On RA Abstracts From Rhetorical Structur

ENGLISH FOR
SPECIFIC
English for Specic Purposes 23 (2004) 280302 PURPOSES
www.elsevier.com/locate/esp
On RA abstracts: from rhetorical structure

to thematic organisation q
Rosa Lores *
Dpto. de Filologa Inglesa y Alemana, Facultad de Filosofa y Letras, Universidad de Zaragoza,

Campus Universitario, Zaragoza 50009, Spain
Abstract
This paper reports an analysis of research article (RA) abstracts from linguistics journals
from two related angles: rhetorical organisation and thematic structure. Based on a small scale
study it reveals two major types of rhetorical organisation, here called the IMRD type and the
CARS type. When thematic analysis, in terms of thematic progression and method of thematic
development, is applied to the two types of structure, distinct patterns of thematic distribution
and choice are revealed, showing that the study of thematization can shed light on the complex
prole of the RA abstract and contribute towards the understanding and explicit description
of these texts. Moreover, the approach taken in this study shows potential for further research
and pedagogic applications.
2003 The American University. Published by Elsevier Ltd. All rights reserved.
1. Introduction
For a long time it has been assumed that writing an abstract in English was a
relatively easy task for the non-native English speaker as compared with the arduous
task of completing the research article (RA) itself. However, this is not necessarily
q
This research was carried out within the framework of the Project entitled El ingles y el conocimiento
cientco: an
alisis pragmatico-cognitivo de la metafora gramatical y su utilizaci
on en la creaci
on, expresi
on y
divulgaci on del conocimiento cientco (BFF 2000-1074), funded by the DGES (Spanish Ministry of
Education and Culture). The rst part of this research, focusing on the rhetorical organisation of abstracts was
presented in the XX AESLA Congress (Jaen, Spain, April 2002). I should like to thank Professor Hickey for
his enlightening comments and the thorough revision of this paper. Needless to say, any inaccuracies are mine.
*
Tel.: +34-976-761-538; fax: +34-976-15-19.
E-mail address: rlores@posta.unizar.es (R. Lores).
0889-4906/$30.00 2003 The American University. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.esp.2003.06.001
R. Lores / English for Specic Purposes 23 (2004) 280302 281
so, due to the fact that the abstract constitutes a genre in its own right which, while
sharing many features of the RA, also diers in several important aspects, one of
which is its rhetorical structure.
Abstracts are a growing eld of study in linguistics. The interest that linguists show
in the genre of the RA abstract stems from the need to understand the mechanisms
which underlie these multifunctional texts. As Ventola (1994a, p. 333) states, abstracts
have become a tool of mastering and managing the ever increasing information ow
in the scientic community. They constitute the gateway that leads readers to take up
an article, journals to select contributions, or organisers of conferences to accept or
reject papers. Moreover, English is the major vehicle by which this particular journey is
made. In fact, it is not unusual for journals published in languages other than English to
expect the author to write an English abstract of their article. This requirement re-
sponds to the function of the English-language abstract, which is to guarantee that the
reported results of scientic work will circulate worldwide (Ventola, 1994a, p. 333).
Taking previous studies in the eld as a starting point (Bhatia, 1993; Nwogu, 1990;
Swales, 1990; Ventola, 1994a, 1994b) the present paper sets out to analyse the global
organisation of RA abstracts from linguistics journals. As a second, but comple-
menting stage, thematic analysis in terms of thematic progression (Danes, 1974) and
method of thematic development (Fries, 1983) will be applied as a tool to explore the
textual mechanisms which lie under the construction and understanding of this genre.
2. The abstract as a distinct genre in academic prose
There seems to be general agreement on three ways in which abstracts dier from
RAs. They dier in their function, in their rhetorical structure and in their linguistic
realisations. These three areas are, undoubtedly, closely connected: the function of
an abstract will determine both its global structure and its linguistic realisation.
Given the brevity of this paper, the present study will be limited to the relationship
between function and rhetorical structure, and it will explore the ways in which the
rhetorical structure can be realised in dierent thematic organisations. Other lin-
guistic features of abstracts, such as the use of cohesive devices, nominalizations,
passive voice and the like, although intimately related to the global organisation and
the function of a text, must lie outside its limits.
An abstract is basically a concise summary of a much longer report. The Amer-
ican National Standards Institute (ANSI) denes it as follows: [it] is an abbreviated,
accurate representation of the contents of a document, preferably prepared by its
author(s) for publication with it (ANSI, 1979, p. 1, in Bhatia, 1993, p. 78).
This denition seems to suggest that abstracts faithfully portray the form and
content of their originals. But, as Nwogu (1990, p. 113) points out, this is not always
necessarily so. In fact, though abstracts may represent the content of their originals,
they do not (always) represent the surface form of these originals.
There seems to be general agreement (Day, 1988; Graetz, 1985; Jordan, 1991;
Ventola, 1994a) in distinguishing two basic types of abstract on functional grounds,
specically indicative and informative abstracts. The function of indicative
282 R. Lores / English for Specic Purposes 23 (2004) 280302
abstracts is to help readers understand the general nature and scope of the research
article; it indicates the subject and the main ndings of the paper, but it does not go
into a detailed step-by-step account of the process involved. On the other hand,
informative abstracts encapsulate the whole paper; as Jordan (1991, p. 507) states,
they act as a report in miniature.
The existence of two types of abstract dierentiated on functional grounds leads
logically to expectations of two matching though dierent rhetorical structures. Yet
only the rhetorical structure of informative abstracts seems to have been acknowl-
edged and described. Generally, it is understood (Cremmins, 1982; Day, 1988;
Graetz, 1985; Jordan, 1991; Nwogu, 1990; Ventola, 1994a, 1994b) that abstracts
display the rhetorical structure of the RA, which for long has been taken to be the
IMRD structure (IntroductionMethodsResultsDiscussion).
In my view, however, the fact that there is a second type of abstract, the indicative,
may encourage the writer to use a dierent rhetorical structure to better perform a
dierent function. The existence of such a dierent rhetorical structure would seem to
be endorsed by studies in the eld. Graetz (1985), for instance, allows up to four
dierent variations to the IMRD structure; Nwogu (1990) considers that only 2 out of
the 11 moves or sections which constitute the RA are obligatory in abstracts.
3. The analysis of data: materials, methods and ndings
The present study has been carried out on a corpus of 36 abstracts taken from
four prestigious publications in the eld of linguistics: Journal of Linguistics, Applied
Linguistics, Linguistics and Journal of Pragmatics (see Appendix A). Nine abstracts
were selected from recent issues of each journal.
3.1. About rhetorical structures
An analysis of the rhetorical organisation of the 36 abstracts selected was carried

