The Semantics of Relative Clause Expressions in Yorùbá

Cite this article as: Akinwande, O. (2026). The semantics of relative clause expressions in Yorùbá. Sokoto Journal of Linguistics and Communication Studies (SOJOLICS), 2(2), 105-116. https://doi.org/10.36349/sojolics.2026.v02i02.009

By

Ọlaide Akinwande

Department of Linguistics, African and Asian Studies,

University of Lagos

oaakinwande@unilag.edu.ng

Abstract

Previous studies have continued to deliberate on the syntax of relative clause expressions in Yorùbá (Bamgboṣe1990, Dechaine 1993, Dechaine 2001, Ilọri 2010, Adeṣuyan 2014, Akintoye 2017, among others), whereas little or no attention is given to the semantics of relative clause constructions in the language. Therefore, the scarcity and undetailed analysis of the semantic interpretations of relative clause constructions in Yorùbá literature necessitates this article for Yorùbá linguistic expansion. The data used for the study were drawn from spontaneous relative clause expressions in Yorùbá from 20 native speakers of the language in Abeokuta South Local Government Area of Ogun State, Nigeria. Analysis in this study is guided within the Information Structure (Chafe 1976, Prince 1981, Halliday 1994, Lambrecht 1994, Ertechik-Shir, 2007, van der Wal et al 2025). The Information Structure approach employed in this paper is contained in Alternative Semantics of Rooth (1992, 2015). The study discovers two materials (NPs) that are crucial in relative clause interpretations. These materials occur internally and externally within the scope of relative clause constructions. The paper illustratively discusses three semantic interpretations of relative clause constructions, namely; appositive, restrictive, and sortal-internal in Yorùbá. On appositive reading, it claims crucial presence of external material; both internal and external materials are essential for restrictive reading, while sortal-internal reading relies on the content of relative clause. The study relates the relative clause interpretations to one of the core aspects of semantics - presupposition.

Keywords: Alternative Semantics, Information Structure, Relative Clause, Semantics, Yorùbá

1.  Introduction

Previous studies have dwelt much on the syntax of relative clause expressions in Yorùbá (Bamgboṣe 1990, Dechaine 1993, Dechaine 2001, Ilọri 2010, Adeṣuyan 2014, Akintoye 2017, among others). There is dearth of scholarly works on the semantic related studies of the expressions. Therefore, the semantic readings of relative clause constructions in Yorùbá needed to be investigated for the logical understanding of the constructions. The paper is organised into six sections. Section two examines literature review on relative clause expressions in Standard Yorùbá and its varieties. Theoretical frameworks adopted for this study are discussed in section three. Section four discusses materials (NPs) in relative clause interpretations. Construals of Yorùbá relative clause expressionsare examined in section five. Section six is the concluding remark of the study.

2.  Literature Review

This section reviews existing works on relative clause expressions in Standard Yorùbá (SY) and its dialects. This step is necessaryto know the available studies on the relative clause expressions in SY and its varieties and the gap that thispresent study intends to fill. To start with, Yorùbá relative clauses typically begin with the introducer tí (Awobuluyi, 1978:35; Bamgboṣe, 1990:209), whereas in Mọ̀bà Yorùbá, relativisation requires the presence of focus marker ni alongside a mid-toned relative clause marker (Ajiboye, 2006:14). Ajiboye (2006) is a syntax-based study which claims that focus marker ni and relative marker ti concurrently present in Mọ̀bà relative clause (RC) expressions. Dechaine (1993) is another syntactic account ofRC in SY. It opines that the structure of relative clause constructions in Yorùbáis like the focus constructions. Dechaine (1993) recounts that like focus constructions, V/VP can be relativised in Yorùbá. Further syntactic analysis of Yorùbá RC is Dechaine (2001). The paper asserts that Yorùbá relative clause is head initial. The study claims that relativisation of both arguments and adjuncts mostly requires the relative clause marker tí following them in the language. Ilọri (2010) is also a syntactic report of the structure of relative clausein Igálà and SY. The study recounts that the structure of RC in Igálà and SY is similar. The work claims that RC marker in Igálà and SY is kì and tí respectively. Ilọri (2010) argues that in SY RC, there is a constituent which has anaphoric link with another constituent serving as its antecedent in the main clause. Ilọri’s work opines that SY relativised items originated within tí-clause before being raised to the pre-tí-clause position. Ilọri submits that Igálà RC marker kì and SY RC marker tí are strong heads that project RC structures in the two languages. Adeṣuyan (2014) also focuses on the syntax of RC in Ìlàjẹ and Ìjẹ̀bú. The work reports that jí and tí express RC in Ìlàjẹ, while kí marks RCin Ìjẹ̀bú. The study argues that the RC markers can project their phrases in the two dialects. Akintoye (2017) examines the syntax of relative cause constructions in Iyàgbà. The paper posits that a simple sentence is reduced to RC by moving the relativised itemsto the initial position of the RC expression. Akintoye reports that ghin and ń mark relative clause in Iyàgbà dialect. Awobuluyi (2021) syntactically recounts that unlike sentential nominalisation markers (pé, kí, bí) and relative clause constructions where nouns or noun phrases usually precede RC tí, there is a particular tí that does not need nouns or noun phrases preceding it, rather, a verb or verb phrase precedes it. Tí in such an expression is tagged “sequel marker”, as shown in (1).

1a. Ǹjẹ́ ó  wọlé  rí, tí kò kí ẹnìkan kan?

Q  3SG enter-house ever that NEG greet person one

‘Did he/she ever go in without greeting anyone?’

 b. Wọ́n mọ̀, tí  wọn ò  sì sọ  nǹkan  kan.

3PL know that 3PL NEG CONJ say something one

‘They knew it, but didn’t say a word about it.’

           (Awobuluyi, 2021:3)

Ọladimeji and Owoyẹle (2023) did a comparative study of relativisation in Uro and Arigidi from the syntactic point of view. They argue that when subject arguments are relativised, there is no overt relative marker, but when object arguments are relativised, relative marker i is employed immediately after the relativised object. Relative clause marker in Ọ̀họ̀rí is ìyé (Akinwande, Ilọri& Ajiboye, 2024). The authors claim that unlike Standard Yorùbá where relative clause introducer tí is optionally dropped in rare expressions (Awobuluyi, 1978:36), the introducer ìyé is overtly obligatory in Ọ̀họ̀rí relativeclauses¹. However, of all the aforementioned works, it is obvious thatthe semantic aspect of relative clause in SY and its dialects is scanty and therefore, it needs to be investigated and documented. Although, Awobuluyi (2013:82-87) briefly discussed two interpretations of relative clause: restrictive identification and referent tracking, but this present study discusses three readings of Yorùbá RC, two of which is different from what Awobuluyi discussed. Besides, Awobuluyi’s study is a descriptive-based interpretations of RC, whereas this current analysis is a theoretical-based account of Yorùbá RC readings.

3.  Theoretical Framework

This section discusses the theoretical framework adopted for the study. The paper employs information structure and alternative semantics theories. The Information Structure-cum-Alternative Semantics theory as adopted in this paper is not two theories but a single theory that effectively accounts for left periphery expressions such as relative clause constructions in Yorùbá. The framework is discussed in the following subsections.

