Date: Sun, 25 May 1997 23:10:09 +0000
From: "full-name" <email>
Subject: Re: Spam - (And not the lunch meat)
Mike:
In response to your comments about spam, I must remind you that what I posted was written by the editor of the Red Rock Eater News Service. Those were not my words, but not far off my own opinions.
>First, let's differentiate between targeted ads and "spam."
No. Let's not. Spam in my mailbox is unrequested advertising. In certain cases I have volunteered to receive promotional material from a company (Some software companies offer updates on their stuff that I like to hear about.). I do not object to receiving this because I asked for it. However, if I order a book on Tibeten Rifle Shooting from amazon.com and then receive an advertisement from them about Tibeten Archery, it is spam, it is evil, and I don't like it.
>I operate a phone coaching service, and have no ethical problem with presenting it in newsgroups where it is relevant. For example, since I know quite a bit about weight-loss psychology, I put ads targeted precisely towards that end in an ng as alt.diet-support. I get the usual feedback to the effect that I am Evil.
You are. But that sort of evil is so common on unmoderated newsgroups that I have been driven from them completely. Not reading the newsgroups has pretty much solved the problem for me. My question has not to do with evil, but are you making any money?
>As I see it, targeted ads are not "spam."
As I see it they are and I will support legislation to stop it.
>"Spam" is, as I understand it, ads that are not precisely targeted to the unmoderated newsgroup or mailing list in question.
See above.
>(Moderated newsgroups or mailing lists exist at the behest of the owner or moderator...so they are in totally separate categories).
In the case of mailing lists, this is not really true. We at Netdynam are not moderated, yet I can kick people off and filter them so that they cannot resubscribe. In the case of someone repeatedly trying to sell a product here, I might very well do so. On Netd, of course, you would have the option of gathering enough support to get me unelected as listowner. On most unmoderated groups you don't even have that option.
>There should never be guilt in responding to targeted ads or to spammers.
Yes there should. Lots of it. (Notice we are trading pronouncements. Is that arguing, discussion or just posturing. Inquiring minds want to know.)
>If the cost is low to email, as Simon mentions, that is the joy of the Internet.
No, it's a curse upon the internet.
[...]
>We use them. Let us save
>our vitriol for the things in life that count.
I agree with this somewhat. I will choose the social battles I want to fight. I choose not to put much effort into this one. You might choose not to put much effort into trying to get homeless alcoholics off the booze. So we differ in where we do our community service. Nevertheless, I will encourage those fighting spam just as they would probably support me.
Methinks.
Simon
[sig file]
-------
Thus, each post can either be considered to be one text -
since they
are each created and sent by one list-member in one chunk - or they can
be considered to be comprised of a number of texts of varying types,
resulting in a type of exchange complex, whose boundaries happen to be
technologically signalled. In terms of Bakhtin's (1986: 72ff)
discussion regarding the boundaries of the utterance, "changes in
speaking subject" are clearly signalled within each post:
However, in the case of the email post, the signals of these changes within posts are managed by the writer, mainly through formatting - a matter of control over the expression plane of mode - and thus the conditions and situations of communication can be regarded as providing for the interpolation of the Other as a case of 'manifest intertextuality' (Fairclough 1992: 117ff), rather than as a signal of the "finalization of the utterance" (Bakhtin: 76), which Bakhtin notes as one of the boundary conditions of a complete utterance. Therefore, following Bakhtin and others (c.f. for example, Stubbs 1996: 32), the complete utterance, or complete text-unit, is considered to be bounded by a change in 'creating' subject, as well as by the various signals of 'finalization' that posts typically employ (see below 1.6.2.1). Thus the post itself is considered as the primary text unit for analysis.
