary like picture friend body Gen 3pl 10 6. Conclusions and further questions We have shown that multifunctional Yorùbá reflexive ara a won is best accounted for if treated as pro-dp based on the typology of proforms (Déchaine & Wiltschko 2002, 2004). Our argumentation relies on several facts. First, the possessive structure of ara a won patterns with that of an independent noun phrase in that it can be used emphatically and play a part in formation of epithets. Second, the NP ara body can occur outside the reflexive structure, i.e. a pro-d type pronominal construction clearly complex and has an NP as a subconstituent. Third, the reflexive construal gets only the strict identity interpretation under VP ellipsis, i.e. the reflexive expression is not a variable. Fourth, the reflexive ara a won can be interpreted as reciprocal and reciprocals are of only one type: D. Although we have established that there exists an unambiguous reciprocal expression eni kòòkan won person distr. 3pl, we still need to determine the environments where eni kòòkan won rules out ara a won and vice versa. We have also left for the future research the question how do the pronominal expressions eni kòòkan won and ara a won interact with the discourse context. Lastly, we have not addressed a gap in our data: why only the weak pronoun series are used for the formation of the reflexive constructions in Yorùbá. Reference Abraham, R. C Dictionary of Modern Yorùbá London: University of London Press. Ajíbóyè, O The syntax of Yorùbá reflexive epithets. in Journal of West African Languages. Vol 31, Ajíbóyè, O Topics on Yorùbá nominals. Ms, University of British Columbia. Awóyalé, Yiwola Reflexivization in Kwa languages. In Dimmendal, G.T. Current approaches to African linguistics 3. Dordrecht: Forris. Awóbùlúyì, O Essentials of Yoruba grammar. Ibadan: Oxford University Press. Déchaine, Rose Marie, Wiltschko, Martina Dissolving condition A. GLOW talk handout, Thessaloniki. Déchaine, Rose Marie, Wiltschko, Martina Decomposing pronouns. In Linguistic Inquiry, Vol 33 (3). Lebeaux, David A distributional difference between reciprocals and reflexives. In Linguistic Inquiry, Vol 14 (4).
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks