Making it clearer: Elaboration strategies in Gĩkũyũ, a Bantu language

--------------------------------------------------------------------------------------------------------------------------Abstract


Introduction
Elaboration is one means of sequencing clauses, a means of forming clause complexes, when using language. Using logically joined sequences of clauses more in communication has been noted to enhance fluency more than using |154 simple sentences does (Srinon & White, 2007) and raises the rating of the language user (Cribb, 2004). The significance of such complexity has been observed even in academic language (Fillmore and Fillmore, 2012). There is need, therefore, to get more insight into clause complexing since it is a feature of language that fosters communicative competence.
The paper is a study on Gĩkũyũ, a Kenyan Bantu language. The language is predominantly spoken in the central region of the country but has spread to other areas (Gichohi & Subiyanto, 2022). Mandillah (2022) stresses the importance of studying native languages, noting that, for instance, they unite the native speakers and also bear the values of the speakers. Based on Halliday and Matthiessen (2014) Functional Grammar Theory, this article sought to examine clause complexing in Gĩkũyũ, specifically focusing on how the joined clauses are related. The thrust was informed by a claim of the theory that some forms of relationships bearing certain logical functions hold among joined clauses. To this end, the article intended to, one, establish the categories of elaboration in Gĩkũyũ clause complexes, and two, to identify the elaboration strategies that the categories employ. The theory refers to these relationships as functional-semantic or logico-semantic relations. The explanation is that, in general terms, the relationship could be one of expansion or projection. In the former, one of the clauses in the clause nexus logically functions to expand the meaning of the other clause while, in the latter, one clause functions to logically project the meaning of the other. (Halliday & Matthiessen, 2014). The paper limits itself to relationships in clause nexus, given that clauses are joined a pair at a time.
The specific logico-semantic relation of concern here is elaboration, one of the means of expansion posited by the theory. This paper brings to the fore the strategies that Gĩkũyũ employs in elaboration, knowledge that would be of importance not only to users of Gĩkũyũ but also linguists interested in the language and Bantu languages in general.

Literature review 2.1. Clause complexing
A clause complex is presented in Functional Grammar Theory by Halliday and Matthiessen (2014) as a sequence of clauses joined by logico-semantic relations. In a study on the usage of complex structures in language among learners of different levels, O'Hallaron (2014) observes that students in the higher levels use complex structures more than those in the lower levels do. The difference is explained by greater use of conjunctive resources by the higher levels' students. O'Hallaron (2014), thus, confirms that the language user's knowledge on conjunctive patterns of a language influences their selection of the conjunctive resources. This article identifies conjunctive resources as one of the means of achieving elaboration in Gĩkũyũ. Chege (2009) identifies conjunctions as one linguistic characteristic that contributes to cohesion in Gĩkũyũ, noting that conjunctions do so by communicating some senses that presume that other components exist in a discourse. Cribb (2004) studies clause complexing patterns employed by learners of English. The study suggests that language users with greater syntactic complexity are generally rated higher in performance and proficiency than those with lower complexity. Relatedly, Srinon and White (2007) observe that the knowledge of the conjunctive resources, and logicosemantic relations in general, can be enhanced through greater exposure to the same (and would be a mark of higher competence (Chomsky, 2015). This confirms O'Hallaron's (2014) stand since the exposure to conjunctive resources is greater in the higher schooling levels than in the lower ones. Srinon and White (2007) add that greater use of logicosemantic relations (and so clause complexes) in writing does not only extend the communication range of a writer but also increase fluency of one's language due to reduced number of clause simplexes (simple sentences) that would otherwise make the writing more disjointed.