out on the basis of Swaless (1981, 1990) description of move structure for RAs.
However, we cannot deny the fact that Swaless identication of stages in RAs in
terms of moves has been often questioned (Crookes, 1986; Dudley-Evans, 1994;
Lewin, Fine, & Young, 2001; Paltridge, 1994).
Move is a functional term that refers to a dened and bounded communicative
act that is designed to contribute to one main communicative objective, that of the
whole text. But the analysis of the research genre seems to lack uniform standards
for move identication, as they include not only lexicogrammatical features but also
cognitive criteria.
Two questions about this approach to genre analysis are pertinent here (Dudley-
Evans, 1994, p. 226): how we make decisions about the classication of moves, and
second, how we can be condent of the validity of the moves and the move cycles
that are posited. As Crookes (1986, p. 61) states, the Swales model is open to the
criticism that, however explicit its criteria and its exemplication, it remains in the
end based on personal and individual judgement, as Swales himself recognizes.
Table 1
IMRD structure CARS structure Combinatory structure
Linguistics 4 3 2
Journal of Linguistics 6 3
Journal of Pragmatics 5 3 1
Applied Linguistics 7 2
Total 22 (61.1%) 11 (30.5%) 3 (8.4%)
Taking into account that a certain scope for individual, subjective judgment in the
identication of moves in abstracts has to be catered for, the analysis of the corpus in
terms of move structure revealed three dierent types of structure:
(i) The IMRD structure, with the highest percentage of representation in the
corpus, is taken to be the characteristic rhetorical organisation of abstracts by
previous studies in the subject (Bhatia, 1993; Graetz, 1985; Nwogu, 1990; Swales,
1990), mirroring the structure of the RA. However, these studies tended to focus
only on the informative type of abstract.
This is in fact the type of structure which our data conrms as being the most
representative in the corpus analysed. Table 1 shows how, on average, this type
represents 61.1% (22 abstracts) of the total number of samples analysed, a per-
centage which, in some cases, like the journal Applied Linguistics, rises to 77.7% (7
abstracts). The sections, or moves, included give the following information:
Section 1 (Introduction). This may outline the authors purpose or objective, the
goals of the research or the problems that the author wishes to tackle.
Section 2 (Methods). Here the author indicates the way the problem has been
studied or the goal set out: this might include the data used and the methodology
followed.
Section 3 (Results). In this section a summary of the general ndings appears.
Section 4 (Discussion). This move might include an interpretation of the results,
some implications for further research or applications of the ndings.
Abstract 1 is a good example of this rhetorical structure which, in my view, fulls
the informative function, a function which pervades the entire paper, step by step
(numbers in bold type have been introduced to separate the four sections of the
abstract):
Abstract 1
1 (Introduction) This article investigates EFL learner use of high frequency
verbs, and in particular use of the verb MAKE, a major representative of this
group. The main questions addressed are: do learners tend to over- or underuse
these verbs? Are high-frequency verbs error-prone or safe? What part does
transfer play in the misuse of these verbs? 2 (Methods) To answer these ques-
tions, authentic learner data have been compared with native speaker data us-
ing computerized corpora and linguistic software tools to speed up the initial
stage of the linguistic analysis. The article focuses on what proves to be the
two most distinctive uses of MAKE, viz. the delexical and causative uses. 3
(Results) Results show that EFL learners, even at an advanced prociency
level, have great diculty with a high frequency verb such as MAKE. They
also demonstrate that some of these problems are shared by the two groups
of learners under consideration (Swedish- and French-speaking learners) while
others seem to be L1-related. 4 (Discussion) In the conclusion, the pedagogical
implications of the study are discussed and suggestions made for using concor-
dance-based exercises as a way of raising learners awareness of the complexity
of high-frequency verbs. (Applied Linguistics 22 (2), 2001)
(ii) Almost a third of the samples analysed (30.5%) were found not to follow the
IMRD rhetorical structure and follow what has been here called the CARS
structure, which, in my view, matches the indicative type of abstract. Generally, in
this structure no mention of the methodology undertaken is made. Firstly, a general
indication of the context in which the research has been carried out is given (e.g.
previous research or the scope within which the present study has to be contex-
tualised). Then, reference may be made to any perceived lacunae in the knowledge
of the question or indeed some kind of counter-claim may be made against it,
linguistically marked by means of conjunctions such as however or but. The nal
section usually announces the principal ndings or the way in which the research is
going to ll the gap found or answer the questions raised. This type of abstract,
then, indicates the scope of the paper and outlines some general ndings. In that
sense it fulls the function of the indicative abstract and, thus, mirrors the struc-
ture, not of the whole RA, but of the Introduction section in RAs, which Swales
(1990) described as the CARS model (Create a Research Space). Swales (1990, p.
141) sees the CARS model as consisting of three moves or sections, each made up
of dierent steps:
Move 1 (Establishing a territory). Three steps make up the structure of the rst
move: (i) claiming centrality, (ii) making topic generalizations, (iii) reviewing
items of previous research.
Move 2 (Establishing a niche). This consists of four steps: (i) counter-claiming,
(ii) indicating a gap, (iii) question-raising, (iv) continuing a tradition.
Move 3 (Occupying a niche). This also consists of three steps: (i) outlining pur-
poses or announcing present research, (ii) announcing present ndings, (iii) indi-
cating RA structure.
My corpus study reveals that only the general structure of moves is followed and,
in each move, no more than one step is selected by the writer. The scope of the
corpus is not large enough to allow us to draw reliable conclusions about which steps
are obligatory and which are optional. Abstract 2 below exemplies the CARS or-
ganisation. Roman numbers I, II and III identify each section.
Abstract 2
I [Establishing a territory] Since the early 1980s, in much of the work within a
genre-analytic approach, communicative goal or purpose has been used as an
important and often primary criterion for deciding whether a particular
discourse falls within a particular generic category (book review, sales letter,
cross-examination, etc.). II [Establishing a niche] However, as the number of
genre studies has increased, and as genre theory has become more complex,
the concept of communicative purpose has also become more complex, multi-
ple, variable and generally hard to get at. We believe that one consequence of
these developments has gone largely unnoticed: that uncertainties surrounding
communicative purpose undermine its claimed role as a means of assigning
genre membership. III [Occupying the niche] In this paper we discuss this par-
adox, illustrate the diculties that can arise, and then suggest a procedure
whereby communicative purpose can be retained as a viable and valuable con-
cept. Applied Linguistics 22 (2), 2001
(iii) A small percentage of abstracts (3 abstracts, found only in two of the four
journals) did not match either of the two structures indicated above. A representative
of this type of organisation is Abstract 3, which starts with a CARS type of structure
in which an IMRD model is embedded. Section 1 corresponds to the CARS move of
establishing a territory, where previous research is reviewed; Section 2 corre-
sponds to the establishing a niche move, where a general assumption is disputed
(see the use of however), and Section 3 corresponds to the occupying a niche
section but abandons the indicative type and moves on to the information type.
Thus, three subsections can be easily dierentiated here, this time mapping on to the
IMRD structure quite closely: in Section 1 the general purpose of the paper is stated
(Introduction); in Section 2 the methodology is briey explained (Methods) and in
Section 3, the ndings are summarised (Results):
Abstract 3
I (Establishing a territory) The natural semantic metalanguage (NSM) theory
developed in such research as Wierzbicka (1972, 1980, 1994a) and Goddard
(1994, 1998) has taken SAY to be a semantic primitive without specifying ex-
actly which semantic domains it covers. II (Establishing a niche) Empirical re-
search shows, however, that the words that might intuitively be taken as the
exponents of SAY in English (say), Hebrew (amar), Japanese (iu), and Span-
ish (decir) very frequently do not translate as each other, so that we cannot
simply say that SAY say amar iu decir. III (Occupying a niche) 1 (In-
troduction) In order to reconcile the translation data with a theory of univer-
sal primitive meaning, 2 (Methods) it is necessary to posit a number of cases
of suppletion and polysemy, the former to account for automatic, non-mean-
ing-based alternations between allolexes reecting a single semantic primitive,
the latter to account for meaning-based alternations between dierent lexe-
mes. 3 (Results) The resulting analysis is relatively surprising, for example,
that tell has two distinct meanings (tell a story vs. tell me what you want), that
tell in the latter meaning alternates allolexically with say, and that ask can
serve as an exponent of SAY when the clause it introduces is formally a ques-
tion but does not pragmatically function as a question. (Linguistics 39 (4),
2001)
This organisation displays some kind of combinatory structure, which, in my view,

