Discourse Analysis Of Different Activities English Language Essay

Discourse analysis incorporates a number of different activities, and merging them all together is too ambitious. However, the natural nature of discourse analysis seems to be agreed upon by the scholars involved in the field. Stubbs (1983) summarizes the realm of discourse analysis like this:

The term Discourse Analysis is very ambiguous. I will use it … to refer mainly to mean the linguistic analysis of naturally occurring connoted spoken and written discourse. Roughly speaking, it refers to attempts to study the organization of language above the sentence or above the clause, and therefore to study larger linguistic units, such as conversational exchanges or written texts. (Stubbs, 1983: 1)

Discourse analysts seem to be interested in analyzing any kind of spoken and written discourse, and there are various distinct analytic patterns that can be applied to both types of discourse. This paper attempts to emphasize on some discourse analytic patterns applied by researchers in the field of learner written discourse analysis. Learner discourse is an invaluable resource that can be used, exploited, and analyzed by interested researchers in an attempt to clarify, classify, and describe the way learners use the language. Learner written discourse and texts often offer a rich resource for a learner’s language development. They can be used for diagnosis and evaluation, language awareness raising, and classroom use. In fact, there’s a good case for learners’ texts being the best resources for a focus on language and they are closer to the developmental stage that other learners are going through (their interlanguage).When learners see their own used as classroom learning material, they become more motivated and as Scott Thornbury (2005) mentions: “serves to break down the distinction between language learner and language user.”

In describing learner language, learner discourse can be dealt with via four major approaches identified by Ellis (1994: 44) including the study of learners’ errors. The focus of this paper is to try to study learners’ errors on some semantic and syntactic areas of discourse analysis naming grammatical and lexical cohesions, coherence (theme and rheme), ellipsis, reference, collocation, and nominalization. While focusing mainly on analyzing written texts for posterior classroom application, all the samples discussed seek to relate analysis of specific written texts to social and cultural contexts in which such texts are written and read. In our examples, we tend to demonstrate the approach of analyzing learners’ written assignments dealing with the issue of “using and learning Arabic by foreigners in the United Arab Emirate” to investigate the particular discourse methods used in the similar texts. In an attempt to relate discourse analysis with corpus linguistics, the researcher later suggests that the use of ‘concordancing’ in the teach­ing and designing of written discourse in language classrooms is encouraging and worthwhile and presents samples of exploited concordances to highlight collocations. It des­cri­bes a possible way of having students approach discourse analysis tasks in an inductive and learner-centered manner. In this paper, there is an emphasis on the relationship between the linguistic features of the written texts and the UAE society in which they are produced.

Written Discourse

Written discourse incorporates communication by exploitation of textual material. It can be outlined in numerous modes. McCarthy defines, discourse analysis as “the study of the relationship between language and the context in which it is used” (McCarthy 1991:5). Written discourse does not have to deal with people speaking all simultaneously or even with spontaneous interruptions. McCarthy (1991:6) states that in written discourse the writer normally has time to prepare the text. He also mentions that in written discourse, “the sentences are usually well formed in a way that the utterances of natural spontaneous talk are not”.

Learner Discourse

Merely testing learners on their aptitude to write, or complete, isolated sentences is clearly inadequate if their overall ability to communicate at the level of discourse is a goal. However, even when whole texts are exploited for assessment or diagnosis, there is an affinity for many teachers not to be able to observe beyond their surface grammar errors, or to appreciate their strengths irrespective of their weaknesses. To guarantee a more efficient, more extensive, and more reasonable judgment, more comprehensive criteria for assessing texts are needed. Ellis (1994: 44) identified four major approaches in describing learner language:

the study of learners’ errors

the study of developmental patterns

the study of variability

the study of pragmatic features

In our attempt to analyze learner’s text, the researcher tries to bear in mind the above-mentioned considerations.

Patterns of Text Analysis

Written texts can be analyzed using various patterns and methods. However, in our paper we tend to analyze texts dealing with the following aspects:

Grammatical cohesion including reference (cataphoric, anaphoric, exaphoric, and endophoric cohesive devices), ellipsis, substitution, nominalization, and conjuncts;

Lexical cohesion including reiteration, superordinate, repetition, and collocation.

We now try to shed some light on the application of each of the abovementioned aspects.