Elaboration
Joining of clauses in writing and speaking is not haphazard but logical. Halliday and Matthiessen (2014) in Functional Grammar Theory explore different kinds of relations that logically glue clauses together. The relations, the logicosemantic or functional semantic relations, function in the logical metafunction of language. This metafunction, which comes to play in clause complexes, falls under the broader ideational metafunction (Halliday & Matthiessen, 2014). It is worth noting that the term 'clause' as used in Functional Grammar has the same meaning as the term 'sentence' as used in formal grammar. This, therefore, explains the expression 'clause complex'. Clause complexes are equivalent to the compound and complex sentences in formal grammar.
This paper focuses on elaboration in Gĩkũyũ clause complexes. Elaboration is one means of expansion, one of the two fundamental logico-semantic relations identified by Halliday and Matthiessen (Halliday & Matthiessen, 2014). Expansion presents phenomena as being of similar order or experience. This means that one of the combined clauses develops the proposition of the other. The former clause is termed as the secondary or continuing clause and the latter the primary or initiating clause. The other means of expansion are extension and enhancement. The second fundamental logico-semantic relation is projection. Both expansion and projection could be realised either paratactically or hypotactically. This paper addresses both paratactic and hypotactic elaboration in Gĩkũyũ. Paratactic and hypotactic joining of clauses has been found to vary. For instance, comparing the rates of occurrence of paratactic and hypotactic patterns in learners of English in different levels, Cribb (2004) notes that parataxis was used more than hypotaxis in lower levels while the frequency was reversed in the upper stages.

Research methodology
The data for this paper is Gĩkũyũ clause complexes, specifically, those manifesting expansion through elaboration. For a detailed description of the data, the paper, being qualitative in nature, adopts a descriptive research design. The complexes were purposively sampled, primarily from printed and oral sources. To get a variety of clause complexes, sources with prose language were purposively picked. One of the printed sources picked was a novel by Ngũgĩ wa Thiong'o, Matigari ma Njirũũngi (Remnant of the Bullet). Also picked were two Gĩkũyũ magazines, Wathiomo Mũkinyu (a True Friend) by Catholic Archdiocese of Nyeri, and Mũruna (Companion), a magazine carried in The County Voice newspaper, and particular books from Gĩkũyũ Bible for non-fictional sources. These were found to be rich in the data needed for the study, hence their choice. Two oral sources were picked, a television programme and a radio talk show. The television program was from a Gĩkũyũ television station known as Inooro T.V. The show was Reke Twaranĩrie (Let us Talk to each other). The radio program was Kĩrĩrĩmbĩ (Flame), which was from Inooro F.M., a radio station that also broadcasts in Gĩkũyũ. The two discussion shows were picked for the spontaneity of language in them.
Data collection was guided by the two modes of elaboration, paratactic and hypotactic, and the categories under each as advanced by Halliday and Matthiessen (Halliday & Matthiessen, 2014). The paratactic categories are exposition, exemplification and clarification while the hypotactic categories are finite clause and non-finite clause forms of elaboration. From both the printed and oral sources, Gĩkũyũ clause complexes portraying elaboration were purposively identified and written down. A list of four complexes from each category of sources was made under each category of elaboration. Since the study aimed at identifying the different elaboration strategies that Gĩkũyũ employs, the lists were deliberately tailored to consist of complexes portraying different strategies in each of the elaboration categories. Further, though minimally, the researchers, through native speaker competence, employed introspection but only to fill in the data gaps that the other sources left (Gahutu, 2012;Chomsky, 2015). For reliability, the data sourced through introspection was subjected to corroboration by other three native speakers of Gĩkũyũ. The data generated a total of thirteen (13) elaboration strategies, eleven (11) under paratactic and two (2) under hypotactic elaboration. Data analysis was based on these strategies. Subsequently, a consolidated list of four (4) clause complexes for each strategy was then generated, giving rise to a sample of fifty-two (52) clause complexes. For discussion and presentation, twenty (20) clause complexes were picked from the sample, each illustrating a different aspect observed in the elaboration strategies.
Thematic presentation of findings (Kombo & Tromp, 2006) was adopted. The data is presented and discussed under the categories of elaboration identified in the complexes and the elaboration strategies marking the categories.

Findings and discussion
The study established that indeed elaboration is one way through which Gĩkũyũ achieves expansion in its clause complexes. The data revealed that Gĩkũyũ elaborates through all the categories identified by Halliday and Matthiessen (2014) under the two modes of elaboration, exposition, exemplification and clarification under paratactic mode, and finite clause and non-finite clause elaboration under hypotactic mode. These are summarised in Table 1 below as adapted from the findings. The study revealed different strategies employed by Gĩkũyũ under the different categories of elaboration as discussed below.