corresponds to the mixed type of informativeindicative abstracts, acknowledged by
researchers such as Ventola (1994a, p. 335).
Table 1 shows the distribution, in numbers (and percentages), of the abstracts
analysed, according to source and rhetorical structure.
3.2. Thematic analysis: thematic progression (TP) patterns and method of thematic
development
As has been shown above, a relationship seems to exist between the abstracts
function (informative, indicative or informativeindicative) and the rhetorical or-
ganisation of the information in this type of text. Since moves perform dierent
rhetorical functions, they require dierent linguistic resources to realise those func-
tions (Swales, 1990). This relationship makes us consider thematic analysis as po-
tentially useful in developing genre awareness, since the choice of rst position in the
sentence is signicant as part of a writers available linguistic resources. In Gosdens
(1993) view, the thematization of certain types of information according to the genre
is one means through which writers can achieve local discourse goals, for instance,
the signalling of moves within sections.
With this purpose in mind, thematic analysis needs to be carried out not in iso-
lated sentences but (i) as thematic progression (Danes, 1974), that is, how thematic/
rhematic material is linked to the material which comes next in the text, and (ii) as
mode of development (Fries, 1983), that is, which material is typically used in the-
matic position.
There is little agreement among scholars as to the precise nature of Themes. Fries
(1983, p. 117) establishes two approaches to the denition of Themes: the com-
bining and the separating approach. Whereas combiners merge the concepts of
Theme/Rheme and Given/New, separators analyse the two sets of concepts as be-
longing to two dierent systems in the information structure of the clause. Halliday
and other scholars working in the systemic tradition represent this last approach,
whereas Prague School followers combine the concept of Theme as point of de-
parture with information focus.
Halliday (1985, p. 37) denes Theme as the element which serves the point of
departure; it is that with which the clause is concerned. This denition has triggered
a interesting and fruitful debate (G omez Gonzalez, 2001, p. 115) among those who
equate the spatial metaphor (the point of departure) and the matter metaphor
(that with which the clause is concerned), and those who adopt a dissociating
view and criticise Hallidays denition as involving two distinct notions which tend
to, but need not, coincide in one wording (Fries, 1983; Downing, 1991; Downing &
Locke, 1992). 1
1
Hallidays denition of Theme has led to a fruitful discussion. See Berry (1996) for dierent systemic-
functional views on Theme.
Thus, whereas according to mainstream systemic functional grammar in English

topical Theme is iconically realized by the element occupying clause-initial position,
extending up to and including the rst transitivity constituent in the clause (that is, a
participant, a circumstance or even a process), the defenders of the dissociating
view exclude from topical status certain clause initial constituents. In terms of the-
matic identication, this means that
1. in Hallidayan terms, everything after a marked theme (in a declarative clause,
any participant which is not realised as Subject, any circumstance or Attribute)
is rheme;
2. within the dissociating view, a marked Theme is followed by an unmarked
Theme. Thus, to identify topical Themes, Downing (1991, p. 125) suggests fol-
lowing Enkvist (1973) and treat as initial elements every constituent from the be-
ginning of a macrosyntagm up to and including the Subject of its main clause,
which allows topical Theme to be captured in a more systemic way.
It seems that the dissociating view is more in tune with what occurs in scientic
discourse. Most topical Themes in scientic and academic discourse in English tend
to be realised in the grammatical category of Subject (Gosden, 1993). Consequently,
in line with previous studies on the role of Theme in this type of discourse (Davies,
1988; Gosden, 1993; Mauranen, 1993), I have included the Grammatical Subject
(GS) as an obligatory element in thematic analysis, even in the sequence where the
Subject is preceded by other experiential material such as a marked contextualising
Adjunct. I shall here adopt the model of thematic analysis devised by Davies (1988),
later applied by Gosden (1993) to RAs, in which both GSs or GSs preceded by
Adjuncts or by other types of Context Frames (CFs) are considered as highlighting
the options for and constraints on thematic choices (Gosden, 1993, p. 60). Thus,
our analysis of thematic elements will run as follows: 2
Comitatives and instrumentals are characterized. . .

Grammatical subject
(Topical) THEME RHEME
In this study I suggest. . .

Adjunct Grammatical subject
Contextual frame (Topical) Theme
THEME RHEME
This conceptualisation of Theme is also acknowledged by Mauranen (1993) in

her dierentiation between orienting Theme and topical Theme. In her view,
2
Themes will be analysed only in main clauses since my intention here is to provide a clearer picture of
thematic patterns throughout RA abstract discourse.
orienting Themes are realised not only by connectors and Adjuncts, but also by
Complements, modal clauses and reporting clauses. This wider perspective of what
constitutes a Context Frame as part of the Theme will be adopted in the present
analysis, in which two angles of thematic analysis will be considered: thematic
progression patterns and methods of thematic development.
3.2.1. Thematic progression (TP)

