2 Beyond the sentence:rhetorical or discourse analysis There were, as we shall see, serious flaws in the

register analysis-based syllabus, but, as it happened, register analysis as a research procedure was
rapidly overtaken by developments in the world of linguistics. Whereas in the first stagé of its
development, ESP had focussed on language at the sentence level, the second phase of development
shifted attention to the levelabove the sentence, as ESP became closely involved with the.emerging field
of discourse or rhetorical analysis. The leading lights in this movement were Henry Widdowson in Britain
and the so-called Washington School of Larry Selinker, Louis Trimble, John Lackstrom and Mary Todd-
Trimble in the United States The basic hypothesis of this stage is succinctly expressed by Allen and
Widdowson (1974) We take the view thar the difficulties which the students encounter arise not so
much rom a defective knowledge of the system of English, but from an unfamiliarity with English use,
and that consequently their needs cannot be met by a course which simply provides further practice in
the composition of

sentences, but only by one which develop a knowledge of how sentences are used in the performance of
different communicative acts. Register analysis had focussed on sentence grammar, but now attention i
shifted to understanding how sentences were combincd in discourse to produce meaning. The concern
of research, therefore, was to identify the organisational patterns in texts and to specify the linguistic
means by which these patterns are signalled. These patterns would then form the syllabus of the ESP
course. The Rhetorical Process Chart below (from EST: A Discourse Approach by Louis Trimble (1985) is
representative this approach:

Level Description of leve! A. The objectives of the total discourse EXAMPLES: 1. Detaiing an experiment
2. Making a recommendation 3. Presenting new hypotheses or theory 4. Presenting other types of EST
information B. Thegeneral rhetorical functions that develop the objectives ol LevelA EXAMPLES: 1.
Stating purpose 2. Reporting past research 3. Slating the problem 4. Presanting information on apparatus
used in an experiment- a) Description b) Operation 5. Presenting information on experimental
procedures C. The specific rhetorical functions that develop the general rhetorical functions of Level B
EXAMPLES: 1. Desciption: physical, tunction, and process 2. Definition 4. Instructions 5. Visua verbal
relationships D. The rhetorical techniques that provide relationships within and between the rhetorical
units of Level O EXAMPLES: Orders 1. Time order 2. Space order 3. Causality and result Il. Pattems
Causality and result 2. Order of importance 4. Analogy 5. Exemplification 6. Iiustration Figure 2:
Rhetorical Process Chart

What is ESP? As in stage r there was a more or less tacit assumption in this approach that the rhetorical
patterns of text organisation differed signiicantly between specialist areas of use: the rhetorical structure
of science texts was regarded as different from that of commercial texts, for example: However, this
point was never very clearly examined (see Swales, 1985, pp. 70-) and indeed paradoxically, the results
of the research into the discourse of subject-specific academic texts were also used to make
observations about discourse in general (Widdowson, I978) The typical teaching materials based on the
discourse approach taught students to recognise textual patterns and discourse markers mainly b means
of rext-diagramming exercises (see below p. 36). The English in Focus series (OUP) is a good example of
this approach.

3 Target situation analysis The stage that wc come to consider now did new to the range of knowledge
about ESP What it aimed to do was to take the existing knowledge and set it on a more scientific basis, b
establishing procedures forrelating language analysis ore closely to learners' reasons for leang Given that
the purpose of an ESP course is to enable learners to function adequately in a target situation, that is,
the situation in which the learners will use the language they are learning, then the ESP course design
process should proceed by first identifying the target situation and then carrying out a rigorous analysis
of the linguistic features of that situationAThe identiied features will form the syllabus of the ESP course.
This process is usually known as needs analysis. However, we prefer to take Chambers' (1980) term of
target situation analysis', since it is a more accurate description of th process concèrned not really add
anything The most thorough explanation of target situation analysis is ti system set out by John Munby in
Communicative Syllabus Design (I978). The Munby model produces a detailed profile of the learners
needs in terms of communication purposes, communicative setting, the means of communication,
language skills, functions, structures etc. (see below p. 55.) The target situation analysis stage marked a
certain 'coming of age for ESP. What-had previously been doue very much in a piecemeal way, was now
systematised and learner need, was apparently placed ar the' centre of the course design process. It
proved in the event to be a false dawn. As we shall see in the following chapters, the concept of needs
that it was based on was far too simple.