Paratactic elaboration
Halliday and Matthiessen (2014) explain that parataxis is the form of connection that combines a pair of similar elements that are equivalent in status. For the purposes of this article, the elements are clauses. The initial clause in the pair is the initiating or the primary clause and the second one is the continuing or the secondary clause. These two could be overtly linked by explicit conjunctive expressions, the linkers, or covertly linked, with the clauses juxtaposed to each other, separated by punctuation marks such as a comma or a colon. The data manifest both strategies and are exemplified in 4.1.1 below. Paratactically, elaboration through exposition, exemplification and clarification was identified. Majorly, paratactic elaboration results in clause complexes with two clauses in apposition with overt elaboration relation markers only used rarely. However, on close comparison between the two clauses in apposition, different categories of paratactic elaboration have been observed to employ varied strategies. The specific strategies are highlighted in the next section. It is worth mentioning here that while translating the data, there are occasions in which the Gĩkũyũ-English translation is direct, with the lexemes in both languages matching. In such cases, a single translation line is given. However, some instances of such word-to-word glossing elicit English structures that are not grammatical. In these cases, a two-line glossing is provided, the first line being the word-to-word translation and the second being the fine translation.

Exposition
In exposition, the thesis in the initiating clause is paraphrased in the continuing clause, thereby emphasising it (Halliday & Matthiessen, 2014). Consider the English clause complex He never thought of achieving little things; achieving great is what was always in his mind. The idea expressed in the two clauses in the clause nexus is the same, but rendered in different words. The effect is emphasis.
The data revealed exposition in Gĩkũyũ with a number of strategies observed, the first being antonymy as illustrated in Example 1 below.
Both the initiating clause ũkũrũ ũkoora (old age disappears) and the continuing clause wĩthĩ ũkamũcoka ũthiũ (youthfulness returns on his face) carry the idea of the face of the person talked about transforming to appear younger. The transformation is carried in the initiating clause by the content words ũkũrũ (old age) and ũkoora (disappears). The idea of transformation is restated by their respective antonyms, wĩthĩ (youthfulness) and ũkamũcoka (returns), in the continuing clause. Another strategy noted was repetition. Some expressions used in the initiating clause were repeated in the continuing clause, thereby reiterating the thesis in the initiating clause. This is exemplified in Illustration 2 below; in which the expression no gũkorwo (probably) is used in both clauses. In the two clauses, the expression introduces the idea of the referents of the clauses being in a state in which they do not recognize themselves.
2. No gũkorwo matiĩyũĩ, no gũkorwo nĩmariganĩirwo maarĩ a. (Thiong'o, 1986: 34) Probably they do not know themselves; probably they have forgotten they were who. Probably they do not know themselves; probably they have forgotten who they were.
Repetition of some expressions in the process of exposition creates emphasis. There was also the use of synonymous repetition (Kemertelidze & Manjavidze, 2013). The idea was reiterated in the continuing clauses in other words although with the same meaning present in the initiating clause. Example 3 below illustrates this.
This is a Gĩkũyũ clause complex with three clauses: mwĩtĩkĩre gũkoroga thubu (agree to stimulate the soup) as the initiating clause, gũkoroga hakiri (to stimulate the mind) and gũkoroga tombo (to stimulate the brain), the two being the continuing clauses. The nominal expressions hakiri and tombo in the subsequent two clauses correspondingly are synonymous, denoting mind or brain. The same meaning is carried in the initiating clause by the word thubu (soup), which is used metaphorically in this context as an alternative of hakiri and tombo.
