A text can be defined as a piece of language use which is experienced and accepted as a communicative unit on syntactical, semantical and pragmatical foundations. What is meant by this is that any text is understandable because it has a unique linguistic (syntactic, morphological and lexical) structure; it is intelligible, seeing as it has a set meaning; and it is explainable for the fact that it can be placed in the context in which it is used. Another characteristic by which a text can be identified is the presence of texture. Texture is acquired when a text functions as a unit with regard to its surroundings. Cohesive markers are used to accomplish this unit and consequently ensure texture.
More than four decades ago, Halliday and Hasan (1976:4) identified five types of cohesive marker, namely reference, substitution, ellipsis, conjunction and lexical cohesion. The use of each of these markers is indicated by certain language forms in the text.
When reference is used for accomplishing cohesion, it creates continuity; in other words, its use indicates that the same reference (or matter) that was introduced earlier in the text is applicable for a second time. Pronouns, adverbs and articles are used, among others, to indicate the use of reference.
Secondly, cohesion can be acquired by means of substitution. When substitution is used to create a cohesive text, a type of substitute word (or counter) is used to exclude the repetition of a word or phrase. Substitutive words such as so, one, do that or do it are used to indicate the realisation of lexical cohesion.
Thirdly, relation between words and sentences can be indicated with an ellipsis; in other words, there is a presumption in the structure that the reader must predict or understand something. The meaning of the syntactical process is usually embedded in the preceding text section.
Fourthly, cohesion can be acquired through conjunction. Conjunction differs from the other cohesion markers in that it does not establish cohesion in the form of continuity with reference to continuity of form (as is the case with ellipsis and substitution); cohesion is established because a particular semantic relation (contiguous, contradictory, explanatory or temporal) is exemplified explicitly between sentences through lexical items like conjunctions (for example and, but and because), adverbs (yet, therefore, thus) and adverbial phrases (for example the following day).
Lastly, lexical cohesion can result in binding. Lexical cohesion is realised by the use of a language’s vocabulary. For this, repetition of the same word/phrase/expression, synonyms, semantic contrast and hyponyms, among others, are utilised.
The Curriculum and Assessment Policy Statement (CAPS) of the intermediary (grade 4–6) and senior (grade 7–9) phases states that learners must use conjunctions such as repetition, synonyms and antonyms when producing essays to create cohesive texts. When these language forms are used to create cohesive chains, they point to the realisation of lexical cohesion, as indicated in the previous paragraph. We wanted to investigate the various language forms that learners use in Afrikaans texts in order to realise lexical cohesion, because the realisation of lexical cohesion is not yet studied in the writing of Afrikaans mother tongue speakers, even though it is implicitly stated in the CAPS.
In this research article, a corpus-based study is undertaken to determine the independent use of language forms for the realisation of lexical cohesion among Afrikaans-speaking grade 6 and grade 9 learners. One grade 6 corpus (20 723 words) and one grade 9 corpus (22 171 words) were analysed. Each of the essays were read more than once, word by word, and analysed according to the measuring framework. Each language form that was used effectively for the realisation of lexical cohesion in the grade 6 and grade 9 corpora and included in the measuring frame work, was coded. ATLAS.ti, which was used for the coding of the data, improves the contribution of the study, because it differs from existing Afrikaans corpus studies that use WordSmith Tools as corpus analysis software. The effective use of language forms for the realisation of lexical cohesion, which indicate links between words within the same sentence (i.e. sentence-internal cohesion marker use) and language forms which indicate meaning relations between words across sentence boundaries (i.e. sentence-external cohesion markers), were coded by means of ATLAS.ti. After the coding of the effective use of each language form for the realisation of lexical cohesion by means of ATLAS.ti, we used ATLAS.ti to create a synthetising report indicating how many times each language form figured in the grade 6 and grade 9 corpus. These frequencies are normalised after cohesion marker use per 1 000 words, as is appropriate in corpus linguistics.
From the analysis it seemed that all the grade 6 and grade 9 participants involved had figured all the different language forms for the realisation of lexical cohesion included in the measuring framework of the grade 6 and grade 9 corpus, namely (i) repetition, (ii) synonyms and nearby synonyms, (iii) antonyms, (iv) superordinates, hyponyms, coordinated members in organised and unorganised sequences and epithet (i.e. inclusion), (v) derived forms and the repetition thereof, as well as (vi) alternate noun phrases. Parallelism as a form of repetition, which is not in the measuring framework, is also used for realising lexical cohesion.
The contribution of this article is a summary of various language forms by which lexical cohesion is realised as measured against learner corpora. Afrikaans teachers can use the findings of this study in the classroom to give guidance on how conjunctives (as stated in the CAPS) can be used to bind consecutive language phrases syntactically – in other words, to make texts cohesive by linking their surface components (i.e. the sentences and the words they consist of) with each other.
The creation of cohesive texts, as effected by making use of various cohesion markers (not solely lexical cohesion), should be encouraged in the classroom because cohesive texts are texts of which the various parts fit together. It is precisely this characteristic of co-interpretation that distinguishes a collection of non-related sentences of a text as a linguistic unit. The use of cohesion markers indicates that an idea is continued from one point of discourse to another. In so doing, semantic continuity is ensured between different parts of a text. It is this semantic continuity which enables the listener or reader to, just as with an incomplete puzzle, provide the missing puzzle pieces so that the whole picture can be obtained and the text can be optimally interpreted. The continuity that is created through the linking of surface components establishes coherence, even though other researchers rather use the term enforce to describe the relationship between cohesion and coherence, while others argue that coherence is reflected by means of the relationship with cohesion. The findings of this article can therefore be used by teachers when teaching writing and presentation as language skills in the CAPS, since the use of cohesion markers has an important text-forming function.
Keywords: cohesion; corpus linguistics; learner writing; lexical cohesion; realising; language forms; text linguistics