4 Skills and strategies We noted that in the first two stages of the development of ESP all the analysis had
been of the surface forms of the language (whether at sentence level, as in register analysis, or above, as
in discourse analysis) The target situation analysis approach did not really change this because in its
analysis of learner need it still looked mainly at the surface linguistic features of the target situation. The
fourth stage of ESP has seen.anattempt to look below the surface and to consider not.the language irself
but the thinking processes that underlie language use. There is no dominant figure in this movement,
although Christine Nutall (1982) and Charles Alderson and Sandy Urquhat (r984) as having made
significant contributions to work on reading skills. Most of the work in the area of skills and strategies,
however, has been done close to the ground in schemes such as the National ESP Project in Brazil (see
below p. 172) and the University of Malaya ESP Project (see ELT Documents 1o7 and Skills for Learning
published by Nelson and the University of Malaya Press) we might mention the work of Francoise Grellet
(x98i), Both these projects were set up to cope with study situations where the medium ofinstruction is
the mother tongue but students need to read a number of specialist texts which arc available only in
English. The projects have, therefore, concentrated their efforts on reading strategies. -

Theprincipal idea behind the skills-centred approach is that underlying all lanuage use there are common
reasoning and interpreting processes, which, regardless of the surface forms, enable us to extract
meaning from discourse. here iš, theretore, no nced to tocus closely on the surface forms of the
language. The focus should rather be on the underlyin interprctive strategies, which cnabte the learner
to cope with the surface forms, for example guessing the meaning of-words from context, using visual
layout to deternine the type of text, exploiting cognates i.e. words which are similar in the mother
tongue and the target language) etc. A focus ific subject registers is unnecessary in this approach,
because the underlying processes are nor specific to any subject register 'It was argued that reading skills
are not language-specific but universal and rhat there is a core of language (for example, certain
structures of argument and forms of presentation) which can be identified as"academic" and which is
not subject-specific. (Chitravelu, 198

5 A learning-centred approach In outlining the origins of ESP (pp. 6-8), we identified threc forces, which
we might characterise as need, new ideas about language and new ideas about learning. It should have
become clear that in its subsequent development, however, scant attention has been paid to the last of
these forces learning. Al of the stages outlined so far have been funda- mentally flawed, in that they are
all based on descriptions of language use. Whether this description is of surface forms, as in the case of
register analysis, or of underlying processes, as in the skills and strategies approach, the concern in each
case is with describing what people do withlanguage. But our concernin ESP is not with language use -
although this will help to define the course obiectives. Our concerm is with language tearning. We
cannot simply assume that describing and exemplifying what people do with language will enable
someone to learn it. If that were so, we would need to do no more than read a grammar book and a
dictionary in order o learn a language. A truly valid approach to ESP must be based on an understanding
of the processes of language learning This brings us to the fifth stage of ESP development the learning-
centred approach, "which will form the subject of this .book. The importance and the implications of the
distinction that we have made between language use andlanguage learning will hopefully become clear
as we proceed through the following chapters

4 An information leaflet about the Brazilian ESP Project summarises its aims as follows: The"specific
purpose" most common within the participant eaiversities the reading of specialist literature in English.
Consequently there is a consensus within the project to focus on the teaching of reading strategies with
the use of authentic materials and the use of the native language in spoken classroom discourse. The
teaching of grammar is based on the minimum necessary for understanding academic texts. The
emphasis is largely on a general course content to cover common problems (such as reading strategies),
rather than specific courses according to the student's subject specialism (e.g. "English for Engineers") a)
Why do you think this approach has been adopted? b) Do you think that the approach is a justifiable
response to the needs of the students? c) How do you think students and teachers wili reacr to this
approach?