In the process of exposition, ellipsis was also noted in the clause complexes. According to Crystal (2011), ellipsis involves leaving out a section of the structure to achieve economy, emphasis or style, the ellipted part being recoverable from the context. This may be the case when the speaker feels that he/she may become unnecessarily repetitive when restating an idea. In Example 3 above, the agglutinated subject prefix and verb mwĩtĩkĩre (you (plural) agree) in the initiating clause is elided in both the second and the third clauses, gũkoroga hakiri (to stimulate the mind) and gũkoroga tombo (to stimulate the brain). The basic form of the two is mwĩtĩkĩre gũkoroga hakiri (you (plural) agree to stimulate the mind) and mwĩtĩkĩre gũkoroga tombo (you (plural) agree to stimulate the brain). Here, ellipsis enables the speaker to avoid redundancy, thereby achieving economy. It also creates parallelism between the second and third clauses, a matter of style.
Code mixing was identified as a factor motivating exposition in Gĩkũyũ. Code-mixing is "where a fluent bilingual talking to another fluent bilingual changes language without any change at all in the situation." (Hudson, 1996: 53). Consider Illustration 4 below: 4. Nĩ gũ'support'; nĩ kũnyitĩrĩra mũciĩ ũyũ. (Inooro T.V, 2016) It is to support; it is to support home this. It is to support; it is to support this home.
Code-mixing between Gĩkũyũ and English occurs in this clause complex in the initiating clause Nĩ gũ'support' (It is to support). The Gĩkũyũ word nyitĩrĩra in the continuing clause restates the sense of the English word support. This demonstrates that other than emphasising the proposition of the initiating clause, exposition in Gĩkũyũ could function to repeat a message previously given in another language. The final exposition strategy observed was the use of overt exposition markers. These were, however, noted to be rare, typical of paratactic elaboration as noted by Halliday and Matthiessen (2014). One such marker is atĩ (that) as exemplified in Illustration 5 below: 5. …tũkiugaga mwana ti wa mũndũ ũmwe; atĩ mwana nĩ ithaga rĩa rũrĩrĩ. (Thiong'o, 1986: 43) …we saying a child is not of person one; that a child is a treasure of the community. …we are saying a child does not belong to an individual; that a child is a treasure of the community.
Example 5 illustrates projection, a relation that projects one clause through another in way of locution or idea, indicating what people say or think. The projected element is that which is said (locution) or thought (idea). (See Halliday and Matthiessen (2014). The example carries a locution, mwana ti wa mũndũ ũmwe; atĩ mwana nĩ ithaga rĩa rũrĩrĩ (a child does not belong to an individual; that a child is a treasure of the community). It is a paratactic clause complex achieving elaboration. The clauses in the nexus carry the meaning that a child belongs to the community. Using atĩ (that) at the beginning of the continuing clause implies that the clause restates the idea in the preceding clause, mwana ti wa mũndũ ũmwe (a child does not belong to an individual). Further, exposition in Gĩkũyũ can be marked by conjunctive expressions acting as linkers. An example here is na njĩra ĩngĩ (in another way/in other words), a linker that could combine the clauses in Example 2 above as shown in Example 6 below. 6. No gũkorwo matiĩyũĩ, na njĩra ĩngĩ, no gũkorwo nĩ mariganĩirwo maarĩ a.
Probably they do not know themselves, in way another maybe they have forgotten they were who. Probably they do not know themselves, in another way/in other words maybe they have forgotten who they were. Other conjunctives applicable in place of na njĩra ĩngĩ (in another way) can be: kana (or), kana tuge (or we say), ũguo nĩ kuga (that is to say) and kuga (atĩ) (meaning (that)).