3.1. Information Structure

Information Structure (IS henceforth) is a discourse-semantics framework that explains and models the cognitive processes involved in speech-act and how information is coded, packaged, and arranged in discourse to ensure precision in the way that the addresser organises the structure of utterances for the addressee to understand the intents/purposes and context for appropriate decoding/understanding/interpretation and feedback (Halliday, 1994; Erteschik-Shir, 2007; van der Wal, 2025). Information packaging in IS is achieved in two major ways. The first is by connecting the way the speaker/addresser encodes information in the utterance and how the addressee/listener interprets it given his/her mental state, especially how the two of them distinguish between what is already known i.e., pre-knowledge information that the two of them had and brought into the conversation generally called the GIVEN (also called topic), which continues to be updated by them as the discourse progresses; and the NEW (called comment) that is just being introduced into the discourse. The second is via the use of syntactic steps such as altering and reordering of the basic word order in the utterance to structurally construct linguistic expressions that satisfy the particular speech-act targeted in the discourse. For instance, word order ensures the GIVEN is placed at the beginning of a sentence (especially in complex sentences), while the COMMENT is relegated and inserted after it. The other two main aspects that play specific roles in IS are: first, the activation status of referents; and second the functions that referents may take within the sentence, most especially in relative clause expressions. Activation of referents in IS refers to mental representation of referents and the expressions used to refer to them. The referents are agentive arguments (such as characters in stories) that play active roles in information packaging. An agentive argument may exist in real world or imaginary world, i.e., mental representation in the mind of the speaker. The agentive argument may be a noun, pronoun or modified noun phrase (Van der Wal, 2025:7). For instance, inYorùbá relative clause, a noun head might be a referent as topic for discussion. Similarly, a pronoun might be a referent (anaphoric linking) that provides full information about the head noun. Referent activation status has influence on which expressions are used to refer to such a referent. If a referent has been previously mentioned in context, then it has been activated but if the referent has not been mentioned in discourse, it is inactive (Ariel, 2001; Gundel, 2003). The functions and assigned roles to each referent in discourse largely depend on how the information would be structured. However, in IS, if a new referent is introduced and has not been mentioned in the speech act, such a referent is inaccessible (inactive); but if the newly introduced referent has been mentioned in expressions, it becomes accessible i.e., active. Conclusively, although the precision about the way IS processing manifests may differ from one language to the other, they do not radically depart from the core/main dictates of the theory. However, one aspect of IS adopted forthis study is Alternative Semantics.

3.2. Alternative Semantics

Rooth’s (1992, 1999, 2015) Alternative Semantics (AS henceforth) is a formal semantics theoretical framework that interprets the meaningof certain natural language expressions, namely; focus, question, topic, relative, disjunction, negative polarity, etc., using the conceptual ideology that such meanings denote sets of alternative propositions rather than just single meaning of entities or referents. The framework unifies ideas on semantic, pragmatic, discourse, and structural operations toalternative phrasal meanings with a focus on how alternatives are generated and interpreted, particularly in relation to relative clause (or relative sensitive marker), multiple focus, question-answer congruence, comparatives, substitution, interpretation of topic, etc.In this theoretical approach, every single expression is associated with two semantic interpretations. The first is the ordinary semantic value, which is the conventional meaning of an expression, and the second is the core/specific value attached to the expressionin the particular construction in which it is used. It is within this core semantic value meaning that a set of likely alternatives is generated for the variable created within the context of construction/speech-act type. For example, in Mary likes Sue, the ordinary semantic value is the proposition that Mary likes Sue but Mary is the target element in the expression Mary likes Sue. The targeted semantic value would be the set of propositions where different individuals, {Esther, Joy, Rebecca, Acquilla, etc.} like Sue. Conversely, if Sue is the targeted element, the targeted semantic value would be the set of propositions where Mary likes different individuals. This distinction allows for a detailed understanding of how alternative semantics affects meaning and how listeners interpret expressions based on which elements are emphasised. The targeted semantic value changes depending on the scope of relative clause which can influence pragmatic inferences that arise from an expression, e.g., expression with relative clause can lead listeners to infer that the relativised element is contrasted with other possible alternatives. For instance, alternative propositions in the semantic value of a question are restricted to some entities while the alternatives in the relative semantic value are unrestricted. For example, alternative proposition in (2a) below is targeted at someone (who invented calculus) but there are three alternatives. It is from these alternatives that the responder picks one entity as correct answer in (2b).

