A Descriptive and Syntactic Analysis of Àhàn Pronouns

This paper examines pronouns in Àhàn, a language spoken in the Southwest Nigeria, specifically in Òmùò-Èkìtì, Ekiti East Local Government Area. In this language, the tone born by a pronoun is conditioned by the environment where it occurs. In other words, a first person pronoun subject can be a high, mid or a low tone. We also observe that in this language, pronoun can change its morphological form when it occurs in a particular syntactic environment as against the other. We shall also show, in the paper that pronouns in Àhàn inflect for tense.


Introduction
Àhàn is a language spoken by about 2000 inhabitants in Omuo-Obadore Ekiti in Ekiti East Local Government Area of Ekiti State in the Southwest of Nigeria.Though, situated within the Ekiti dialect extraction of Yoruba speaking language area , Àhàn is a distinct language from Yorùbá.Linguists have still not been able to find the proper classification for the language.However, Àkànbí (2014) following Elugbe (2012) classifies it as a branch of Ahanoid under Defoid sub-group.Earlier, Williamson (1989: p. 201), Capo (1989: p. 281), andCrozier andBlench (1992: p. 13) put Àhàn under the Akokoid branch of Defoid family group .Bennet and Sterk (1977) also classified Àhàn as one of Akokoidlects.Elugbe (1989) grouped Àhàn under Edoid sub-family.We believe that as more works are done on the language, the proper classification will be arrived at.
All languages in the world appear to have pronouns.This is so because pronouns are words used instead of a noun.Pronouns come in different forms depending on the syntactic position they appear in the sentence.Apart from the appearance of pronouns in different forms, pronouns also inflect for number.This means that there is first person, second person and third person singular and plural pronoun subject and object.Scholars, at one time or the other have discussed pronouns in different languages of the world.In the literature, pronouns are grouped under functional category.The focus of this paper is to examine the pronouns of Àhàn.We are going to attempt both its descriptive and syntactic analysis.The paper is divided into three sections.Section one is the introduction.Section two looks at the pronouns of Àhàn.Section three will be the conclusion.

The Pronouns of Àhàn
Just like in other languages, the following pronoun types are attested in Àhàn. 1 i.Personal pronouns ii.Anaphoric pronouns iii.Relative pronouns iv.Interrogative pronouns In this present paper, our focus will be on the first two of these pronouns i.e. personal pronouns and anaphoric pronouns while the remaining two-Relative pronouns and Interrogative pronouns-will be the subject of other paper in some other time.
In the literature, scholars have given two different names to pronouns found in the languages of the world; short pronouns and long pronouns.However, in recent grammatical theories, these names have been changed to weak pronouns and strong pronouns.Today, short/weak pronouns and long/strong pronouns are also called antilogophor and logophor respectively; (See Ajíbóyè (2004), Adéṣọlá (2004Adéṣọlá ( , 2005)), Manfredi (1987), andSells (1987)).We shall look at the personal pronouns in 2.1.