Exemplification
In exemplification, the continuing clause is more precise on the notion carried by the initial clause, often by providing a particular example (Halliday & Matthiessen, 2014). This is illustrated in the sentence: The woman's leadership earned her admiration outside Africa, for instance, many leaders from Europe showed their desire to partner with her. The continuing clause, many leaders from Europe showed their desire to partner with her, gives an example of a place outside Africa, Europe, where the woman mentioned in the initiating clause got admiration. This is the notion in the initiating clause, the woman's leadership earned her admiration outside Africa. The two clauses are explicitly linked by the expression for instance.
The data did not reveal extensive use of explicit markers of exemplification in Gĩkũyũ. The majority of clause complexes bearing exemplification had their two clauses placed in apposition, separated by punctuation marks such as semi-colon as exemplified in 7 below: 7. Thĩ ĩkĩhana ta yaninamana; mwana akĩgarũka ithe. (Thiong'o, 1986: 3) World appeared like upside down; child turned against father. The world appeared upside down; the child turned against the father.
The initiating clause thĩ ĩkĩhana ta yainamana (the world appeared upside down) implies that things went awry while mwana akĩgarũka ithe (child turned against father) exemplifies what exactly went dismally wrong. There is no overt marker introducing the continuing clause except a semi-colon separating it from the primary one. Nevertheless, some conjunctive expressions can overtly link the clauses to mark exemplification. An example is na (and) as in Example 8 below: 8. Fr. Domenico Gillio nĩ amaheire o mũndũ wĩra wake na Sr. Irene akĩheo wĩra wa gũthomithia. (Catholic Archdiocese of Nyeri, 2015: 11).
Fr. Domenico Gillio gave them each person work his/her and Sr. Irene she was given job of teaching. Fr. Domenico Gillio assigned duty to each person and Sr. Irene was assigned a teaching job.
Na (and) introduces the continuing clause Sr. Irene akĩheo wĩra wa gũthomithia (Sr. Irene was given a teaching job). The clause is an exemplification of the individuals that were assigned different roles by Fr. Domenico Gillio as mentioned in the initiating clause. In this context, it is possible to substitute na (and) with several other linkers such as ta (like), ta rĩu (like now), and kwa mũhiano/mũhano (for example).