2a. [Who invented calculus_i, Leibniz, John or James?]

b.  [Leibniz invented it_i].

      (Rooth, 2015:9)

This model is applied on the treatment of semantic interpretations of relative clause constructions in Yorùbá.

4.  Data and Methodology

This study employed a qualitative research approach, providing an overview of how relative clause expressions are interpreted in Yorùbá. It is assumed that the qualitative research method can effectively handle the semantics of relative clause expressions in the language. However, the data used for the study were drawn from spontaneous relative clause expressions in Yorùbá from 20 native speakers of the language in Abeokuta South Local Government Area of Ogun State, Nigeria. To achieve the quality and accuracy of the data, adult native speakers of Yorùbá, whose age range between 30 and 60 years, and who have full command of the linguistic competence of their language, were targeted during spontaneous speech form of Yorùbá relative clause expressions, using digital recording device. Thereafter, the elicited data were transliterated and appropriately glossed with the two tiers of literal/grammatical and logical translations in line with Leipzig glossing rules’standards. Thegrammatical glossing was done morpheme-by-morpheme. Orthographic conventions were followed. Tone is represented as high (´), low (`), and mid (unmarked). The data were later analysed using information structure and alternative semantics theories to uncover the semantic denotations associated with the relative clause expressions in Yorùbá.

5.  Analysis and Discussion

This sectionis devoted to the data analysis and discussion of findings in connection with the semantics of relative clause expressions in Yorùbá.

5.1. Materials in Relative Clause Interpretations

It is essential to firs tset record straight that in treating the semantics of relative clause constructions in Yorùbá, two materials are important to take into account - internal and external. These materials are discussed and illustrated in the following subsections.

5.1.1. Internal Materials

Internal material is a noun phrase (NP) that occurs within the scope of relative clause constructions. Internal material in relative clause is that head noun which occurs within the relative clause itself (Comrie, 1989:145-146; Aldridge, 2017:41). The head noun that appears inside the relative clause rather than outside is internal material (de Vries, 2018). Example of internal materials in relative clause interpretation is illustrated in (3) and (4).

3.  Ajá dúdú [ tí [Adé sọ ajá dúdú ti  jẹ ẹran]]

  Dog black REL Ade say  PERF eat meat

  ‘The black dog that Ade said has eaten the meat.’

4.  Ìwé [tí [Túndé rà ìwé ní ọjà dára]].

  Book REL Tunde buy LOC market fine

  ‘The book that Tunde bought in the market is fine.’

Fromthe example in (3), Adé ‘personal name’ and ẹran ‘meat’ are the internal materials, which show up within the domain of relative clause there, while Túndé ‘personal name’ and ọjà ‘market’ in (4) are the internal materials, which appear inside the relative clause therein. The internal materials are crucial in interpreting the relative clause in (3) and (4). In other words, if Adé ‘personal name’, ẹran ‘meat’, Túndé ‘personal name’, and ọjà ‘market’are absent in (3) and (4), the expressionsbecome impotent. The presupposition of AS in (3) and (4) is that Adé ‘personal name’, ẹran ‘meat’, Túndé ‘personal name’, and ọjà ‘market’have semantic value meaning attached to the relative clause interpretation.

5.1.2. External Materials

External material is the NP that features outside the domain of relative clause constructions. The head of noun phrase outside the relative clauseis external material (Keenan, 1985:143). External material in relative clause expression is exemplified in (5) and (6).

5.  Iyán [tí [mo jẹ iyán ní òde yẹn]] pọ̀

  Yam REL !SG eat LOC function DEM much

  ‘The pounded yam that I ate at the party/function is much.’

6.  Ìdòwú_i [tí [ó_i  ṣe ìdánwò]] yege

  Idowu REL 3SG.HTS  do exam pass

  ‘Idowu who wrote the examination passed.’