The concept of thematic progression (TP) was rst introduced by Danes (1974) as
an organising principle that accounts for the ordering of information in discourse be-
yond the level of the sentence. The notion of TP is one of the constituent elements of the
wider functional sentence perspective (FSP) proposed by scholars of the Prague School.
Daness model is based on the Prague Schools combining approach to the de-
nition of Theme. He (1974, p. 114) denes TP as the choice and ordering of ut-
terance themes, their mutual concatenation and hierarchy, as well as their
relationship to the hyperthemes of the superior text units (such as the paragraph,
chapter,. . .) to the whole text and to the situation. In his view (1974, p. 111), it is
not R [Rheme] alone, but its connexion with the given T [Theme] that is commu-
nicatively relevant. He points out that R is signicant as the conveyor of New
information: it represents the core of the utterance and it pushes the communication
forward. However, being informatively insignicant, it is Theme that plays the main
constructing role. This is Daness most important contribution: to realise that a close
relationship exists between the thematic organisation of texts and their inner con-
nexity, a relationship in which Rhemes also have a part to play.
In Danes view there are three basic types of Thematic Progression:
(1) Single Linear TP. The Rheme of sentence 1 becomes the Theme of sentence 2.
Example from corpus
Results indicate that these learners of English employ different production
strategies based on the length of the coda; with feature change favoured for
single codas R1 . These modifications T2 can be explained by several linguis-
tic constraints including L1 transfer, markedness, and sonority, as well as by nat-
ural phonological processes, the latter of which appear especially salient of the
coda segments have been acquired. (Applied Linguistics 22/23, 2001)
(2) TP with a Continuous (or Constant) Theme. Here the same Theme appears in
a series of sentences, to which dierent Rhemes are linked.
Example from corpus:
This paper T1 assesses, and contests, the long tradition of attacks on the use
of invented sentences in language teaching. It T2 seeks to separate arguments
against them which rely on parody and ridicule, from more reasoned asser-
tions. (Applied Linguistics 22/23, 2001)
(3) TP with derived Themes, in which dierent Themes are derived from the same
hypertheme. (No examples found in corpus).
Previous studies of thematic progression in RAs (Nwogu, 1990; Nwogu & Bloor,
1991) reveal a tendency for abstracts to display both simple linear and constant the-
matic patterns. The studies by Nwogu (1990) and Nwogu and Bloor (1991) were based
on both professional and popular medical texts, and observed a high frequency of
both linear and constant themes in the two subgenres. As for the third type of thematic
progression pattern (derived themes), it was found only in the professional texts.
In the present study, Daness TP model is used as an analytical tool. The fact that
TP takes a combining approach to the concept of Theme might seem contradictory
with the separating approach adopted in the present study. However, as Hasan and
Fries (1995, p. xxxiii) state: [t]here are some fairly obvious grounds for suggesting
that the patterns of thematic progression concern only topical theme. Here, then, I
sustain a dissociating view, which for analytical purposes implies that the GS, being
the typical realisation of the topical Theme, is always included as part of the thematic
element. Moreover, as the present study is carried out in rather short texts, TP will be
addressed as operating between adjacent clauses.
3.2.2. The method of thematic development

The method of thematic development (Fries, 1983, 1995a, 1995b) can be dened
as the typical thematic selection, that is, material which is typically used in thematic
position. Fries (1995a, 1995b) suggests that patterns of Theme selection provide
important insights into the understanding of a specic genre.
The notion of typicality (the typical thematic selection) is approached by Fries
(1983) from two dierent but closely related angles: (i) transitivity, that is, which
participants are selected as Theme, and whether there are particular selections which
can be described as characterising a genre; and (ii) lexical selection: whether the
Themes are typically people, institutions, concrete objects, abstractions, etc.
Taking both Friess method of thematic development and our understanding of
Theme, we will be exploring this relationship from two dierent but complementary
perspectives:
1. The syntactic realisation(s) of Themes, both as sentence-initial GS or as GS pre-
ceded by a Context Frame (CF).
2. The GS domain (Gosden, 1993), that is, the semantic roles that GS encodes. In
my view, the adoption of a semantic perspective may contribute relevant insights
to the exploration of thematic choices.
In Gosdens view, there are four main domains or roles to be enacted by GSs:
(A) Participant domain, as
Discourse participant (we).
Participant viewpoint (our).
Interactive participant (citations).
(B) Discourse domain, as
Discourse event /process (e.g. interpretation, conclusion, argument, explanation).
Macro discourse entity (e.g. report, paper, thesis).
Micro discourse entity (e.g. gure, diagram).
Interactive discourse entity (e.g. previous studies).
Empty discourse theme (it/there).
(C) Hypothesized and Objectivized domain, as

Hypothesized viewpoint (e.g. the possibility of variations, the apparent
contradiction).
Objectivized viewpoint (e.g. a signicant dierence, the surprising feature).
Hypothesized entity (e.g. a simple model).
Empty Hypothesized and Objectivized theme (e.g. it is clear that).
(D) Real World domain, as
Mental process (e.g. calculation, comparison, analysis).
Real world entity (e.g. system, framework).
Real world event/process (e.g. the preparation of samples).
Empty real world theme (e.g. it was found that).
It is hoped that this double thematic analysis as TP and as method of thematic
development will allow us, on the one hand: (i) to establish the grammatical and
lexico-semantic features of the Themes in the corpus of RA abstracts of linguistics,
and (ii) to see whether dierences are found at those two levels which may sustain the
hypothesis that dierent types of abstract demand dierent thematic organisations.
3.2.3. Thematic analysis: TP patterns

In order to explore the relationship between thematic progression and rhetorical
organisation in abstracts of papers on linguistics, 10 abstracts of each main type of
structure found (IMRD and CARS) were analysed. The third, minor, type of
structure found was not explored, as its combinatory nature reects too much var-
iation to yield reliable conclusions. I am aware that the limited scope of the present
corpus study will only allow us to draw general conclusions. However, this rst
tentative attempt may serve as a starting point for future work in the eld.
3.2.3.1. TP patterns in informative abstracts (IMRD structure). The 10 abstracts

analysed which display an IMRD structure share certain distinctive features as re-
gards thematic progression patterns. For the sake of clarity, a distinction will be
made between what we here call TP within moves, and TP across moves.
(A) Within moves, we observe that some moves tend to be shorter than others, and
therefore, no TP patterns can be appreciated. This is the case of the Introduction
section, generally one-sentence long:
Example 1
(Introduction) This paper analyzes the discourse conditions governing the
choice between extraposition and nonextraposition of that-clause and inniti-
val-VP subjects. (Linguistics 39 (4), 2001)
The Methods section poses some identication problems in the discipline of
Linguistics, and, I would also say, in the soft disciplines in general, as what accounts
for method can range from theoretical issues to laboratory work. Secondly, it varies
in length, ranging from a long ve-sentence section to a mere Adjunct. This is the
case in:
Example 2
(Methods) On the basis of a large corpus of naturally occurring data, (Re-
sults) it is shown that non-extraposition requires that the content of the subject
be discourse-old or directly inferable. (Linguistics 39 (4), 2001)
Whenever the Methods section accounts for more than two sentences, the con-
stant TP pattern seems to be used within the move. This is the case in 5 of the 8
abstracts from the corpus where the Methods section is at least two-sentence long:
Example 3
(Methods) (1) The narratives were played to listeners at their unmodied rates
and at three computer-manipulated rates: (2) all passages were adjusted to the
Mean Mandarin rate, the Mean English rate, and a Reduced Rate, 10 per cent
slower than the Mean Mandarin rate. (Applied Linguistics 22 (3), 2001)
The longer move is the Results section, where a simple linear TP pattern pre-
dominates. This is the pattern found in 4 of the 8 abstracts whose Results section
consists of 2 sentences or more:
Example 4
(Results) (1) The most basic contribution of this study is the development of a
coding system that describes the types of responses elicited during
poetry reading. (2) Poetry reading is described as a close reading, meaning
construction task that involves high levels of close consideration, analysis
and elaboration of textual meaning. (Applied Linguistics 22 (3), 2001)
Example 5
(Results) (1) Results indicate that these learners of English employ
different production strategies based on the length of the coda; with feature
change favoured for single codas; epenthesis for two-member codas; and
absence for three-member codas. (2) These modifications can be explained by
several linguistic constraints, including L1 transfer, markedness, and sonority,
as well as by natural phonological processes, the latter of which appear especially
salient if the coda segments have been acquired. (Applied Linguistics 22 (3), 2001)
The Discussion section is again a shorter move consisting of one or two sentences
at most. Out of the 4 samples in which it consists of more than one sentence, the
constant TP pattern is used in three cases:
Example 6:
(Discussion) (1) This coding system reveals how non-native readers of poetry
notice form and consider the gap between input and output, thus extending
their understanding of the potential users and meanings of an existing linguistic
structure. (2) In addition; it shows how non-native readers view the distance
between the poems content and their own knowledge of the target culture and
thus nd their cultural awareness enhanced. (Applied Linguistics 22 (3), 2001)
and a simple linear TP pattern is found in just one instance:
Example 7:
(Discussion) (1) The paper ends with a discussion of the syntactic position
of nonextraposed sentential subjects and concludes that it cannot be the
same as that of fronted sentential complements. (2) This means that the com-
mon discourse properties of fronting and nonextraposition must be linked to
their common linear ordering properties, rather than to a common syntactic
position. (Linguistics 39 (4), 2001)
(B) Across moves, no clear pattern emerges. Both the simple linear TP and the
constant TP pattern are used, although there is a certain tendency for the linear
pattern to be used in the transition between Results and Discussion:
Example 8:
(Results) The most basic contribution of this study is the development
of a coding system that describes the types of responses elicited during
poetry reading. (Discussion) This coding system reveals how non-native read-
ers of poetry notice form and consider the gap between input and output, thus
extending their understanding of the potential uses and meanings of an existing
linguistic structure. (Applied Linguistics 22/23, 2001)
Example 9:
(Results) Regression analyses projected that the Mandarin-speaking list-
eners would prefer the same ideal rate for Mandarin-accented speech that
they did for native English speech; while the other ESL learners would
prefer Mandarin-accented English to be spoken at a rate slower than native
English speech but faster than the Mandarin speakers natural rate. (Discus-
sion) This result may reect a dierence in processing costs for familiar and un-
familiar accents. (Applied Linguistics 22/23, 2001)
3.2.3.2. Thematic progression patterns in indicative abstracts (CARS structure). The