Clarification
According to Halliday and Matthiessen, 2014, in clarification, the continuing clause explains the proposition of the initiating clause, this way, giving a clarification. Consider Example 9 below. 9. Ngarũro na Mũriũki matiamenyaga magikinya kamũkawa-inĩ: rũgano rwa Matigari rwamatwarĩte bũrũri ungĩ. (Thiong'o, 1986: 20) Ngarũro and Mũriũki were not knowing as they reached at the small hotel: the story of Matigari's had taken them to country another. Ngarũro and Mũriũki hardly realized as they reached the small hotel: Matigari's story had taken them to another country.
The initiating clause Ngarũro na Mũriũki matiamenyaga magikinya kamũkawa-inĩ (Ngarũro and Mũriũki hardly realized as they reached the small hotel) indicates that the two individuals, Ngarũro and Mũriũki, did not realize that they were reaching the hotel. The continuing clause, rũgano rwa Matigari rwamatwarĩte bũrũri ungĩ (Matigari's story had taken them to another country) clarifies why that was the case: they were engrossed in a story that another character, Matigari, was narrating to them. Clarification markers too were noted to be scarce in Gĩkũyũ. The clauses making up the complex are chiefly juxtaposed. Halliday and Mathiessen (2014) indicate that a colon (:) is applicable in writing to indicate juxtaposition. Illustration 9 above shows the form of punctuation in Gĩkũyũ. However, conjunctive linkers like na arĩ (at least), kwaria ma (surely speaking/to speak the truth/in fact) and na ma (in truth) are applicable. Example 10 below demonstrates how kwaria ma (surely speaking/to speak the truth/in fact) can be used to link the clauses in 9 above.
10. Ngarũro na Mũriũki matiamenyaga magikinya kamũkawa-inĩ: kwaria ma rũgano rwa Matigari rwamatarĩte bũrũri ungĩ. Ngarũro and Mũriũki were not knowing as they reached at the small hotel; surely speaking, the story of Matigari's had taken them to country another. Ngarũro and Mũriũki hardly realized as they reached the small hotel; surely, Matigari's story had taken them to another country.
According to Halliday and Matthiessen (2014), the clarifying clause could be an explanatory or an evaluative comment. These strategies were evidenced in Gĩkũyũ. The continuing clauses in Examples 9 and 10 are examples of explanatory comments. Example 11 below illustrates evaluative clarification. 11. Na mũcene nĩ mũrimũ… ũkanagia na andũ magĩkũra; agakũra ee mũndũ nginya mũhĩrĩga no ũmũhe mũhĩrĩga wa mũcene (Inooro TV, 2016). And rumourmongering it is a sickness…. that comes alongside people as they grow; he/she grows being a person even the clan can give him the clan of gossip.
Rumourmongering is a sickness…. that comes with age; (such that) one can be classified as a gossiper by the clan.
The continuing clause agakũra ee mũndũ nginya mũhĩrĩga no ũmũhe mũhĩrĩga wa mũcene (he/she grows as a person that the community can classify as a gossiper) is an evaluation on the sort of rumourmongers the referents of the initiating clause are. The initiating clause is na mũcene nĩ mũrimũ… ũkanagia na andũ magĩkũra (and rumourmongering is a sickness ... that comes with people as they grow). Note the change of number from plural in the initiating clause to singular in the secondary one. In the initiating clause, the speaker talks of andũ magĩkũra (as people grow). However, while elaborating in the continuing clause, the speaker instead says agakũra ee mũndũ… (he/she grows as a person…). This can be seen as a non-fluency characteristic of spontaneous language. Halliday and Matthiessen (2014) add that clarification occur via alteration of polarity, from negative to positive or the other way around. Example 12 shows the change of polarity in Gĩkũyũ, from negative in the initiating clause to positive in the secondary.
12. No ũguo ti kũiganu; ũrĩa mũgũthiĩ kũrũta wĩra nĩgũo gũkuonania nama mũrĩ arũtani akinyanĩrũ (Catholic Archdiocese of Nyeri, 2015). But that is not enough; how you will go to work is what will show truly you are teachers qualified. But that is not enough; the way you work out there is what will truly demonstrate that you are qualified teachers.
This clause nexus addresses a group of people graduating as teachers. They are informed in the initiating clause, the negative one, that their graduation will not be the key indicator of their qualification as teachers. The polarity becomes positive in the continuing clause, which clarifies to the addressees the key indicator in the view of the addressor: diligence in their teaching. Table 2 below summarises the elaboration strategies this study identified in paratactic Gĩkũyũ clause complexes.