Iyán ‘pounded yam’ and Idowu ‘personal name’ are external materials in (5) and (6), which feature outside the scope of relative clauses therei.e., in the specifier position of the relative clauses. Iyán ‘pounded yam’ and Idowu ‘personal name’ are head nouns that the relative clauses modify in (5) and (6). Iyán ‘pounded yam’ and Idowu ‘personal name’ are essential for the interpretations of the relative clause constructions in (5) and (6). The IS of the expressions in (5) and (6) appears that iyán ‘pounded yam’ and Idowu ‘personal name’ have semantic value in the mind of the speaker(s) there. For instance, following AS approach, other available fooditems in (5) may be {ẹ̀bà, iyán, àmàlà ‘cassavaflower’ with soup, rice, drinks, etc.) as set of propositions of different food items, but the targeted food that carries the semantic value is iyán ‘pounded yam’. Similarly, other available people/students who wrote the examination in (6) may be {Sọlá, Dàda, Ìdòwú, Túndé, etc.} as set of propositions of different people/students who sat for the examination, butthe targeted proposition with core semantic value is Ìdòwú ‘personal name’. The illustrations in (5) and (6) presuppose that iyán ‘pounded yam’ and Ìdòwú ‘personal name’are set apart as relative clause heads.

5.2. Construals of Yorùbá Relative Clause Expressions

This section discusses the semantic interpretations of relative clause expressions in Yorùbá. Following information structure-cum-alternative semantics theory, three semantic interpretations of relative clause expressions- appositive, restrictive, and sortal-internal are discussed and illustrated in this section.

5.2.1. Appositive Reading

Appositive reading implies that a relative clause is anaphorically linked to the antecedent in the discourse position (i.e., co-index). It is crucial that appositive reading in relative clause contains an antecedent that connects it with the anaphoric item in the RC domain, otherwise, the appositive interpretation would disappear. Therefore, in appositive reading, the external material (antecedent) is more important than the relative clause. In line with the explanation, postnominal relative clause type matches the appositive reading, as illustrated in (7) and (8).

7a. Olùkọ́ rí Táyọ̀_i [tí  ó_i   ń  kàwé]

  Teacher see Tayọ REL 3SG.HTS PROG read-book

  ‘The teacher saw Tayọ who isreading.’

b.  Ṣọlá rí  Báyọ̀_i [tí  ó_i  jẹ ẹ̀bà ní ìjẹkújẹ̀ ní òde]

  Ṣọla see Bayọ REL 3SG.HTS eat ẹba LOC badly LOC function

  ‘Sola saw Bayọ who ate eba badly at a/the function/event.’

8a. Olùkọ́ na Àríkẹ́_i [tí  [ó_i   pariwo.]]

  Teacher beat Arikẹ REL 3SG.HTS kill-noise  

‘The teacher beat Arikẹ who make noise.’

b.  Dàda jẹ  ìrẹsì_i [tí  ó_i  gbóná]

  Dada eat rice REL 3SG.HTS hot

  ‘Dada ate rice that is hot.’

