Current Research in African Linguistics: Papers in Honor of Ọladele Awobuluyi Edited by Ọlanikẹ Ọla Orie, Johnson F. Ilọri and Lendzemo Constantine Yuka Current Research in African Linguistics: Papers in Honor of Ọladele Awobuluyi Edited by Ọlanikẹ Ọla Orie, Johnson F. Ilọri and Lendzemo Constantine Yuka This book first published 2015 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2015 by Ọlanikẹ Ọla Orie, Johnson F. Ilọri, Lendzemo Constantine Yuka and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-7812-X ISBN (13): 978-1-4438-7812-8 CONTENTS Profile of Ọladele Awobuluyi ..................................................................... x Foreword ................................................................................................... xii Curriculum Vitae of Ọladele Awobuluyi.................................................. xvi Introduction ............................................................................................ xxvi Part I: Applied and Socio-linguistics Chapter One ................................................................................................. 2 Bridging Interdisciplinary Gaps in the Study of Nigerian Languages Tunde Adegbọla Chapter Two .............................................................................................. 10 African Language Dilemma in African Renaissance and Development Ayọ Bamgboṣe Chapter Three ............................................................................................ 27 Orthographics as a Science: The Gbexological Approach Hounkpati B. C. Capo Chapter Four .............................................................................................. 51 Bilingualism and Language Maintenance in Small Language Communities: The Case of Gungbe Christine Iyetunde Ofulue Chapter Five .............................................................................................. 73 From Monne to Muezzin: A Translator’s Voyage to Instruction Yetunde Oluwafisan Chapter Six ................................................................................................ 85 Term Creation Methods in Yorùbá Scriptural Translation Funmi O. Olúbòḍé-Sàwè ̣ Contents viii Part II: Phonology and Morphology Chapter Seven .......................................................................................... 116 High Vowel Alternations in Igbo (Olo Ezeagu) and Yoruba Akinbiyi Akinlabi and George Iloene Chapter Eight ........................................................................................... 140 Vowel Harmony in Emergent Grammar: The Case of Yoruba Diana Archangeli and Douglas Pulleyblank Chapter Nine ............................................................................................ 178 Harmony across Borders: Trinidad Yoruba High Vowel Patterns Ọlanike Ọla Orie Chapter Ten ............................................................................................. 195 Morphological Evidence for the Computation of Numbers in Edoid, Akokoid, Akpes, Nupoid and Yoruboid Francis O. Oyebade Part III: Syntax and Semantics Chapter Eleven ........................................................................................ 220 More on the Source of the High Tone Syllable in the Infinitive Phrase of Yorùbá: Evidence from Ànàgó in Benin Republic Michael A. Abiọdun Chapter Twelve ....................................................................................... 228 Sentence-final ni in Yorùbá Olúṣèỵẹ Adeṣọlá Chapter Thirteen ...................................................................................... 251 The Morpho-Syntax of Genericity in Yorùbá O̩ládiípò ̣Ajíbóyè Chapter Fourteen ..................................................................................... 271 The Hausa Verb Structure: An Overview Ahmed Haliru Amfani Current Research in African Linguistics ix Chapter Fifteen ........................................................................................ 287 What “Spell-out” Reveals: How Niger-Congo Prosodification Constrains the Syntax-Semantics Interface Rose-Marie Déchaine Chapter Sixteen ....................................................................................... 353 Yorùbá Symmetrical Verbs’ Projections Johnson Fọlóṛunṣó ̣Ìlòṛí & Simeon O. Ọláògún Chapter Seventeen ................................................................................... 368 The DP Structure in Mǝ̀dʉ́mbὰ and the Nature of Universal Grammar Constantine Kouankem Chapter Eighteen ..................................................................................... 389 Causative Constructions in Igbò Sennen Agbo Maduabuchi Chapter Nineteen ..................................................................................... 411 Headedness and Demarcation between Nominal Compounds and Noun Phrases in Ígálà Gideon Sunday Omachonu Chapter Twenty ....................................................................................... 434 Òwe as a Property of the Yorùbá Language: A Prolegomenon Ọlásopé O. Oyèláràn Chapter Twenty One ................................................................................ 