indicative abstract has been found to display a distinct rhetorical organisation which I
have here called the CARS structure, as it coincides with the structure which Swales
(1990) observed in the Introduction of RAs. This global organisation consists of three
moves (Establishing a territory, Establishing a niche and Occupying a niche) which, I
will argue, show distinct patterns of thematic progression. As was the case with
informative abstracts, ten texts of the 36 abstracts selected, which conform to the
CARS pattern, have been analysed in depth in order to reach some tentative
conclusions which deserve further research. Again, conrming previous ndings about
TP patterns in abstracts (Nwogu, 1990; Nwogu & Bloor, 1991), in the abstracts
analysed two types of pattern seem to combine in the arrangement of information: the
simple linear type and the constant or continuous theme type. However, the way in
which these two types combine seems to depend on the move or section that the writer is
dealing with, and diers in certain ways from the combination patterns of IMRD
abstracts.
(A) Within moves, we again notice that there are some sections longer than others,
and clearer TP patterns can be observed. The rst move (Establishing a territory) is
the longest, with normally 2 or 3 sentences in each text analysed. The TP pattern
generally displayed here is the simple linear. This pattern is found in 8 instances out
of 10, one more displaying a constant TP pattern and the remaining one being only
one sentence long. The simple linear pattern suits this rst section in which the writer
claims some kind of central role for the subject or tries to back up his/her research by
referring to previous studies:
Example 10
(Establishing a territory) (1) This paper deals with one of the most produc-
tive word-formation patterns in German : -ung nominals denoting events;
such as Entdeckung or Erblingdung. (2) They are derived by attaching the
suffix -ung to a verb stem. (3) Not all of these derivations are grammatical,
a fact that has often been related to the inherent temporal properties (the Ak-
tionsart) of the base verb. (Linguistics 39 (2), 2001)
Move 2 (Establishing a niche) is generally too short to draw reliable conclusions.
In fact, in 7 abstracts, Move 2 is only one sentence long. However, whenever it
stretches across more than one sentence, that is, in the remaining 3 cases, the
combination pattern is constant TP:
Example 11
(Establishing a niche) (1) It can be easily shown, however, that verbs belong-
ing to one and the same Attionsart sometimes allow the derivation of an
-ung nominal and sometimes do not [. . .]. (2) For instance; die Erblind-
ung des Mannes the mans becoming blind; corresponding to the sentences
der Mann erblindet the man becomes blind, is possible, whereas *die
Erbluhung die Rose erbl uht the rose begins to blossom, is not. (3)
But the two verbs; erblindenand erbl uhen, belong to the same class of inchoa-
tive verbs. (Linguistics 39 (2), 2001)
Move 3 (Occupying the niche) also tends to display the constant TP pattern. This
is what happens in 5 out of the 10 instances:
Example 12
(Occupying the niche) (1) On the basis of these empirical facts mainly drawn
from examples of predicative possession the concept of bridging function
is introduced [. . .]. (2) Bridging functions serve the purpose of legitimizing/
facilitating combinations of functions that are otherwise incompatible with each

other. (3) The recurrent patterns and the concept of bridging function are dis-
cussed with reference to recent hypotheses on the nature of possessions and the
relationship that holds between classes of functions. (Linguistics 39 (2) 2001)
(B) Across moves, the linear TP pattern tends to be used between move 1 and 2:
Example 13
(Establishing a territory) Comitatives and instrumentals are characterized
by markedly different behavior as to their combinability with additional
features [. . .]. (Establishing a niche) However; if expressions of predicative
possession involve a marker that also has the function of encoding the
instrumental; such combinations of functions generally involve the comitative
as well. (Linguistics 39 (2) 2001)
However, transition from Move 2 to Move 3 constitutes some kind of change of

direction, and the focus is here on the discourse itself (This paper) or on the discourse
participant (I/we).
Example 14
(Occupying the niche) In this paper; we discuss this paradox, illustrate the dicul-
ties that can arise, and then suggest a procedure whereby communicative purpose
can be retained as a viable and valuable concept. (Applied Linguistics 22 (2), 2001)
Example 15
(Occupying the niche) This paper aims to (re)conrm the original analysis.
(Journal of Linguistics 37, 2001)
3.2.3.3. Summary of ndings. When comparing the TP patterns displayed in each

type of structure, and although no well based conclusions can be drawn from such a
small corpus, we can observe a tendency for the linear patterns across the Results/
Discussion section in IMRD abstracts, and for the same pattern to be used in CARS
abstracts, with the exception of Move 3, which constitutes a complete change of
viewpoint. Within moves, each structure and each move within the structure display
a dierent combination: whereas no clearcut pattern can be observed in the IMRD
structure, and linear and constant TP patterns combine within sections, in the CARS
structure a more consistent distribution of patterns seems to be displayed: linear for
the rst move and constant for Moves 2 and 3.
3.2.4. Thematic analysis: method of thematic development

As explained above, two aspects will be analysed in this section: the syntactic re-
alisation of the Themes and the Grammatical Subject (GS) domain (Gosden, 1993).
3.2.4.1. Method of thematic development in informative abstracts (IMRD structure)

