<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffcc" text="#000099">
<font face="Palatino Linotype"><br>
Chris' question about the concepts of 'genre' and 'register' is to the
point, because these notions are often ill-defined, vague or confused.
Let me summarize my view of these notions:<br>
<br>
<big><b><font color="#990000">Register</font></b></big><br>
<br>
One line of research that has often used the notion of 'register' is,
indeed, Systemic Grammar, and apart from references to Michael Halliday
himself, there is much work by Jim Martin, and others. See, e.g., <br>
</font>
<blockquote><small><font face="Palatino Linotype">Leckie-Tarry, H.
(1995). Language & Context. A functional linguistic theory of
register. Edited by David Birch. London: Pinter </font></small><br>
</blockquote>
<font face="Palatino Linotype"><br>
Another -- more quantitative -- line of research on register is by
Biber & Finegan, see e.g.,<br>
</font>
<blockquote><small><font face="Palatino Linotype">Biber, D., &
Finegan, E. (Eds.). (1994). Sociolinguistic perspectives on register.
New York Oxford: Oxford University Press. </font></small><br>
</blockquote>
<font face="Palatino Linotype">Here registers are defined by relative
frequencies of (mostly) grammatical structures, such as the use of
various pronouns, tense, etc. For instance, the pronouns 'I' and 'you'
are -- obviously-- used more often in informal conversations than in
scholarly articles, whereas past tenses are more typical of news or
storytelling, than in weather forecasts, etc. Clusters of such
(relatively frequent) properties define a register -- and such
registers may be associated with specific genres, as I did in the
examples. However, they only define a frequent (but seldom exclusive)
grammatical pattern of such genres. Strictly speaking, thus, they are
merely abstract configurations of grammatical properties of discourse,
which one sometimes tries to give a name, e.g., a 'narrative' register,
etc. - which however should not be confused with that of genre or text
type (see below).<br>
<br>
<font color="#990000"><b><big>Genre</big></b></font><br>
<br>
Genre, on the other hand, is currently mostly defined in terms of a
kind of social practice (see, e.g. Fairclough, etc.), rather than as a
'type of text or talk'. Thus, informal conversations, parliamentary
debates, news reports in the press, poems or TV soaps, are among the
hundreds of genres <i>conventionally</i> distinguished and used, as
such, in interaction and communication. Traditionally, genres tended to
be described in terms of their discursive structures -- including their
grammatical structures (register) - but such an account is at best
incomplete - while often neither sufficient, nor necessary. <b>Rather,
genres should largely be accounted for in terms of the properties of
their <i>context</i> rather (or not only) in terms of their 'textual'
properties.</b> Thus, what defines a parliamentary debate as a genre,
as a social practice, is its Setting (Time, Place, etc.), its
participants (MPs, etc.) and participant roles (government, opposition,
party membership, etc.), the specific -- political -- actions performed
(e.g., present a motion, etc.), its specific goals, as well as the
knowledge and other social cognitions (e.g., politcal ideologies) of
the participants. And although parliamentary debates has a few unique
discursive (register) properties, such as fixed formulas ("my
honourable friend", etc.), a formal register, strict turn taking
regulation (by the Speaker), and preferred (political) topics, among
others, few of these are exclusive for parliamentary debates, and at
most (proto) typical - because most of them also define other formal
meetings. For details on this important role of context analysis, see
my new, forthcoming book on Context and Discourse with Cambridge
University Press (2008). <br>
<br>
Indeed, a more cognitive approach to genres is in terms of <i>prototypes</i>
(a notion investigated by Eleanor Rosch). For detail see the entry
"Prototype Theory" in Wikipedia. <br>
<br>
My current view on genre would be that it is one of these complex
notions that need a complete - <i>multidisciplinary - theory,</i>
which would involve (a) a characterization of context properties (b) a
description of prototypical gramamatical and (other) discursive
structures, (c) a cognitive theory of the production, comprehension and
memory aspects of the actual use of genres by language users, (d) a
social theory of the social situations, participants, etc in which
genres are used (not the same as a context theory!), (e) an
anthropological theory of the uses and variations of genres in
different cultures, and (f) a historical theory of the diachronic
changes of genres, among others. <br>
<br>
Hence a theory of register is merely a small (linguistic, grammatical)
aspect of a theory of genre. <br>
<br>
<big><b><font color="#990000">Text type</font></b></big><br>
<br>
To further complicate matters, we may want to maintain the notion of
'text type', as a more restricted discursive characterization of
(fragments of) discourse, such as narration, argumention, exposition,
etc., defined in terms of (schematic, semantic) structures, such as the
canonical schematic structure of a story as defined by Labov &
Waletzky in 1967 and beyond (see the special of <i>Narrative Inquiry</i>
of 30 years of research on such narrative structures (1997) - or on the
other hand, the work on argumentation structures (e.g. by Van Eemeren,
et la.). One example is my characterization of the canonical structure
of news reports in the press (see, e.g. <i>News as Discourse,</i>
1988). However, these are NOT (complete) genre descriptions (which are
multidiscplinary, see above) but only descriptions of the abstract
(schematic, superstructural) structure of the 'format' of (parts of)
discourses. For instance, within the <i>genre</i> of an informal
conversation (or a parliamentary debate, or a life story) we may, of
course, have narrative and argumentative structures -- besides
conversational, interactional structures -- and not vice versa: we do
not have 'genres' within genres.<br>
<br>
<b><big><font color="#990000">Style</font></big></b><br>
<br>
And finally, besides genre, register and text type, there is the notion
of 'style', which is even more elusive, and that also deserves a
multidisciplinary approach. The more linguistic approach would
typically define style as the result of (mostly surface structure,
grammatical: phonological, visual, lexical or syntactic) variation to
express more or less the same local or global meaning, as controlled by
different properties of the context (formal vs., informal style, etc.).
But also here a broader notion might be useful, because there are
different ways of news repòrting (see the tabloids vs. the "quality"
press), different ways of telling stories, that are not (only) covered
by grammatical variation. Thus, the crucial defining notions of style
are, e.g. (a) variation - where something remains the 'same' (b) social
context - as the condition for variation, and (c) personal, social or
situational uniqueness, etc. -- hence the style of a person, a
movement, etc. Part of this definition of style can be given in terms
of register, as explained, namely the combination of the choice and
frequency of grammatical resources. This is also why style and register
are often confused.<br>
<br>
This is, in summary, how I would define these notions. Crucial is, in
my view, NOT to limit description to linguistic grammars (as has been
an early tendency of systemic grammar and other linguistic approaches),
but to include all kind of discursive structures at all levels
(grammar, style, rhetoric, schematic structures, pragmatics,
interaction, etc.), a context analysis, and the cognitive and
sociocultural approaches as we know them from the ethnography of
speaking, anthropological linguistics, etc. <br>
<br>
I hope this is clear... <br>
<br>
Warm regards to all.<br>
<br>
Teun<br>
<br>
________________________________________<br>
<br>
Teun A. van Dijk<br>
Universitat Pompeu Fabra<br>
Dept. de Traducció i Filologia<br>
Rambla 30<br>
08002 Barcelona<br>
<br>
E-mail: <a class="moz-txt-link-abbreviated" href="mailto:teun@discourses.org">teun@discourses.org</a><br>
Internet: <a class="moz-txt-link-abbreviated" href="http://www.discourses.org">www.discourses.org</a><br>
<br>
<br>
<br>
<br>
</font>
</body>
</html>