469 The Structure of the Lamnso' Proverb Lendzemo Constantine Yuka Contributors ............................................................................................. 492 CHAPTER SIXTEEN YORÙBÁ SYMMETRICAL VERBS’ PROJECTIONS JOHNSON F. ILỌRI UNIVERSITY OF LAGOS AND SIMEON OLAOGUN ADEKUNLE AJASIN UNIVERSITY The syntactic structure of Yoruba symmetrical verbs constructions is still a puzzle. The problem is that a Yoruba symmetrical verb is able to function in a pair of clauses with different syntactic structures but a somewhat unified semantic interpretation. This paper examines the syntactic and semantic structures of the projections of these verbs within the contexts of the pair of clauses in which they occur first to determine the nature of the Subject-V-Object → Object-V-Subject swap and second to account for the difference in syntactic structures and why they don’t seem to trigger salient changes in the semantic interpretations of the pair of sentences. Relying on minimalist grammar and formal semantics assumptions, we hope to show in this paper that the syntax of the resultant pair of clauses is driven by a single semantic and syntactic base where the subject is an experiencer and the V object complement a theme argument. The disparity in the surface syntax results from an internal merge operation which targets either of the two arguments with some subtle semantic implication which rests on speaker’s preference. Therefore, what looks like a swap superficially is actually the result of a syntactic movement driven by a semantic choice between the experiencer subject and the theme object complement of V such that the resultant clause’s subject position becomes open to either of them. Chapter Sixteen 354 1. Introduction Yoruba symmetrical verbs are so-called because their subject and object can be swapped or freely interchanged without any appreciable change in the logical interpretation of the pair of sentences projected by them. As a result, every Yoruba symmetrical verb is able to function in a pair of clauses with different syntactic structures but a somewhat unified semantic interpretation as exemplified in (1). 1a. Olú ń bí inú Olu prog ? stomach ‘Olú is angry.’ b. Inú ń bí Olú stomach prog ? Olu ‘Olú is angry.’ While scholars have long pointed out the fact that sentences like (1a) and (1b) are semantically and syntactically related in the language, the problem is how to account for the nature of this structural relatedness, i.e. how (1a) derives from (1b) or vice-versa, and how such related pair of sentences come to having the same logical interpretation despite the difference in the syntactic functions of the arguments involved in their projections (Awobuluyi 1978:61). The aim of this paper is to show among other things that each of the pair of clauses in Yoruba symmetrical verb projections (e.g. 1a and 1b) is independently derived from the same semantic and syntactic VP base and that it is not the case that one is derived from the other. 2. Semantics of Yoruba Symmetrical Verbs Most of the verbs in this group are experience verbs. By experience, we mean the semantic features of the verbs are such that the effect of the action/state they denote is not concrete or physically visible. Rather, such effect is only psychologically felt by the referent of the experiencer argument which they select as subject. This is evident in the fact that Yoruba symmetrical verbs are usually two-place predicates that obligatorily assign the theta role of patient/experiencer to one of their arguments, the subject to be precise. In other words, Yoruba symmetrical verbs usually have two participating arguments to which they assign theta roles, and one of the roles they constantly assign as a matter of obligation Yorùbá Symmetrical Verbs’ Projections 355 is experiencer. Using the VP-Internal-Subject hypothesis, the argument structure of these verbs can be characterized as: V{experience, theme}. This is exemplified in (2). 2. In (2), the experiencer is usually a [+HUMAN] entity whether man, a personified non-human, or a pronoun referring to any of such. The structural architectural arrangement is such that the theme argument is selected and merged with V as direct complement to derive a V-bar before the experiencer is then merged to the V-bar as subject within the resultant core VP. This fact is observable in the semantic interaction between Yoruba symmetrical verbs and their theme complements, as an appreciable number of them are idiomatically fused with their object complement in such a way that it is sometimes difficult to separate them semantically. This is evident in the word-by-word glossing and the logical interpretation of the symmetrical V-object expressions in (3). 