Cohesion

Cohesion is the way a text is held together and has meaning (Holland and Lewis 2001:53), and the way unrelated structural elements “are linked together, through the dependence of one on the other for its interpretation” (Halliday and Hasan 1976:27 in Holland and Lewis 2001:55). Cohesive links can work within the text, endophorically, or outside the text, exophorically (Holland and Lewis 2001:53). Within the text they may be anaphoric when they refer to something that has already been mentioned, or cataphoric when they look to something that is yet to be mentioned (Holland and Lewis 2001:53). Halliday and Hasan propose grammatical cohesion and lexical cohesion, each with their own sub classifications, as the two categories of cohesion (1976 in Holland and Lewis 2001:55). Written texts can be analyzed by detecting the grammatical connections between individual clauses and sentences of the text known in linguistics as grammatical cohesion. Grammatical cohesion is what meshes the text together. Renkema defines cohesion as the phenomenon of connectedness of sentences or utterances in discourse (Renkema, 2004). Cohesion in text conveys meaning to the reader.

Grammatical Cohesion

Cohesion in a text, as mentioned earlier, can be established in two ways; grammatical and lexical cohesion. We shall begin by looking at grammatical cohesion. Halliday and Hasan are pioneers in the study of grammatical cohesion. They mention that there are cohesive relationships between the sentences in a text and these relationships create texture. Texture distinguishes a text from something, which is not a text. McCarthy (1991:34) defines grammatical cohesion as “the surface marking of semantic links between clasuses and sentenses in written discourse, and between utterances and turns in speech.”

Halliday and Hasan consider grammatical cohesion through reference, substitution, ellipsis and conjunctions. Reference is further subdivided into the categories of personal, such as pronouns (e.g. he, she, it, him, they, etc.), demonstratives (this, that, these, those), the article the, and items like such a (Halliday and Hasan 1976: 37-9) and comparatives (Holland and Lewis 2001:57-8). All three forms of referential cohesion can work either endophorically or exophorically (Holland and Lewis 2001:57-8). Substitution, usually working anaphorically, replaces a reference with a dummy object, while ellipses are of a similar nature, except the missing reference is replaced by nothing (Holland and Lewis 2001:58). Ellipsis is when expected elements of the text are omitted because they have already been mentioned formerly in the text or that they are spontaneously inferred. The last category of cohesion is conjunctions which operate through grammatical relationships between clauses or propositions, and are subdivided into the categories of additive, adversative, causal, and temporal (Halliday and Hasan 1976 in Holland and Lewis 2001:60).

Lexical Cohesion

Halliday and Hasan consider lexical cohesion through reiteration and collocation (Angermeyer 2002:365). Reiteration could include the re-mentioning of an item using the original noun phrase, a synonym, a near-synonym, a superordinate expression, or a general word (Halliday and Hasan 1976 in Holland and Lewis 2001:63). Following such reiterations we can possibly create a lexical chain in a text.

With regards to collocation, Halliday and Hasan note that it is a “problematical” category of lexical cohesion, (1976 in Angermeyer 2002:365), and Hasan later rejects collocation as a lexically cohesive category (1984 in Angermeyer 2002:365). Similarly, Hoey considers the various forms of repetition as the key aspect of lexical cohesion (1991 in Angermeyer 2002:365).

Corpus Analysis

‘Corpus’ means a body, and ‘corpus analysis’ thus refers to the analysis of a body of language data. A corpus can be small (for example one newspaper article or letter) or large (several million words of naturally occurring spoken or written language). Linguistically speaking, corpus is any collection of natural language examples. It is a collection written and/or spoken examples of the usage of a language, employed in linguistic analysis. Presently, corpus analysis employs computer applications, called concordancers, in the analytic procedure. The computer applications designed for this type of analysis include concordance programs that can, for instance, recognize specific words selected by the researcher and demonstrate how frequently these words are used in discourse. Analyses of large corpora of spoken and written English have revealed the frequency and co-occurrence of many different lexical and grammatical items. This co-occurrence is called ‘collocation’. These analyses have been capable of illustrating enormously facts about language that could hardly be inferred intuitively.

For the purpose of this paper we tend to rely mostly on two major corpora naming the Collins WordbanksOnline English corpus sampler which is composed of 56 million words of contemporary written and spoken text as well as Vienna-Oxford International Corpus of English which is a structured collection of language data of English as a Lingua Franca.