The IS in (7) shows that Táyọ̀ and Báyọ̀ ‘personal names’ are the external materials that are outside the scope of the relative clause constructions in (7a&b) and at the same time linking the anaphoric elements in the domain of relative clause. Similarly, the IS in (8) appears tha tÀríkẹ́‘personal name’and ìrẹsì‘rice’are the external materials in (8a&b) which are also linking the anaphoricelements in the scope of relative clauseconstructions.Therefore, these external materials in (7&8)and the anaphoric item ó are referential elements that interpret the appositive reading.The comparative aspect of alternative semantics views the external material and anaphoric element ó in (7) and (8) as overlapping tools for interpreting appositive reading inYorùbá relative clause expressions. In other words, the anaphoric item is supplying information about the external material that performed an action in (7) and (8). The presuppositional logic of AS perceives Táyọ̀ and Báyọ̀ ‘personal names’as external materials that carry semantic values in terms of the person that is reading a book in (7a) and someone who ate food badly at a function in (7b). (7a&b) presuppose a set of propositions which could be different people, {Kẹ́hìdé, Táyọ̀, Bọ́sẹ̀, Túndé, Báyọ̀, etc.} but the semantic values that link the anaphoric element ó targeted Táyọ̀ and Báyọ̀ ‘personal names’. Likewise, a set of propositions in (8a) could be {Àjíún, Àríkẹ́, Àṣàkẹ́, Àmọ̀pé, etc.) but the person that is semantic value-based is Àríkẹ́ ‘personal name’. In the same vein, Dàda ‘personal name’ could have eaten {cold rice, warm rice, stale rice, hot rice, etc.} but the semantic values in (8b) show that Dàda ‘personal name’ ate hot rice.

5.2.2. Restrictive Reading

The second interpretation attached to Yorùbá relative clause expressions is restrictive reading. This is a semantic scale, where both the internal and external materials are crucial for the interpretation of relative clause expressions, as demonstrated in (9) and (10).

9.  Owó [tí Ìyábọ̀ fi <owó> ra ata] pọ̀

  Money REL Iyabo use  buy pepper much

  ‘The money that Iyabo used to buy pepper is much.’

10. Aṣọ [tí  Kẹ́mi rà <aṣọ> ní àná]  pupa

  Cloth REL Kẹmi buy  LOC yesterday red

  ‘The cloth that Kẹmi bought yesterday is red.’

As evident in (9), the external material, owó ‘money’ and internal material, Ìyábọ̀ ‘personal name’ are important here. In the same vein, the external material, aṣọ ‘cloth’ and internal material, Kẹ́mi ‘personal name’ in (10) are equally crucial. The presuppositional sense of restriction in (9) as evident in IS of alternative semantics is that the context is not talking about any other money but the money that Ìyábọ̀ ‘personal name’ used to buy pepper for cooking soup. Any other money could be interpreted as money expended to buy {fish, snails, meat, pepper}, etc., but the context narrowed it down to only the money expended to buy pepper. Likewise, in (10), the IS narrowed it down to the red cloth that Kẹ́mi ‘personal name’ bought yesterday. The context in (10) presupposes that Kẹ́mi ‘personal name’ could have bought another cloth {yesterday, day before yesterday, last week, upper last week, last month, etc.}, and the cloth could be {white, black, purple, red, yellow, etc.} but the context in (10) limited it to the red cloth that Kẹ́mi ‘personal name’bought yesterday. However, the restrictive reading in (9) and (10) is different from appositive reading in (8b) in that thehotrice eaten in (8b) has an anaphoric item that connects them together, whereas there is no anaphoric element under restrictive reading in (9) and (10).

5.2.3. Sortal-Internal Reading

Sortal-Internalin relative clause interpretation refers to expressions, where semantic properties of a head noun are determined within the relative clause itself, rather than by external materials (Abubakari, 2019:31-35). This description implies that the content of the relative clause is more important than the external material. Therefore, the external material can be semantically realised from the content of the relative clause, as illustrated in (11) and (12).

11. Ewúrẹ́ jẹ iṣu_i  [tí  Báyọ̀ sè  iṣu_i]

  Goat eat yam  REL Bayo cook

  ‘The goat ate the yam that Bayo cooked.’

12. Ìgè gbé omi_i [tí  Ìdòwú pọn omi_i]

  Ige carry water REL Idowu fetch

  ‘Ige carried the water that Idowu fetched.’