Personal Pronouns
In line with the vast majority of pronominal paradigms in the world's languages, the traditional singular categories-speaker, addressee and other, are coded by three different morphemes in Àhàn.These three coded morphemes are known as first person, second person and third person respectively.The tables in (2a and 2b) below show the various forms of Àhàn pronouns both in the subject and object positions.Pronouns in Àhàn do not inflect for features such as gender or animateness, which, from our observation, is a phenomenon common to virtually all Benue-Congo languages.However, pronouns in the language have morphological inflection.This makes the pronouns in Àhàn to be n different forms for case endings and this is dependent on the syntactic position each of these pronouns occupies in a sentence.As will be shown from the various examples we shall present, the morphological form that these pronouns take either in the subject or object position is always different.We observe that the third person singular pronoun object in Àhàn is phonetically zero.In the following examples , the various pronouns identified in Àhàn which are listed in (2) above are used sententially.
( The various pronouns in the subject position in Àhàn are used in (3 -5) as seen from the examples above.First, second and third person pronouns are used in their singular and plural forms at the subject position.A close observation of (3b) and (5a) shows that the form of these pronouns is the same, even together with their tone2 .The difference between these two pronouns, i.e. 1pl subject and 3sg subject can only be interpreted in context.However, some speakers of Àhàn language pronounce the 3sg as /xa/.But this is limited to just few old people in the community.We assume that one of the forms must be the original.Probably synchronically, consonant [x] of the pronoun got deleted and leaving only the vowel.Another reason may be due to the fact that articulatory organs normally want to go for the easiest pronunciation whenever there is alternative.We presume that this is what Awóbùlúyì (1967: p. 3) calls "economy of effort".And since it is easier for the articulatory organs to pronounce the vowel [a] rather than the velar fricative [x], it is easier to drop [x].In the same vein, the 2sg pronoun object and 2pl pronoun object also have the same form morphologically.This is evident in (46) below.The difference can only be realised through context.
We present, in the data below, the usages of these pronouns in the object position.As we have noted earlier , the 3sg pronoun object in Àhàn is phonetically zero .This is shown in (8) above where the object is not phonetically realised.We assume that the reason for this dropping is to disambiguate third person singular pronoun in the object position from that of first person plural pronoun in the subject position (see 6b).Though, the verb used in ( 6) to ( 7) above is inherently low tone verbs, it can then be presumed that probably the dropping of the 3sg pronoun is peculiar to the environment of low tone verbs.But this is not the case.In ( 9) to ( 11) below, the verb with mid-tone is used, yet, the object still drops.
( Still on this lack of phonetic content of 3sg at object position, we argue that the phenomenon has to do with language internal evidence.But before we return to further explanation on these examples, we shall give more data in order to make our general comments on them.The data in (12 -13) show the various occurrences of the personal pronouns in Àhan at the subject position .It will be observed that the same pronoun is realized in different forms, even in the same syntactic position.For instance, first person singular pronoun is realized as mí in (13a) and mà in (13b).There are two things to note.One, there is a change of tone and two, there is a change of form (see 13).But there is only a change of tone in (12); the form does not change.We posit that the change in form has to do with the type of tense of the following verb.What we have in (13a) is perfective aspect and that in (13b) is past/present tense.Therefore, we posit that pronouns in Àhan inflect for tense.Again, if we compare (12a) and (12b), even though, the segmental part of the first person plural pronoun does not change, there is a tone change.The change of tone on this pronoun is due to the type of verb that follows it.In (12a), we have stative verb fẹ̀yẹ "love" and in (12b) we have aktionsart verbthèrù "eat".Piernezi and Verze (2000: p. 15) say of aktionsart verb as representing …ways in which languages systematically divide eventualities into categories that are crucial to the meaning of verbs, verb phrases and sentences.
As we have observed in the data presented, the first person plural pronoun subject and the third person singular pronoun in the subject position have the same form.The form for these two pronouns is [a] with a tone that conforms to the tone of the immediately following verb.The only way to disambiguate these pronouns is through context.However, as it has been shown, when the third person singular pronoun occurs in the object position, it normally does not have phonetic content.We give ( 14) and ( 15 The tone on the morpheme of the pronoun is not fixed.The tone on the morphemes of both the 1pl and the 3sg pronouns varies.The variation depends on the contiguous item or the tense of the verb.For instance, (15b) takes a low tone instead of the mid tone which is the default tone on all the pronouns.The tone on the pronoun in ( 15b) is low because of the negative morpheme that has a low tone.Besides, this low tone marks the past tense on the verb.Further examples are given to buttress this point and to show that the following verb together with its tense dictates the tone of the preceding pronoun.As shown in ( 17), all the logophoric pronouns in Àhànhave mid tone .Each of the pronouns in ( 17) is used in the sentences in (18 -23) both at the subject and object positions.As can be seen from the data in (18 -23), there is no noticeable difference between the forms of these logophoric pronouns both in the subject and object positions.Only that the third person logophor in the object position, as seen in ( 22b) has just un as opposed to un un in (22a).We propose that the reason for the change is as a result of the serial verb that we have in ( 22).In the absence of the serial verb, the 3sg logophor reverts to unun as seen in (23c) below.
One other thing we observe on the issue of the third person pronoun in this language is the eratic nature of it.This may not be too surprising because even in English language, the third person pronoun behaves in a way different from all other pronouns in the language.This same phenomenon is observed in Yorùbá where the third person pronoun is not consistent in form like the other pronouns in the language.For instance, it is always phonetically zero in the environment before negation and future aspect in the Yorùbá language as shown in ( 24) and ( 25) below.The behaviour of this pronoun in the environment like this makes some scholars like Stahlke (1974), Awóbùlúyì (2001Awóbùlúyì ( , 2006)), Abíọ́dún (2007)