The model that I have developed looks at posts from both perspectives
mentioned above: firstly, from that of the internal organisation of
each of the basic discourse units (e.g. Turns, Stages, phases (or
‘Parts’); c.f. also discussion Mod 2: II) and the
idea that each post may be comprised of a series of turns and shifts of
register, and that writers will mark the boundaries of these text units
using a variety of signals; and secondly, from the perspective of the
post as complete utterance, as "primary text", one which makes
reference to other texts and contributions in more or less overt ways,
and locates itself in the context of an ongoing conversation:
My objective in developing this model is to account for the means by
which posters adopt, and creatively reproduce, what I have previously
referred to as the ‘group norms of interaction’
(Don 1997)
by which members of the email list attempt to make their
meanings intelligible to the other listmembers using only graphic
channel means. My investigation is therefore concerned with:
As argued in Mod 2: I[note 1], resources of the graphic channel (Hasan 1985) have not previously formed the main communicative means for group practices to develop via what Hasan terms ‘process sharing’ (opcit: 58). An email list provided a way of investigating features of process sharing in a written-only context, such as the way in which posters regularly incorporate the previous posts of other listmembers, the way in which posters stage their texts, and the ways in which attitudinal meanings functioned to both signal staging and to signal affiliation with respect to other listmembers. A further related objective was to reach conclusions about the ways in which posters creatively manipulate these norms of interaction as they set about constructing for themselves, by means of this communicatively restricted graphic channel, discursive identities or personas.
It is my position that ‘the post’ represents a type of speech genre, what Bakhtin (opcit: 78) refers to as the "relatively stable typical forms of the construction of the whole"(italics in original). A linguistic analysis of the representative texts in this study, and my active participation in the interaction of the list has allowed me to identify a number of ‘relatively stable typical forms of construction of whole posts’, and to note that listmembers use these forms as flexible templates or conventionalised social practices in which they manage their identities as textual personae. This aspect of textual identity, or what I term poster identity, is taken up in more detail under a discussion related to the regular stylistic features used by poster identities in the later sections (i.e. Chapters 4 & 5[note 2]) of the original thesis.
In describing the typical post, rather than adopt the term 'generic structure potential' mentioned above, instead, I use the term rhetorical organisation potential (first proposed in Module 2), for a number of reasons. The first relates to the applicability of the term rhetorical as an attribute of text organisation: most of the contributions to the group discussion forum I see as organised by argument in the service of identity maintenance. In other words, the social purpose of all of the contributions is to maintain identity - to explore difference and similarity (alignment/disalignment or affiliation/disaffiliation) with respect to others. The means by which list-members explore difference or construct solidarity is through argument, using a form of expository discourse - by positioning themselves and others in relation to ideological value systems, via reference to ideologically-charged tropes such as family, work, aesthetics, religion, gender, and so on. These types of text belong to the category of what Martin (1985) describes as expository writing, both hortatory and analytic, in which writers' social purpose is defined as persuading to and persuading that respectively (opcit: 17). In the service of these arguments, the use of evaluation and the evocation of attitude remains an obvious rhetorical device. In order to characterise the use of these rhetorical devices, I introduce a set of 'tenor variables' below in 1.3.1.1 (see Fig 1.1).
Secondly, rather than adopt the term 'structure', which has overtones of rigidity and obligatory elements, 'organisation' in this paper refers to a looser sequence of text events. It refers to a general tendency of posts to be sequenced, or ordered, in typical or conventional ways, which in turn allows for noticeable marked behaviour. This marked behaviour - when it occurs - tends to function as an overt signal of attitudinal stance - even if no explicit evaluative lexis is present due to the way conventions tend to set up expectancy of typical sequences: to break with convention is a means of calling attention to that act. Furthermore, while 'structure' connotes a somewhat static view of text creation and interpretation , 'organisation' allows for a more 'mutable' conception of a generic prototype, one which can accommodate recursivity, and transition phases or boundary conditions between stages or segments, rather than strict boundaries between the stages of any text (see for example Hunston 1989). As will be discussed again below, the relationship between the various stages or segments of texts, and the boundaries of the texts themselves, is conceived of as layered by frames of coherence (a term I first introduced in Don 1997: c.f. below 1.4), and these 'layers' can be partially accounted for by differences between constituency and dependency relations. In other words, 'coherence' in the context of an email post to a list can be described as a matter of embedding at one layer, and as a matter of reference and sequence at another. The conception of 'layering', 'framing', or 'tracks' owes much to the notion of metacommunicative, logical levels of analysis prevalent in cybernetics (c.f. Bateson 1972, Wilden 1980, Lemke 2000) and in Systemic Functional Linguistics (henceforth SFL) to account for the fact that features at one level of analysis contribute to meaning at the next level 'up', and rely on sequencing in order that these meanings be made and interpreted. It also owes its conception to the work of Goffman (1981), particularly his notion of footing, referred to again below (1.4.1). These perspectives can be used both to account for texts as organised wholes after the event, or as sequences of textual events which make their meaning by reference to ever wider intra- and intertextual chains from a more dynamic perspective of text as utterance (or reading event) . From this perspective as well, insights into the nature of the "reading event" provided by Sinclair and others (e.g.1993, Hunston 1989, 2000, Francis 1994) - particularly the notion of the signalling of coherence on both the autonomous and interactive planes of discourse have influenced my investigation into how both reference and discourse markers (analogous to signals of Engagement under the appraisal framework) contribute to the articulation of a text's organisation. Bhatia (2004) also outlines a view of genre in which the description of genre can be 'characterized at various levels of generalization' (p. 59), and his approach can be usefully compared with the one adopted in this paper. Both these approaches are discussed again below.