Syntactic realisations of theme. Theme is generally realised as the GS (unmarked
Theme), especially as far as the Introduction and the Methods sections are con-
cerned. The other realisation is that of CF + GS. Within the concept of CF
(marked Theme), we here include the empty it construction, mainly used in the
Discussion section (it is concluded, it is argued), where, in this way, real world
entities are given focus, avoiding the introduction of a discourse participant
(I/we).
It is in the Results and in the Discussion section where the sequence CF + GS tends
to be widely used. In both sections, CFs are mainly subordinate clauses and textual
Themes organising discourse:
Example 16
(Results) While signifying within an instrumental domain such as record-
keeping or accounting; they may simultaneously participate in the mise-
en-scene of conversational action or serve as metaphoric graphic gestures.
(Journal of Pragmatics 33, 2001)
Example 17
(Results) On the contrary; sentences invented by a teacher for a specific
context may have advantages, which are less easily attained by the use of at-
tested examples. (Applied Linguistics 22 (3), 2001)
Example 18
(Discussion) While this approach prevents the researchers from using the
narratives simplistically as an objective source of ethnographic data, it al-
lows for a complex, theoretically and sociohistorically informed, investigation
of social contexts of language learning and of individual learners trajectories
[. . .]. (Applied Linguistics 22 (2), 2001)
Example 19
(Discussion) In addition, it shows how non-native readers view the dis-
tance between the poems content and their own knowledge of the target
culture and thus nd their cultural awareness. (Applied Linguistics 22 (3),
2001)
GS domains. We observe that the text opens in the Introduction section with GSs
within the Discourse domain (e.g. This study, This paper, Researchers, The paper,
etc.). Then, in the three subsequent sections, the Real-World domains predominate
(e.g. Three participants naturalistic production, The behaviour of discourse OK, These
modications, Holistic students, Sentences, etc.). Therefore, there is a transition from
a more text-related perspective to a more objective, external-world related view (see
Abstract 1 above for illustration).
3.2.4.2. Thematic development in indicative abstracts (CARS structure)

Syntactic realisations of Theme. There is a clear dierence between sections as far
as thematic realisations are concerned. Whereas in Section 1 (Establishing a territory)
it is GS which predominates, in Section 2 (Establishing a niche), it is the sequence
CF + GS which is normally used, with the CF being generally realised by the sen-
tence connector however. This sentence connector functions as a marker of the gap or
lacuna found in the research or tradition presented in the previous section:
Example 20
(Establishing a niche) However; the claim that aL is a complementizer also has
been widely challenged. (Journal of Linguistics 37, 2001)
Other textual Themes used as markers of contrast have also been recorded:
Example 21
(Establishing a niche) Nevertheless; currently three main approaches to dis-
course anaphora can be identied [. . .]. (Journal of Pragmatics 32, 2001)
In the nal section (Occupying the niche), the sequence CF + GS also predomi-
nates, where CF is generally realised by an Adjunct which refers to the discourse
proper (In this paper, In this article) or by a textual Theme:
Example 22
(Occupying the niche) In this paper; we argue that bu is not a clitic-like element
and claim that it is a focus-sensitive operator. (Linguistics 39 (2), 2001)
Example 23
(Occupying the niche) Furthermore; it is argued that -ung nominalization pre-
serves the event structure of the underlying verb but makes it accessible for
modication on the syntactical level. (Linguistics 39 (2), 2001)
GS domains. As far as GS domains are concerned, we observe a ow from the

Real world domain in the rst move (e.g. Welsh, The Chinese negation marker bu,
Comitatives and instrumentals, etc.) to the more visible participant domain in Move 3
(I, we). In certain abstracts GSs as Real World domain are preceded, in the rst
move, by CFs (reporting clauses). The transition section can either refer to discourse
or to the real world. (See Abstract 2 above for illustration.)
Example 24
(Establishing a territory) Huang 1988assumes that bu must cliticize onto
auxiliaries/modals or the following verb [. . .]. Unlike Huang; Ernst 1995
argues that bu is proclitic on the adjacent word. (Linguistics 39 (4), 2001)
3.2.4.3. Summary of ndings. As can be seen in Table 2, the IMRD structure shows
the higher percentage of GS as Theme (59 themes out of 72 themes), and the dis-
tribution of GS or the sequence CF + GS depends on moves (see Table 3): GSs are
more frequent in the Introduction section (14 out of 15 themes), and the sequence
CF + GS in the Results section (5 out of 20).
In the CARS structure (see Table 2), the distribution of GS and the sequence
CF + GS is quite even (38 versus 34, respectively). The sequence CF + GS predom-
inates in Moves 2 and 3, especially in the former (12 out of 16 themes), whereas
Move 1 tends to display GS as Theme, with no CF (22 out of 30 themes)
(see Table 4).
As for the realisation of CFs, in the IMRD structure, they are mainly textual
Themes (On the contrary, however, at the same time) and also subordinate clauses. In
the CARS structure, CFs are Adjuncts and textual Themes (in Move 2, character-
istically, the adversative sentence connector however).
As far as the GS domain is concerned, the IMRD pattern situates the perspective
in the discourse itself (Introduction) and then moves onto the real-world domain in
the subsequent sections. Sometimes, a move back to the discourse can be recognised
in the nal section.
In the CARS structure, the participant domain predominates as discourse par-
ticipant in Move 3 (I/we) and as Interactive participant (citation) in Move 1;
otherwise the Real-World domain perspective is taken in the rst move. As a result,
Table 2
GS CF + GS Total number
IMRD 59 (81.9%) 13 (18.1%) 72
CARS 38 (52.7%) 34 (47.3%) 72
Table 3
IMRD GS CF + GS Total number
Introduction 14 1 15
Methods 19 3 22
Results 15 5 20
Discussion 11 4 15
Total 59 13 72
Table 4
CARS GS CF + GS Total number
Move 1 (Establishing a territory) 22 8 30
Move 2 (Establishing a niche) 4 12 16
Move 3 (Occupying the niche) 12 14 26
Total 38 34 72
in this structure the discourse participant is more visible and more interactive, in
contrast to the IMRD structure, which is more objective and external to the par-
ticipants involved in the interaction.
4. Concluding remarks
This paper set out to analyse the global organisation of RA abstracts of linguistics
as a rst stage in the exploration of the textual mechanisms which lay under the
construction and understanding of this complex genre. With this purpose in mind,
thematic analysis was applied, as a potentially useful tool in exploring genre, since
thematic choice is signicant as part of a writers available linguistic resources and is
one means through which writers can achieve local and global discourse goals. In
this sense, both thematic progression patterns and the methods of thematic devel-
opment were analysed in the corpus selected.
On the level of rhetorical structure, this small scale study has shown that, even if
the majority of the abstracts of papers on linguistics analysed display the rhetorical
structure commonly accepted to be the canonical global organisation of abstracts,
one which mirrors the organisation of the RA, a signicant percentage, about one-
third of the samples, display a dierent structure, one which matches the organisation
of the Introductory section of RAs: the CARS structure (Swales, 1990). Moreover, an
analysis of the moves indicates the existence of a minor rhetorical organisation which
mixes both types, which we have here called the combinatory type. As I see it, the
three types full three dierent functions, generally acknowledged for the RA ab-
stract: the informative, the indicative and the informativeindicative function.
The fact that there seem to be distinct structures fullling dierent functions may
explain why previous studies did not agree on what the rhetorical organisation of
abstracts is and described it in very exible terms.
The analysis of rhetorical structure carried out following Swaless (1981, 1990)
model of move structure was then backed up by a thematic analysis of the two
main types of rhetorical structures found: the IMRD and the CARS type of ab-
stract. The exploration of Theme, which focused on the analysis of TP patterns and
of method of thematic development (studied at the level of the syntactic realisations
of Themes and of GS domains), yielded interesting results. To start with, the
variation found in TP patterns sustains our perception of distinct rhetorical
structures. The starting point for this part was previous research in the eld
(Nwogu, 1990; Nwogu & Bloor, 1991) which had revealed that RA abstracts tend
to combine simple linear and constant TPs. Even though my corpus was too small
to draw reliable conclusions, the data reveal a certain consistency in the TP patterns
underlying each type of rhetorical organisation. Thus, the ndings conrm the
combination of two main types of patterns (simple linear and constant), and a
claim can be made that this combination is not random. On the contrary, the
dierent sections or moves which constitute each type of abstract seem to display a
distinct combination of thematic patterns, both within the moves or across
boundaries between moves.
As far as the method of thematic development was concerned, even if the un-
marked Subject was demonstrated to be the most typical realisation, it was also
shown how a more complex picture emerges when both Context Frames (CFs) and
GS domains were included in the analysis. As far as CFs are concerned, it was shown
how they are not distributed at random but they tend to concentrate on certain
moves. If we understand CFs as the material which orients readers in their journey
along the text (Mauranen, 1993), we may conclude that writers intervention is more
visible in one type of structure (CARS), and in certain moves in each structure
(Results and Discussion in IMRD and Move 2 and 3 in the CARS structure). The
analysis of GS domains contributed to this perception revealing how writers guide
readers through dierent domains depending on the type of structure we are dealing
with. Thus, in the IMRD structure, the starting point is the discourse itself (Intro-
duction), but the perspective is immediately changed to the Real world domain
(Methods), a focus which tends to be maintained for the rest of the text. In the
CARS structure, on the contrary, the journey is made along the opposite route:
generally the focus is rst placed on the Real world domain (Move 1) and ends in the
Participant domain (Move 3). Applying Gosdens (1993) cline of visibility-invisi-
bility, from the most visible (Participant domain) to the least visible (Real-World
domain), we can conclude that dierent patterns of what we here call textual dy-
namics appear: whereas in the IMRD abstract the writer, as a participant persona
tends to hide behind real-world entities and processes, in the CARS structure, the
writer chooses to present himself/herself as a visible participant in the research
community. These ndings reect dierent positions that writers can adopt when
presenting their research through abstracts in the discipline of Linguistics: they may
opt for presenting their papers as the investigation of a relevant real world problem,
accepted as such by the community or, they may position themselves as active
agents, main actors in a play in which interaction with the reader takes place. In my
view, and following Hylands (2000) suggestions, these two divergent positions may
be explained in terms of dierent traditions. Hyland states that in the soft disci-
plines there are less clearly paths of research and topics are more diuse than in the
hard disciplines, and so writers are in need of devising ways of engaging readers and
persuading them to read further, which can account for the more interactional
attitude of the CARS structure. Linguistics, it seems, draws on both traditions.
The fact that two types of textual dynamics have been found in a corpus of texts
belonging to the same genre and to the same discipline demonstrates that abstracts
have a complex generic prole which deserves further investigation in which com-
parison with other disciplines or even with other types of abstract (e.g. conference
abstract) can cast a light on a common core of dening features for this genre.
Appendix A
1. An empirical study of the relationship between cognitive style and the use of
communication strategy. Applied Linguistics 22 (2), 2001.
2. Genre identication and communicative purpose: A problem and a possible so-