The Anaphoric Pronouns in Àhàn
Anaphor is a kind of reflexivization which is a phenomenon that is present in all the languages of the word.Cornish (1986: p. 1) describes anaphor as …a very simple and straightforward phenomenon, consisting of the avoidance of redundancy or repetition by the use of a semantically (and, optionally, lexically and phonologically) attenuated expression in place of the full, lexical expression initially used.By virtue of its pairing with the latter as "antecedent", the attenuated expression (the anaphor) repeats the reference, or the sense, which it has already established.
The simple interpretation one can give to the quotation above is that anaphora prevents repetition of the already mentioned item in one part of the structure in another part of the same structure.
In the literature, anaphor is divided into two types namely reflexives and reciprocals.We observe that reciprocals in Àhàn are not distinct from reflexives .Therefore, we shall limit ourselves to reflexive anaphor in this section.
These two forms of anaphora, i.e. reflexives and reciprocals, have no clear cut distinction in many of the African languages (Zeller, 2011).This fact has earlier been noted by Ajíbóyè and Amorskaite (2005: p. 1) in the Yorùbá language that …correspondence between form and meaning in the realm of reflexive and reciprocal construal 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…Thus, in Yorùbá, one and the same phrase is associated with three different interpretations.
It is observed by many linguists that many of the Benue-Congo languages of Bantu extraction fall within the type of languages where there is virtually no way of differentiating reflexives from reciprocals.While some languages, like English, have both types distinctly clear, a language like Àhàn has no clear cut demarcation b etween the two types.The English examples in ( 26) are regarded as anaphors while those in ( 27 The italicised words in (26 a-f) are known as reflexives, while those italicised in (27a) and ( 27b) are recipro-cals.These two forms, in the literature, are called anaphors.As can be observed in the data, reflexive anaphors must have the same φ-Features (Phi-Features) with their antecedents.This is to say that they must agree in person, number and gender (for languages that manifest gender) with their antecedents.Apart from this, anaphors must be bound within their governing categories.Haegeman (1991: p. 192) puts this requirement in this form: The reflexive and its antecedent must agree with respect to the nominal features of person, gender and number.Lack of agreement leads to ungrammaticality.
Going by the quotation above, the structures in ( 28) are ungrammatical in the English language.28 (a) *I hurt himself (b) *They hurt yourself (c) *He hurt herself (28a) is ungrammatical because of lack of agreement in person; (28b) is ungrammatical for lack of person and number correspondence while (28c) is ungrammatical because there is no gender agreement between the reflexive and its antecedent.
In Principles and Parameters Theory (PPT), the Binding Theory (BT) distinguishes pronouns from anaphors through what it called binding principles.In the BT, three principles, A, B and C are posited to be satisfied by each of the nominal elements.These principles are stated in (29).
29 [A] An anaphor is bound in its governing category [B] A pronoun is free in its governing category [C] An R-Expression is free everywhere (Chomsky 1981: p. 188) Haegeman (1991: p. 229) defines governing category thus: The governing category for A is the minimal domain containing it, its governor and an accessible subject/ SUBJECT.
While she (Haegeman) explains accessible subject/SUBJECT in this way: A is an accessible subject/SUBJECT for B if the co-indexation of A and B does not violate any grammatical principles.
In the various literatures, nominal elements are distinguished in terms of feature specifications in the ways specified in (30).(cf.Chomsky (1981), Haegeman (1991), Carnie (2001), Kim and Sells (2007)), etc.We have gone this far in explaining the property of the anaphors so as to have a better understanding of our analysis of Àhàn reflexives .The next section discusses reflexive pronouns in Àhàn .