Hypotactic elaboration
According to Halliday and Matthiessen (2014), hypotaxis is the bonding in a nexus of elements (clauses) whose status is not equal. One clause is dominant, or independent, since it has the ability of meaningfully standing alone. The other clause is dependent and relies on the dominant clause to complete its meaning. The two clauses are overtly joined by conjunctive expressions, the binders. It is the dependent clause that elaborates the thesis in the independent clause through description. The data revealed a couple of things in relation to hypotactic elaboration in Gĩkũyũ. First, the elaborating clauses are basically non-defining relative clauses. Halliday and Matthiessen (2014) call this kind of elaboration 'nonrestrictive' or 'descriptive' elaboration. Further, the description could be on the whole dominant clause or part of it. In the latter, it is a constituent of the dominant clause that is elaborated. The elaborated part, whether the whole dominant clause or part of it, is termed as as domain, a domain of the dependent clause. Both finite and non-finite forms of elaboration were identified in Gĩkũyũ clause complexes.

Finite clause elaboration
In this category, the verb in the dependent clause is finite (Halliday & Matthiessen, 2014). An example is the clause which led to his dismissal in the clause complex The administrator was rude to the clients, which led to his dismissal. The finite element in the elaborating clause is led. Both the written and the spoken sources only yielded a few finite clauses whose domain is the whole initiating clause. This is a clear indication that this kind of elaboration is not common in daily use in Gĩkũyũ. nevertheless, Example 13 below illustrates of such a situation.
It should be noted that in the example, and in all illustrations of hypotactic elaboration, the binder, which is within the elaborating clause, is underlined. The domain is in bold.
(County Media, 2013, p. 3) Ayinde was sentenced jail term that date 26 month of July, (a thing) which has finished for his daughter misery. Ayinde was sentenced on 26 th July, which has ended his daughter's misery.
The subordinate clause in the illustration is ũrĩa ũninĩire mwarĩ mĩnyamaro (which has ended his daughter's misery), a non-restrictive relative clause. It is a comment on the notion in the whole dominant clause Ayinde atuĩrĩirwo kĩoho kĩu tarĩki 26 mweri wa Mũgwanja (Ayinde was sentenced on 26 th July), which is repeated in ũndũ (something) just before the elaborating clause. This is what ended the unhappiness of Ayinde's daughter.
The more common domains of finite elaborating clauses in Gĩkũyũ are constituents of the dominant clause. When this is the case, the domain could be at the beginning of the clause or otherwise as shown in Example 14 below, whose domain in the initial position.
(And then) the rear side of that dwelling, which faces the west, you will make for it six frames.
The domain here is mwena wa thutha wa gĩikaro kĩu (the rear side of that dwelling), a nominal group. It is elaborated by the non-restrictive relative clause ũrĩa ũrorete ithũĩro (which faces West). The initiating clause is (Ningĩ) mwena wa thutha wa gĩikaro kĩu ũkaũthondekera mbaũ ithathatũ ((And then) the rear side of that dwelling you will make for it six frames). In Example 15 below, a hypothetical clause complex, the domain is in the mid position of the dominant clause.
The father gave children his all, who had surrounded him, word his of final.
The father gave all his children, who had surrounded him, his final word.
The finite elaborating clause is iria ciamũrigicĩirie (who had surrounded him) and its domain is the nominal element ciana ciake ciothe (all his children). The domain occupies a mid-position in the initial clause ithe nĩaheire ciana ciake ciothe mũgambo wake wa mũthia (the father gave all his children his final word). The elaborating clause only provides some extra information on ciana ciake ciothe (all his children), that the children were around their father. The domain could also occupy the final position as illustrated below: 16. Nĩkwarĩ mũndũ wetagwo Jusuf, ũrĩa arutwo matuĩte Baranaba. (Gĩkũyũ Bible, 2008, Acts. 4:36) There was a person called Joseph, whom the disciples had named Barnabas.
The elaborating clause is ũrĩa arutwo matuĩte Baranaba (whom the disciples had named Barnabas). Its domain is mũndũ wetagwo Jusufu (a person called Joseph), which is the last constituent the initiating clause nĩkwarĩ mũndũ wetagwo Jusuf (There was a person called Joseph). Ideally, the elaborating clause comes immediately after the domain. The elaborating clause here is o we wahotithirie gwĩka ũguo (he (who) enabled me to do that) and its domain is Kristo Jesu Mwathani (Christ Jesus Lord). They are detached by the nominal expression ngatho (thanks). The use of overt relation markers in hypotactic elaboration in Gĩkũyũ was noted to be intensive, unlike in paratactic elaboration, where such markers were scarce. In finite clause elaboration, the markers were binders in the form of relative pronouns, which introduce the finite elaborating clauses and refer back to the domains of the clauses. For instance, in the Gĩkũyũ clause complex nĩkwarĩ mũndũ wetagwo Jusuf, ũrĩa arutwo matuĩte Baranaba (there was a person called Joseph, whom the disciples had named Barnabas), example 16 above, the relative pronoun is ũrĩa, a pronoun equivalent to the English pronoun whom. It refers back to the domain mũndũ wetagwo Jusuf (a person called Joseph) and introduces the finite elaborating clause arutwo matuĩte Baranaba (the disciples had named Barnabas). The finite element in the clause is matuĩte, translating to (they) had named.
It is worth noting that the head word in the domain bears the class prefix which controls concord with the relative pronoun (Bresnan & Mugane, 2006) in number and size. For instance, the domain in the Example 17 above, mũndũ wetagwo Jusuf (a person called Joseph) the head word is mũndũ, which belongs to class 1. Thus, it takes ũrĩa (whom), the concordial relative pronoun for this class. Changing to the diminutive form of mũndũ yields kamũndũ (a small person) which belongs to class 12 and will therefore take the pronoun karĩa (whom). The augmentative form will be kĩmũndũ (a big/bad person) of class 7 and its respective pronoun will be kĩrĩa (whom).