lution. Applied Linguistics 22 (2), 2001.
3. Language learning memoirs as a gendered genre. Applied Linguistics 22 (2), 2001.
4. Texts, corpora, and problems of interpretation: A response to Widdowson. Ap-
plied Linguistics 22 (2), 2001.
5. The grammatical and lexical patterning of Make in native and non-native stu-
dent writing. Applied Linguistcs 22 (2), 2001.
6. Linguistic constraints on the acquisition of English syllable codas by native
speakers of Mandarin Chinese. Applied Linguistics 22 (3), 2001.
7. The philosopher pulled the lower jaw of the hen. Ludicrous invented sentences
in language teaching. Applied Linguistics 22 (3), 2001.
8. The task of poetry reading and second language learning. Applied Linguistics 22
(3), 2001.
9. What speaking rates do non-native listeners prefer? Applied Linguistics 22 (3),
2001.
10. Categories of the noun phrase in Jarawara. Journal of Linguistics 36, 2000.
11. Indicators and procedures: Nevertheless and But. Journal of Linguistics 36, 2000.
12. On the distribution of resumptive pronouns and Wh-trace in Welsh. Journal of
Linguistics 36, 2000.
13. Whats What in French questions. Journal of Linguistics 36, 2000.
14. Be going to: An exercise in grounding. Journal of Linguistics 37, 2001.
15. Positional eects in a monostratal grammar of German. Journal of Linguistics
37, 2001.
16. Temporal dependency and the syntax of subjects. Journal of Linguistics 37, 2001.
17. The morphosyntax of WH-extraction in Irish. Journal of Linguistics 37, 2001.
18. Why are categories adjacent? Journal of Linguistics 37, 2001.
19. Lucky Talk in celebrating the Chinese New Year. Journal of Pragmatics 32,
2000.
20. What makes you think you exist?: A speech move schematic and its application
to Pinters The Birthday Party. Journal of Pragmatics 32, 200.
21. Discourse anaphora: Four theoretical models. Journal of Pragmatics 32, 2000.
22. Semantic elds and frames: Historical explorations of the interface between lan-
guage, action, and cognition. Journal of Pragmatics 32, 2000.
23. Speech act metaphor in theatre. Journal of Pragmatics 32, 2000.
24. Can I be with?: Negotiating play entry in a bilingual school. Journal of Pragmat-
ics 33, 2001.
25. Discourse ok revisited: Default organization in verbal interaction. Journal of
Pragmatics 33, 2001.
26. Interaction by inscription. Journal of Pragmatics 33, 2001.
27. Retrieving talk from the simple past and the present progressive on alternative
radio. Journal of Pragmatics 33, 2001.
28. A complementation approach to Chinese passives and its consequences. Linguis-
tics 39 (2), 2001.
29. On the event structure of -ung nominals in German. Linguistics 39 (2), 2001.
30. Some truths and nontruths about even if. Linguistics 39 (2), 2001.
31. The principle of pragmatic detachability in borrowing: English-origin discourse