3a. kán ojú → kánjú quick eye ‘to be in a hurry/haste’ b. je̩ ìyà → jèỵà/jìyà eat suffering ‘to suffer’ c. bí inú → bínú ?give-birth stomach ‘to be angry’ d. ta òṣì → tòṣì sell poverty ‘to be poor’ e. ṣé ̣ ìṣé ̣ → ṣíṣèẹ́ ̣ break/gather poverty ‘to be poor’ f. ti ojú → tijú ?push eye ‘to be shy’ g. kan ara → kanra Subj Experiencer V' V ti ‘push’ Obj Theme ojú eye ‘to be shy’ VP Chapter Sixteen 356 sour body ‘to be peevish’ h. s̩e àánú → ṣàánú do mercy ‘to be merciful’ i. ṣe iwèrè → ṣiwèrè/ṣièrè do madness ‘to be mad/insane’ f. póṇ (ní) ojú → póṇlójú1 ?red foc eye ‘to cause to suffer’ g. já àyà → jáyà cut/snap chest/mind ‘to be afraid’ 3. Syntax of Yoruba Symmetrical Verbs Since the grammatical function subject argument of Yoruba symmetrical verbs is not volitional agent, one may be tempted to assume that the projection of their VP does not involve the outer vP shell. As a first approximation, one can say that their inner core VP is directly merged to the Infl head2 and the experiencer subject from inside the VP is raised to Spec-IP to derive a convergent clause expression, as in (4). The justification for the raising/internal merge operation in (4) is predicated on the EPP feature requirement of the Infl head, I, which must be satisfied in a Spec-Head feature checking relationship between I and its Spec position. It would then mean that these are the structural steps/operations that derive Yoruba symmetrical verb clauses where the subject is an experiencer argument, e.g. (1a). Yorùbá Symmetrical Verbs’ Projections 357 While the analysis represented by (4) appears simple and straightforward enough in explaining the derivation of the symmetrical verb clauses of the type in (1a), the derivation of the type in (1b) where the theme argument shows up in the grammatical function subject position poses a formidable challenge. Two questions which any descriptively adequate analysis of clauses like (1b) must answer are pertinent: (i) how does the theme argument which is base generated as V complement in (2) gets to Spec-IP to become the grammatical function subject of the derived clause? And (ii) how does the experiencer argument which is base generated as subject become the grammatical function object in (1b)? Our aim in this section is to show the inadequacies of the proposed projection in (4) and use that as a springboard to present a more unified and descriptively adequate analysis of the syntax of Yoruba symmetrical verb clauses. One simple way of answering (i), in line with (4), is to say that it is the theme argument that get raised in clauses like (1b) instead of the experiencer that is raised in (1a). Such a position would produce a structure of the type in (5). The problem with the structure in (5), however, lies in the fact that it violates two major economy principles of minimal derivation namely, shortest-move and object feature-checking3 operations. The implication of this is that the analysis represented in (5) is descriptively inadequate on the one hand because it does not tell the whole story and practically misleading on the other because it violates well-formedness principles of convergent derivations, as it does not account for the accusative feature- Chapter Sixteen 358 checking of the object theme argument. Similarly, the analysis in (4) cannot be adequate because it also does not account for the accusative feature-checking of the theme argument which requires a projection above the core VP into which the V and its object will have to raise for spec-head accusative feature-checking. In other words, there is more to the syntactic derivation of clause expressions like (1a) and (1b) than the earlier proposed configurations in (4) and (5) have shown. 3.1. The Syntactic Template for Clause Projection Every syntactic derivation in minimalist grammar is assumed to begin by selecting word items from the lexicon and merging them in line with binary principle to build syntactic structures from bottom to the top of the target projection. Clause projections begin by constructing the inner core VP in line with the argument structure of a V head as indicated in (2): V merges with its subcategorized complement to project V' and the subject merges with V' to project VP in line with the VP-Internal-Subject hypothesis. We assume that this projection is the base from which Yoruba symmetrical verb clauses are derived. We observe that the two sentences in Yoruba symmetrical verb clauses, just like active and passive sentences in English, are both syntactically and semantically related in that one is usually a paraphrase of the other. In other words, for mnemonic reasons, one can say that one of them is the active form while the other is the passive4. We can differentiate between the two forms by saying that the active form contains the experiencer argument as subject while the passive form contains the same experiencer argument as the grammatical function object. So, in the pair in (6), we can also say (6a) is the active form while (6b) is its passive counterpart. 6a. Akin bí inú Akin give-birth stomach ‘Akin was angry.’ b. Inú bí Akin stomach give-birth Akin ‘Akin was angry.’ The question however is how the form in (6b) derives from its active paraphrase in (6a). As earlier proposed, we are of the opinion that the structure represented by the schema in (2) is the base VP structure from which each of the two forms is independently derived. Therefore, the issue Yorùbá Symmetrical Verbs’ Projections 359 of one clause being derived from the other does not arise. To derive the so- called active form in (6a), the syntactic steps in the projection in (7) suffice. In (7), the verb bí is merged with the theme complement inú to derive V’ after which the experiencer subject Akin is merged with the V’ to project the inner core VP which we regard as the base structure. To derive a convergent clause of the type in (6a) from this base , the two economy conditions said to have been violated in earlier structural proposals in (4) and (5) namely shortest move and feature checking must be taken into consideration. Therefore, we assume that a predicate head which projects a PredP is merged to the inner core VP to create a Spec-Head configuration for accusative feature checking between the symmetrical verb bí and its object complement inú. A light v which projects the outer vP shell is then merged to predP to create an intermediate specifier position for the cyclic movement/raising of the experience subject Akin to the specifier position of IP where it checks its nominative feature against the Infl head. This implies that the Infl head, tense to be precise in this particular case, is Chapter Sixteen 360 merged to the outer vP shell to project an I’, and for I to check its EPP feature, it attracts the subject Akin from the specifier position of the outer vP shell (i.e. Spec-vP) to Spec-IP. These steps, we opine, produce the convergent clause type in (6a). On the derivation of (6b) from the same core VP base in (2), it is evident that the same derivational steps can be assumed up to the projection of PredP. It is at that point that the difference between clauses such as (6a) and (6b) sets in. Since the logical object is to become the subject in this context, we are of the opinion that it is the object inú that raises from the object position within the inner core VP to the subject position of the outer vP shell and cyclically to the subject position in the IP. These steps are schematically represented in (8). One crucial fact that is evident in (8) is that the experiencer subject Akin remains in-situ inside the core VP as the last phonetically realized item, linearly, in this convergent derivation. The implication of this revelation is that the experiencer argument which seems to appear as the object in the phonetic spell-out of (6b) is no object afterall, but a perpetual thematic Yorùbá Symmetrical Verbs’ Projections 361 subject. The fact that the thematic subject role is preserved for the experiencer in the pair of clauses in every Yoruba symmetrical verb sentences supports this claim. The structural implication of (7) and (8) is that the Yoruba symmetrical verb clause’s subject position can be occupied by either of the grammatical function subject (e.g. Akin in 6a) or object (e.g. inú in 6b) depending on the choice of the speaker at a given time. If the VP subject (experiencer) is preferred, it would be attracted and raised to spec-IP, and if it is the object (theme), it would equally be raised to the same spec-IP position. 4. Telling Apart Symmetrical Verb Clauses from Similar Expressions There are other structurally similar constructions in Yoruba and other languages, e.g. English, which appear similar to the pair of clauses in Yoruba symmetrical verb clauses. The question is: how can we differentiate the symmetrical verb clauses from such expressions? We shall take a look at the structure of some pair of clauses that could be mistaken for symmetrical verb clauses in Yoruba on the one hand and the structure of English active/passive constructions on the other to see how to tell apart such expressions from the Yoruba symmetrical verb clauses. 4.1. Similar Expressions in Yoruba In this subsection, we shall take a look at the pairs of Yoruba clause expressions in (9) to (11) which though look similar to symmetrical verb clauses but are different expressions. 9a. Títí wù mi Titi attract 1sg-acc ‘Titi is attractive to me.’/’I like Titi.’ b. Mo wu Títí 1sg attract Titi ‘Titi is attracted to me.’/’Titi likes me,’ 10a. Òjó lu Akin Ojo beat Akin ‘Ojo beats Akin.’ Chapter Sixteen 362 b. Akin lu Òjó Akin beat Ojo ‘Akin beat Ojo’ 11a. Wó̩n s̩e ìkà fún ejò 3pl do wickedness give snake i. ‘They did wickedly to the snake.’ ii. ‘They were cruel to the snake.’ b. Ejò (náà) s̩e ìkà fún wo̩n Snake Det do wickedness give 3sg ‘The snake (too) was cruel to them.’ These pairs of clause expressions in (9-11) are similar to Yoruba symmetrical verb clauses in two major ways: first, the subject-object swapping which appears to be one of the identifying marks of symmetrical verb clauses is possible with them. This is evident in the subject-object swap in (9a & b), (10a & b), and (11a & b); second, experience verbs, e.g. wù ‘to be attracted to’ in (9a&b), are allowed in them. The implication of this is that the use of experience verbs is not exclusive to symmetrical verb clauses in the language. However, despite these similarities, the pairs of sentences in (9-11) are structurally different from symmetrical verb clauses in fundamental ways. The following are the observed differences: — Performative action verbs, e.g. lù ‘to beat’ in (10), is possible in the clause type in (9-11) but only experiencer verbs are allowed in symmetrical verbs clauses; — There is thematic role change in the subject-object swap in (9-11). For instance, the object mi ‘1sg’ performs the role of experiencer in (9a) but theme in (9b). Similarly, Títí is the theme in (9a) but the experiencer in (9b). This is in contrast to symmetrical verb clauses where subject-object swap does not bring about any thematic role change between the subject and the object arguments. For example in (12), Adé is the experiencer in both (12a) and (12b), while ikó ̣‘cough’ is the theme in both. 12a. Adé ń hú ikó̩ Ade prog dig cough ‘Ade is coughing.’ Yorùbá Symmetrical Verbs’ Projections 363 b Ikó̩ ń hú Adé cough prog dig Ade ‘Cough is disturbing Ade.’/ ‘Ade is coughing.’ This implies that the swap in symmetrical verb clauses is meaning preserving, unlike in sentences (9) – (11); — Unlike in symmetrical verb clauses, there is no mutual entailment between the pair of clauses in (9-11). For instance, using both native speaker intuition and cancellation test, it is evident that neither (9a) nor (4b) entails each other. In other words, the truth condition of (9a) excludes (9b), and vice versa. That (9a) is true does not necessarily mean that (9b) is true, and vice-versa. Ditto for (10a) and (10b), and (11a) and (11b). This fact is formally captured in the cancellation test in (13). 13a. 9A & ¬ 9B ≠ C then, 4A → 4B b. 9B & ¬ 9A ≠ C then, 4B → 4A 4.2. English Active/Passive Constructions Like English active and passive sentences, the two clauses in Yoruba symmetrical verb constructions are semantically and syntactically related in that one is a paraphrase of the other. As earlier mentioned (section 3.1.), we can differentiate between the two clauses by the fact that one (which we may call the active form just for mnemonic purposes) contains the experiencer argument as grammatical function subject while the other (designated as the passive form for the same reason) contains the same experiencer argument functioning as the grammatical object. The paraphrase relation here is no doubt connected to the fact that the subject and object arguments in these pair of clauses play the same thematic roles in both the so-called active and passive forms. This is similar to the behaviour of subject and object arguments of verbs in English active/passive constructions which as a general rule assume the same thematic roles in both active and passive forms despite the change in grammatical function and distribution. For instance in (14), the DP, the boys, performs the same agent role in the active (14a) and passive (14b) Chapter Sixteen 364 respectively despite being coded in a by-phrase in (14b). Similarly, the quantifier everything is constant in the assumed role of theme in the two clauses despite the difference in its syntactic distribution. 14a.The boys stole everything. b. Everything was stolen by the boys. (14a), without mincing words, entails (14b), and vice-versa. This somewhat parallel structural behaviour of Yoruba symmetrical verb clauses and English active/passive constructions has theoretical implications as it aligns with claims in the literature, especially Baker’s (1988) uniform theta assignment hypothesis (UTAH), that UG correlates thematic structure with syntactic structure in a uniform manner. The fact that the arguments in the two types of constructions, i.e. Yoruba symmetrical verb clauses and English active/passive constructions, maintain their semantic roles after subject-object swap so to say, is evidence that the object arguments which show up as the subject in the passive forms actually originate as V complement in the base core VP before being raised to that position through internal merge in the process of derivation. This position for which we argued on the derivation of Yoruba symmetrical verb clauses appears similar in fundamental ways to the argument advanced for the derivation of English passive constructions by English linguists, e.g. Radford (1997:183-185). Given this striking similarity between Yoruba symmetrical verb clauses and English active/passive constructions, can we then say that Yoruba symmetrical verb clauses are structural equivalents of English active and passive constructions? Do the above-mentioned similarities warrant such a claim? Our take on this is that despite the striking similarities between Yoruba and English in respect of these two types of constructions, there are salient dissimilarities between them, which are though language specific, rule out such consideration. These are -- English passive constructions are open to any lexical verb irrespective of whether it is an action or experience verb, e.g. steal in (14). By contrast, as earlier shown in this study, Yoruba symmetrical verb clauses allow experience verbs only. -- V obligatorily takes