Non-finite clause elaboration
The dependent clause in this category is non-finite, having a non-finite verb (Halliday & Matthiessen, 2014). This can be illustrated by the clause complex The administrator was rude to the clients, annoying them a great deal. The elaborating clause is annoying them a great deal, bearing the non-finite verb annoying.
Non-Finite elaborating clauses in Gĩkũyũ were noted to be uncommon. Such clauses are the same as the English toinfinitive, -ing infinitive or the -ed/-en infinitive clauses. Non-finite elaboration was noted to be possible specifically for domains that are constituents of the main clause. In such cases, and since Gĩkũyũ is agglutinative, a pro-form replicates the domain in the infinitive verb in the elaborating clause, a characteristic absent in English clause complexes. This paper calls the pro-forms domain prefixes. Such are underlined in the examples below. The domain in Example 18 below, a hypothetical example as well, is at the beginning of the dominant clause.
The initiating clause here is Mwana ũcio nĩakũngũĩire ũhotani ũcio (the child celebrated that victory). Anyitĩte gĩkombe-rĩ ((while) holding the trophy) is the dependent clause, whose domain is mwana ũcio (the child). The nonfinite element is the verb nyitĩte ((while) holding) in the agglutinated expression anyitĩte ((while) he was holding) in conjunction with the expletive suffix rĩ attached to the noun gĩkombe (trophy). Nyitĩte in combination with rĩ corresponds to holding, an -ing non finite verb. Examples 19 and 20 below are hypothetical scenarios too. Example 19 has the domain in mid position: 19. Ahũrirwo nĩ kĩrĩndĩ kĩnene na mahiga, kĩrakarĩtio nĩ mĩtugo yake.
He was beaten by a crowd huge with stones, (the crowd) annoyed by behaviours his. He was beaten by a huge crowd with stones, (the crowd) annoyed by his behaviour.
The domain is kĩrĩndĩ kĩnene (a huge crowd), an element of the dominant clause ahũrirwo nĩ kĩrĩndĩ kĩnene na mahiga (he was beaten by a huge crowd with stones). The element is elaborated by kĩrakarĩtio nĩ mĩtugo yake ((the crowd) annoyed by his behaviours). The non-finite element in the elaborating clause is rakarĩtio in the agglutinated expression kĩrakarĩtio and carries the meaning of the infinitive verb annoyed. Conversely, the domain in Example 20 appears finally in the initiating clause. 20. Ateng'eri nĩmaheirwo ngari ni Mũseveni, acanjamũkĩte mũno ni ũhotani wao.
The athletes were given vehicles by Museveni, (he) thrilled greatly by their victory. The athletes were given vehicles by Museveni, thrilled by their victory.
The domain is Mũseveni (Museveni), hosted by the initiating clause Ateng'eri nĩmaheirwo ngari ni Mũseveni (the athletes were given vehicles by Museveni). The elaborating clause is acanjamũkĩte mũno ni ũhotani wao ((he) thrilled by their victory), bearing the non-finite component canjamũkĩte. Canjamũkĩte means thrilled, similarly an -ed infinitive verb.
In non-finite elaboration, the relation is marked by the concordial class markers. The markers are affixes in the verb of the subordinate clause, an infinitive verb, and echo the domain. In example 19 above, mwana ũcio-rĩ, anyitĩte gĩkombe-rĩ, nĩakũngũĩire ũhotani ũcio (that child, holding the trophy, celebrated that victory), the domain is mwana ũcio (that child), which is echoed by prefix a-the concordial class marker, in the verb anyitĩte (holding). Prefixes have been found to be a significant feature in morphosyntax of Luhyia, also a Kenyan Bantu language, in changing common nouns into names of places (Mandillah, 2022). The elaboration strategies identified in hypotactic Gĩkũyũ clause complexes are summarised in Table 3 below:

Conclusion
This study aimed at exploring elaboration in Gĩkũyũ clause complexes. The findings demonstrated that, indeed, Gĩkũyũ avails varied resources to its users for making a statement clearer by means of elaboration. The resources give the language users a choice between paratactic and hypotactic techniques of clause complexing, which results in different categories of elaboration under the two. Paratactically, the resources enable Gĩkũyũ users to restate, illustrate or explain their statements while, hypotactically, the users are able to describe their statements in whole or in part through explanation or evaluation. One major difference noted between paratactic and hypotactic elaboration in the language is that while the former was observed to rely mainly on juxtaposition of the clauses in the clause complexes with scarce use of explicit markers of elaboration, such as conjunctive expression, the latter purely relies on explicit markers, which include relative pronouns and concordial class markers.

Contribution of the study
The study has contributed to a deeper understanding of Bantu languages in the perspective of Functional Linguistics, adding to the corpus of the already existing related knowledge. Specifically, it demonstrates the applicability of Systemic Functional Linguistics in the analysis of Gĩkũyũ clause complexes. This contribution is of great significance owing to the fact that Gĩkũyũ is an agglutinating language as opposed to English, the language used by Halliday and Matthiessen (2014) in describing Functional Grammar.

Recommendations
The study brings to the fore different ways in which expansion in Gĩkũyũ clause complexes is achieved through elaboration. This article recommends similar studies in other Bantu languages to see how the findings would compare to those on Gĩkũyũ. Secondly, Halliday and Matthiessen (2014) identify extending and enhancement as other means of expansion in clause complexes. This article recommends deep analysis of achieving expansion in Gĩkũyũ and related languages through these other two means. Such studies will further illustrate the significance of Systemic Functional Linguistics in analysing the grammar of Bantu languages.