markers in Pennsylvania German. Linguistics 39 (2), 2001.
32. To be with X is to Have X: Comitatives, instrumentals, locative and predicative
possession. Linguistics 39 (2), 2001.
33. Discourse constraints on (non)extraposition from Subject in English. Linguistics
39 (4), 2001.
34. Suppletion, lexical meaning, semantic primitives, and translation data. Linguis-
tics 39 (4), 2001.
35. The Chinese negation marker bu and its association with focus. Linguistics 39
(4), 2001.
36. The dimensions of topic-comment. Linguistics 39 (4), 2001.
References
ANSI (1979). The American standard for writing abstracts. New York: ANSI Publication.
Berry, M. (1996). What is theme?. In M. Berry, C. Butler, R. Fawcett, & G. Huang (Eds.), Meaning and
form: Systemic functional interpretations (pp. 164). New Jersey: Ablex Publishing Company.
Bhatia, V. K. (1993). Analysing genre: Language use in professional settings. London: Longman.
Cremmins, E. T. (1982). The art of abstracting. Philadelphia: ISI Press.
Crookes, G. (1986). Towards a validated analysis of scientic text structure. Applied Linguistics, 7(1), 57
70.
Danes, F. (1974). Functional sentence perspective and the organisation of the text. In F. Danes (Ed.),
Papers in functional sentence perspective (pp. 106128). Prague: Academia.
Davies, F. (1988). Reading between the lines: Thematic choice as a device for presenting writer viewpoint
in academic discourse. Especialist, 9(1/2), 173200.
Day, R. A. (1988). How to write and publish a scientic paper. Cambridge: Cambridge University Press.
Downing, A. (1991). An alternative approach to theme: a systemic-functional perspective. Word, 42(2),
119143.
Downing, A., & Locke, P. (1992). A university course in English grammar. London: Prentice-Hall.
Dudley-Evans, T. (1994). Genre analysis: An approach to text analysis for ESP. In M. Coulthard (Ed.),
Advances in written text analysis (pp. 219228). London: Routledge.
Enkvist, N. E. (1973). Theme dynamics and style: An experiment. Studia Anglica Posnaniensia, 5, 127
135.
Fries, P. H. (1983). On the status of Theme in English: Arguments from discourse. In J. S. Pet o & E.
Sozer (Eds.), Micro and macro connexity of texts (pp. 116152). Hamburg: H. Buske.
Fries, P. H. (1995a). Themes, methods of development, and texts. In R. Hasan & P. H. Fries (Eds.),
Current issues in linguistic theory: Vol. 118. On subject and theme. A discourse functional perspective (pp.
317359). Amsterdam: John Benjamins.
Fries, P. H. (1995b). A personal view of theme. In M. Ghadessy (Ed.), Thematic development in English
texts (pp. 119). London: Pinter.
G
omez Gonz alez, M. A. (2001). The themetopic interface. Evidence from English. Amsterdam: Benjamins.
Gosden, H. (1993). Discourse functions of subject in scientic research articles. Applied Linguistics, 14(1),
5675.
Graetz, N. (1985). Teaching EFL students to extract structural information from abstracts. In J. M. Ulijn
& A. K. Pugh (Eds.), Reading for professional purposes. Methods and materials in teaching language (pp.
123135). Amersfoot: Leuven.
Halliday, M. A. K. (1985). An introduction to functional grammar. London: Arnold.
Hasan, R.& Fries, P. H. (Eds.). (1995). On subject and theme. A discourse functional perspective.
Amsterdam: John Benjamins.
Hyland, K. (2000). Disciplinary discourses. London: Longman.
Jordan, M. P. (1991). The linguistic genre of abstracts. In A. Della Volpe (Ed.), The seventeenth LACUS
forum 1990 (pp. 507527). Lake Blu, IL: LACUS.
Lewin, B. A., Fine, J., & Young, Y. (2001). Expository discourse: A genre-based approach to social science
research texts. London: Continum.
Mauranen, A. (1993). Theme and prospection in written discourse. In M. Baker, G. Francis, & E. Tognini-
Bonelli (Eds.), Text and technology. In honour of John Sinclair (pp. 95104). Amsterdam, Philadelphia:
John Benjamins.
Nwogu, K. N. (1990). Discourse variation in medical texts: Schema, theme and cohesion on professional and
journalistic accounts. Nottingham: Department of English Studies, University of Nottingham.
Nwogu, K. N., & Bloor, T. (1991). Thematic progression in professional and popular medical texts. In E.
Ventola (Ed.), Functional and systemic linguistics (pp. 369384). Berlin: Mouton de Gruyter.
Paltridge, B. (1994). Genre analysis and the identication of textual boundaries. Applied Linguistics, 15(3),
288299.
Swales, J. (1981). Aspects of article introductions. Birmingham: The University of Aston.
Swales, J. (1990). Genre analysis. English in academic and research settings. Cambridge: Cambridge
University Press.

Ventola, E. (1994a). Abstracts as an object of linguistic study. In S. Cmejrkova, F. Danes, & E. Havlova
(Eds.), Writing vs. Speaking: Language, text, discourse, communication (pp. 333352). T ubingen:
Gunter Narr.

Ventola, E. (1994b). Problems in producing scientic abstracts in L2 . In S. Cmejrkov
a & F. Sticha (Eds.),
The Syntax of sentence and text (pp. 283303). Amsterdam, Philadelphia: John Benjamins.
Rosa Lores Sanz is a Junior Lecturer in the Department of English Language and Literature at the
University of Zaragoza in Spain. She has an M.Phil. in translation from the University of Salford (En-
gland) and a Ph.D. in translation and pragmatics from the University of Zaragoza. She has published
several articles on the applicability of the systemic-functional grammar perspective to translation and
contrastive linguistics in general.

On RA Abstracts From Rhetorical Structur

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

On RA Abstracts From Rhetorical Structur

Hochgeladen von

Copyright:

Verfügbare Formate

ENGLISH FOR

On RA abstracts: from rhetorical structure

Dpto. de Filologa Inglesa y Alemana, Facultad de Filosofa y Letras, Universidad de Zaragoza,

2. The abstract as a distinct genre in academic prose

3. The analysis of data: materials, methods and ndings

3.1. About rhetorical structures

An analysis of the rhetorical organisation of the 36 abstracts selected was carried

This organisation displays some kind of combinatory structure, which, in my view,

Thus, whereas according to mainstream systemic functional grammar in English

Comitatives and instrumentals are characterized. . .

In this study I suggest. . .

This conceptualisation of Theme is also acknowledged by Mauranen (1993) in

3.2.1. Thematic progression (TP)

3.2.2. The method of thematic development

(C) Hypothesized and Objectivized domain, as

3.2.3. Thematic analysis: TP patterns

3.2.3.1. TP patterns in informative abstracts (IMRD structure). The 10 abstracts

3.2.3.2. Thematic progression patterns in indicative abstracts (CARS structure). The

facilitating combinations of functions that are otherwise incompatible with each

However, transition from Move 2 to Move 3 constitutes some kind of change of

3.2.3.3. Summary of ndings. When comparing the TP patterns displayed in each

3.2.4. Thematic analysis: method of thematic development

3.2.4.1. Method of thematic development in informative abstracts (IMRD structure)

3.2.4.2. Thematic development in indicative abstracts (CARS structure)

GS domains. As far as GS domains are concerned, we observe a ow from the

2. Genre identication and communicative purpose: A problem and a possible so-

31. The principle of pragmatic detachability in borrowing: English-origin discourse

Das könnte Ihnen auch gefallen