n-participle popularly called the passive participle form, e.g. stolen (in 14) and seen (as in 15), in English passive constructions but experience verbs in Yoruba symmetrical verb constructions are never inflected. Yorùbá Symmetrical Verbs’ Projections 365 15 a. Hundreds of passers-by saw the attack. b. The attack was seen by hundreds of passers-by. Radford 1997:183); -- English passives regularly take the auxiliary be (e.g. was in examples (14) and (15)) as some kind of support within Infl but Yoruba symmetrical verb constructions require no such aux element. -- English passives optionally contain a by-phrase in which the agent subject, if mentioned, shows up as direct object complement of by. However, no such structural requirement exists in Yoruba symmetrical verb constructions. Given the veracity of these observations, we are of the opinion that the two types of constructions are more likely to be differing manifestations of a single UG syntactic/semantic phenomenon. 5. Conclusion In this paper, we have provided a solution to the long-standing grammatical riddle posed by Yoruba symmetrical verb constructions. We assume a single syntactic and semantic base for the pair of projected clauses and reject the consideration that either of the two forms is derived from the other. We proposed that the object complement of symmetrical verbs in the so-called active form of the clause, which superficially becomes the grammatical function subject in the so-called passive form, originates as direct theme object complement of V before being cyclically raised through spec-PredP and spec-vP to spec-IP to derive the clause form in which the theme argument functions as the grammatical subject. For the other form in which the experiencer argument functions as the grammatical subject (i.e. the so-called active form of the clause, e.g. (1a)), we proposed that the experiencer originates in spec-VP as VP-internal- subject before being raised cyclically through spec-vP to spec-IP to become the subject of the derived convergent clause expression. We also showed that the semantic-syntactic base of this pair of clauses and the derivational steps proposed for them align with the UG principle that correlates thematic structure with syntactic structure in a uniform fashion, which Baker (1988) called the uniform theta assignment hypothesis (UTAH). Chapter Sixteen 366 Notes 1. This is ambiguous as it can also mean ‘to have a physically red eye-ball’ probably as a result of anger. 2. Note that the Infl head in this type of construction in Yoruba could be tense (T) or aspect (Asp). 3. Shortest-move is an economy principle of minimalist grammar requiring that a constituent should move the shortest distance possible in any single movement operation (Radford (1997:271). Similarly, feature checking is one of the pillars of convergent derivation. Minimalist program assumes words carry grammatical features which must be checked in a Spec-Head configuration in the course of any derivation. Any word whose features fail to check properly would cause the resultant derivation to crash. In the configuration in (4), the accusative feature of the theme argument is not checked. 4. Note that this study is not claiming that there are passive constructions in Yoruba. The use of the terms active and passive in this paper is purely for mnemonic purposes. References Awobuluyi, O. 1978. Essentials of Yoruba Grammar. Ibadan: Oxford University Press. Baker, M. 1988. Incorporation. Chicago: University of Chicago Press. Chomsky, N. 1995. The Minimalist Program. Cambridge, Mass.: MIT Press. —. 1998. ‘The Minimalist Inquiries: the Framework’. MIT Occasional Papers in Linguistics, No 15. —. 2002. On Nature and Language. Cambridge: Cambridge University Press. Collins, C. 2011. Introduction to Minimalist Syntax. African Linguistics School Handout, Porto-Novo. Haegeman, L. 2006. Thinking Syntactically: A Guide to Argumentation and Analysis. Malden, Mass.: Blackwell Publishers Limited. Heim, I. & A. Kratzer 2000. Semantics in Generative Grammar. Malden, Mass.: Blackwell Publishers Limited. Koopman, H. 1984. The Syntax of Verbs. Dordrecht: Foris. Koopman, H. & D. Sportiche 1991. ‘The Position of Subjects’ McCloskey (ed.) The Syntax of Verb-Initial Languages. North Holland: Elsevier. Palmer, F. R. 1976. Semantics (Second Edition). Cambridge: Cambridge University Press Partee, B., Meulen, A., & R. E. Wall .1990. Mathematical Methods in Linguistics. Dordrecht/Boston/London: Kluver Academic Publishing,. Yorùbá Symmetrical Verbs’ Projections 367 Pylkkanen L. 2008. Introducing Arguments. Linguistics Inquiry Monograph 49. Cambridge, Mass.: MIT Press. Radford, A. 1997. Syntax: A Minimalist Introduction. Cambridge: Cambridge University Press. —. 2004. English Syntax: An Introduction. Cambridge: Cambridge University Press. Yusuf, O. 1997. Transformational Generative Grammar: An Introduction. Ijebu-ode: Shebiotimo Publications. Zimmermann, T. E. & W. Sternefeld 2013. Introduction to Semantics: An Essential Guide to the Composition of Meaning.Berlin/Boston: De Gruyter Mauton. `