Ewúrẹ́ ‘goat’and iṣu ‘yam’ in (11) are external materials and are not really relevant in this reading. The only internal material in the context of (11) is Báyọ̀ ‘personal name’, which is crucial to the sortal-reading. For instance, if the content of the relative clause tí Báyọ̀ sè ‘that Báyọ̀ cooked’ is interrogated (in question-answer approach), the external material² would appear. Question and answer approach is an aspect of alternative semantics of Rooth which is also applicable for analysing sortal-internal reading of relative clause expressions in Yorùbá. This is demonstrated in (13).

Question: 13a. Kí ni  ó   jẹ iṣu [tí Báyọ̀ sè]?

   What FOC 3SG.HTS eat yam REL Bayo cook

   ‘What ate the yam that Bayo cooked?’

Answer:  Ewúrẹ́ ‘goat’

Question: b. Kí ni  Báyọ̀ sè [tí  ewúrẹ́  jẹ?]

   What FOC Bayo cook REL goat eat

   ‘What did Bayo cook that the goat ate?’

Answer:  Iṣu ‘yam’

Evidently, in (13), without the content of relative clause, tí ewúrẹ́ jẹ ‘that goat ate’, which holds the internal material, the first external material in the GIVEN information in (11) cannot be realised. However, the alternative answers to the question in (13a) presuppose {dog, goat, cat, sheep, etc.} but the appropriate alternative/targeted answer, which is the external material that ate the yam in the GIVEN information in (11) is ewúrẹ́ ‘goat’. Hence, the apt alternative answer is ewúrẹ́ ‘goat’ in (13b). Likewise, if the content of the relative clause in (12) is interrogated, the external material would show up, as illustrated in (14) below.

Question: 14a. Ta ni ó  gbé omi [tí Ìdòwú pọn?]

   Who FOC 3SG.HTS carry water REL Idowu fetch

   ‘Who carried the water that Idowu fetched?’

Answer:  Ìgè‘personal name’

Question: b. Kí ni Ìdòwú pọn [tí Ìgè gbé?]

   What FOC Idowu fetch REL Ige carry

   ‘What did Idowu fetch that Ige carried?’

Answer:  Omi ‘water’

Obviously, the second external material in the GIVEN information in (12) appears when the content of the relative clause in (12) is interrogated. The alternative answers to the question in (14a) presuppose various liquid and beverages, such as drinks as available in Yorùbá setting, {ẹmu ‘palm wine’, ọtí bùrùkùtù ‘bùrùkùtù drink’ (made with millet), ọtí àgàdàǹgidí ‘àgàdàǹgidí drink’, (made with unripe plantain), omi ‘water’, etc.} but the appropriate answer in the context of (12) is omi ‘water’.

Table 1:  Summary Table of the Relative Clause Interpretations

RC Readings	External Material	Internal & External Materials	RC Content
Appositive	√	X	X
Restrictive	X	√	X
Sortal-internal	X	X	√

 

The keys to understanding the readings, i.e., X and √ as used on the table, are interpreted as follows: X indicates that the material/RC content is not crucial for RC interpretation, while √ indicates that the material/RC content is essential for that RC reading. Having discussed extensively on the semantics of RC in this study coupled with the earlier syntax-based studies, we wish to encourage scholars to shift attention to the integration of syntax, semantics, and pragmatics in the analysis of Yorùbá relative clause constructions for more gainful understanding of RC and Yorùbá grammar in general.

6.  Conclusion

This study has shown that Yorùbá relative clauses are not only syntactic constructions as most earlier studies have emphasised. By shifting attention from structure to meaning, the present study reveals dimensions of relative clause usage that syntax alone cannot explain. This semantic perspective contributes to a more complete understanding of Yorùbá grammar and highlights the need for future research that integrates syntax, semantics, and pragmatics in the analysis of relative clause constructions. In essence, this work has shown that Yorùbá relative clauses have three semantic interpretations - appositive reading, where external NP is valued over the internal NP; restrictive reading, which relies on the strength of both internal and external NPs; and sortal-internal reading, that leans on the content of the relative clause itself.