Reflexive Pronouns and Other Reflexive Expressions in Àhàn
Àhàn language manifests some reflexive expressions that can be used to indicate that a semantic or a syntactic argument of a predicate is co-referent with another argument of that predicate, typically, the subject.(31 -33), one will observe that the reflexive pronoun in Àhàn comprises of a compound word made up of two morphemes ọla-x.This phenomenon is not peculiar to Àhàn .Even, English language that is genetically unrelated to Àh àn uses a kind of compound word (x-self) for its reflexives (see 26 above).In Yorùbá language, the reflexive anaphor is ara-x.Atóyèbí (2011: p. 3) notes this in Yorùbá and says that : Typologically, Yorùbá belongs to the group of languages in which the lexical source of the reflexive is contained in the expression of a body part, i.e., the word for "body" ara, which combines with a possessive pronoun.By reason of the fact that the reflexive occurs in a possessive frame, the final vowel of ara is always lengthened to reflect possession, hence, araax.
Therefore, the phenomenon of having words corresponding to self to form reflexives is pervasive in world languages.Van den Berg (2007: p. 35) reports that: Creole reflexives derived from the noun for "body" or other noun denoting body parts are often associated with substrate influence in the literature, even though the claim for substrate influence as the source for the Creole body-part reflexives is not particularly strong.
Àhàn uses the same form to express both the reciprocal and reflexive.In other words, there is a kind of reflexive-reciprocal polysemy in the language.This is why there is a high degree of ambiguity in the reflexives in the language.Put in another way; reflexives and reciprocals in Àhàn are functionally similar and tend to behave syntactically alike.We presume that this may be the reason why both reciprocals and reflexives are treated on a par both being anaphors subject to the same binding principle "A" in BT of PPT. In (34)  Àhàn falls within the use of body -noun + possessive attribute.On the reciprocal reflexive ambiguity, many other languages also do not make a clear cut demarcation.Heine and Miyashita (ibid.)cite examples of some world languages where there is no clear cut demarcation between the reciprocal and reflexive.Lango and Luo, two West Nilotic languages of Nilo-Saharan language family spoken in Kenya, use the same morpheme for both the reciprocal and reflexive.( 39) and (40) below, taken from Hein and Miyashita (2008: p. 191) 39) is historically derived from a noun meaning body, and that this marker is used both for reciprocal and reflexive meaning.
As has been said, anaphors are handled in Binding sub-theory of GB syntax.Words like ọla-mi "myself", ọla-rin "himself", ọla-rẹ́n "yourself", etc. are known as reflexives in Àhàn.Such items cannot be used to refer to some entity outside the structure within which they occur.They must be bound by some other entity in the same structure.Where this does not happen, such a sentence will be ungrammatical.We illustrate our explanation through the examples in (41).41 (a) Mà i yẹ̀ ọla-mi i 1sg see body I saw myself (b) *ŋgháiyẹọla-ŋwan i 2sg see body them *You saw themselves The sentence in (41a) is grammatical because the φ-Features of both the anaphor and its antecedent agree.But (41b) is ungrammatical because the φ-Features of the antecedent and its reflexive do not agree.The antecedent ŋgha is a second person singular pronoun whereas that of the reflexive ọla-ŋwan is third person plural.
One other issue in Binding Theory is the c-command.An anaphor must be c-commanded by its antecedent.This notion is stated in (42).
42 A bound constituent must be c-commanded by an appropriate antecedent.We shall illustrate (41a) on a tree diagram in (43) to show how the rule in ( 42) is applied.43 With a closer observation on the configuration in (43) in comparison with the rule in (42) it will be observed that the antecedent and anaphor no longer c-command each other and therefore the antecedent could not bind the anaphor; yet, the sentence is grammatical.What is the solution for this?The solution is to modify the notion of government and state it in terms of m-command as stated in (44a) and (44b) below rather than c-command.(ii) β does not dominate ɑ (iii) every maximal projection dominating ɑ also dominates β and (iv) ɑ does not equal β (Hornstein et al. 2004: p. 79) The configuration in (43) above conforms to the rule in (44b).

Conclusion
We have looked at the descriptive and syntactic analysis of Àhàn pronouns in this paper.We assert that pronouns in the language are conditioned by the tone of the verb immediately following it.We also adduce that pronouns in Àhàn also inflect for tense.Examples for these are given in the body of the paper.We have also shown in the paper that the syntactic and semantic behaviour of the third person singular pronoun in the language is unpredictable.This behaviour, as explained, is not peculiar to this pronoun in Àhàn alone, it is a phenomenon that cuts across many of the languages of the world.On the issue of reciprocals, we opine that there is no clear cut demarcation between reciprocals and reflexives in Àhàn.However, we observe that pronouns in Àhàn also conform to the rules of grammar within the Government and Binding of Chomsky's (1981) framework.
12 (a) Á fẹ̀yẹTúndé 1pl.love Túndé We love Tunde (b) À thèrù a tujù 1pl.eat Prep.Farm We ate at the farm 13 (a) Mí ké ruujù 1sg Perf.go farm I have gone to the farm (b) Mà ruujù Isg.pst.go farm I went to the farm ) as further examples of the similar forms of both the 1pl and the 3sg pronouns in the subject position.14 (a) á á 1pl.come We came (b) á á 3pl.come He came 15 (a) áká yún 1sg Perf.goWe have gone (b) à à thèrù 3sg NEG.eat He did not eat Adé gbì aa nèyì yún Adé take we two go Adé took two of us/Ade went with two of us 20 ) are reciprocals.26 (a) I hurt myself (b) John hurt himself (c) Mary looked at herself in the mirror (d) We know ourselves (e) The boys did the work themselves (f) You must do the work by yourself 27 (a) We love each other (b) They helped one another to do the work 44 (a) Government ɑ governs β iff (i) ɑ m-cammands β and (ii) β m-commands ɑ (b) M-Command ɑ m-commands β iff (i) ɑ does not dominate β; and some others believe that Yorùbá does not have the third person

àtì Ṣadé i rànjè ọla-ŋwan i
below, the italicised word can be interpreted both as reflexive and reciprocal.34(a)Bọ́láHein and Miyashita (2008: p. 174) opine that languages derive their reflexives anaphors from a range of quite different forms ranging from pronouns, noun phrases and adverbial modifiers.Buttressing this point further, they present the table reproduced in (35) below to show how world languages normally derive their reflexives.