Notes

¹Akinwande, Ilọri & Ajiboye (2024) discusses typology of relative clause in Ọ̀họ̀rí Yorùbá, which are prenominal, postnominal, circumnominal, and correlative relatives.

²From the illustrative examples in (7), (8), (11), and (12), it is possible for a relative clause expression to have two external materials, while some may have one external material, depending on the IS of the relative clause expression.

References

Abubakari, H. (2019). The syntax and semantics of relative clause in Kusaal. Ghana Journal of Linguistics, 8(2), 27-62.

Adesuyan, R. A. (2014). Ìtupalẹ̀ àwọn wúnrẹ̀n onítumọ̀-gírámà nínú ẹ̀ka-èdè Ìlàjẹ àti Ìjẹ̀bú. PhD Dissertation, Obafemi Awolowo University, Ile-Ife.

Ajiboye, O. (2006). A morpho-syntactic account of niin Mọ̀bà Yorùbá. Journal of West African Languages, XIII 2.

Akintoye, O. S. (2017). Relative and conditional clause constructions in Ìyàgbà dialect. Journal of Languages and Education, 3(2), 15-28.

Aldridge, E. (2017). Internally and externally headed relative clauses in Tagalog. Glossa: A Journal of General Linguistics, 2(1), 1-33.

Akinwande, O., Ilọri, J. F. & Ajiboye, O. (2024). Relative clause constructions in Ọ̀họ̀rí Yorùbá. Language Review: An International Journal of Linguistics, 3(1), 1-17.

Awobuluyi, O. (1978). Essentials of Yoruba grammar. Oxford University Press, Ibadan.

Awobuluyi, O. (2013). Ẹ̀kọ́ gírámà èdè Yorùbá. Atman Limited, Osogbo.

Awobuluyi, O. (2021). Verb relativization revisited. Ihafa Journal African Studies, 12(1), 1-10.

Bamgboṣe, A. (1990). Fọnọ́lọ́jì àti gírámà Yorùbá. University Press Plc.

Comrie, B. (1989). Language universals and linguistic typology: Syntax and morphology. Blackwell and Chicago Press.

Dechaine, R. M. (1993). Serial verb constructions. In J. Jacobs, A. von Stechow, W. Sternefeld, & T. Vennemann (eds.), Syntax: An international handbook of contemporary research, 799-825.

Dechaine, R. M. (2001). On the left edge of Yorùbá complements. Lingua, 111, 81-130.

Dechaine, R. M. (2002). Decomposing focus: Evidence from Yorùbá. In Jones (ed.), Papers in Linguistics, 46(8), 143-160.

De Vries, M. (2018). Relative clauses in syntax. Oxford Research Encyclopedia of Linguistics.

Erteschik-Shir, N. (2007). Information structure: The syntax-discourse interface. Oxford University Press, New York.

Halliday, M. A. K. (1967). Notes on transitivity and theme in English. Cambridge, MIT Press.

Halliday, M. A. K. (1994). An Introduction to functional grammar. London, Educational Arnold.

Ilọri, J. F. (2010). Nominal constructions in Igálà and Yorùbá.PhD Thesis, Adekunle Ajasin University.

Ọladimeji, O. A. & Owoyele, O. J. (2023). Relativization in Uro and Arigidi: A government and binding theory approach. International Journal of Social Science, Humanity and Management Research, 2, 68-77.

Rooth, M. (1992). A theory of focus interpretation. Natural Language Semantics, 1(1), 75-116.

Rooth, M. (1999). Association with focus or association with presupposition? In Peter Bosch and Rob Van der Sandt (eds.), Focus, 232-244.

Rooth, M. (2015). Replication data for alternative semantics. Harvard Data Verse Network.

Keenan, E. (1985). Relative clause. In: T. Shopen (ed). Language Typology and Syntactic Description, 2, 141-170.

Van der Wal, J. (2025). On the expression of information structure in Bantu.Leiden University Press, Netherlands.