The use of the term 'genre' appears problematic for discourse analysis, and particularly for the data I use for several reasons. The notion of genre has been widely used as a term to cover a range of linguistic entities in both literature and language studies during the past decades, including that of email interaction (c.f. Gruber 2000). It has been adopted to account for the unfolding of text types in SFL in particular, and linked in this regard to the nature of the social purpose for which the text was created (see for example Martin 1985, 1992, 1997, Christie & Martin 1997, Eggins & Martin 1997, Martin & Veel 1998). At the same time, the use of the term 'genre' is itself contested within SFL and some scholars, notably Hasan (1995, 1999), argue strongly that the term should be used to refer to "text types" rather than to a separate level of discourse function, in contrast to Martin (op cit) and others who prefer to conceive of genre as a separate discourse level subsuming Register. Its use in systemics in general seems partly reflected in Bakhtin's use of the term, particularly as presented in the translation of a collection of his later works (opcit 1986, and under the name Voloshinov 1986). At the same time, Martin's conception of genre in SFL can be traced mainly to Labov's (1992, Ch 9) work on narrative (see also Bamberg (ed) 1997), while Hasan's use of the concept can be traced to the work of the Firthian linguist, Mitchell (1957, cited in Eggins & Slade 1997). The use of the term 'genre' is also prevalent in folkloristics and ethnomethodology (e.g. Briggs & Bauman 1992, Hanks 2000), as well as literacy pedagogy (e.g. Christie and Martin 1997, Martin 1985, Dudley-Evans 1986, 1997, Freedman & Medway (eds) 1994, Swales 1990, Hyland 2000, Lewin et al 2001, Burns & Coffin (eds) 2001, Bhatia 2004, Ravelli and Ellis (eds) 2004, Lewin, Fine and Young 2001), and many linguists and researchers alike use it as a term of pre-theoretical convenience (e.g. Biber 1988). This chapter uses the theoretical notion of genre in the process of outlining the framework for analysing written interaction based on the analysis of texts produced in a specific mailing list. This is because the description of the organisation of email posts at first glance suggests that they could be classed as forming a type of genre, but I would prefer to retain the term genre (or more specifically, core-genre) - what Bhatia (opcit) refers to as generic value, and Bakhtin as primary speech genre - to refer to more regular, abstract cases of textual organisation, and in this sense, my use of the term follows that of Martin (1985, 1997). Because the main tools of analysis I use are informed by SFL, reference to genre in this thesis will assume this theoretical perspective. In addition, I have also used approaches which do not explicitly make reference to 'genre', such as that of Goffman (1974, 1981) in order to provide another perspective for analysing and interpreting the data used.
Fundamentally, I view 'genre' as a function of the dynamics of its
texturing (Fairclough 2003: 100ff), and thus it can both account for,
and take into account the legitimated options for action that any
participant can undertake at any given juncture of the unfolding of the
discourse. In a similar fashion, genre figures in what SFL treats as a
realisation relationship - with "the ways in which field, mode and
tenor variables are phased together in a text" (Martin 1997: 12). This
view of texturing I see as analogous with my own term rhetorical
organisation potential (ROP) which, as acknowledged above, makes
reference to Hasan's (1985) term generic structure potential. In the
case of specific institutional practices, which include the
activity-sequences legitimated within a 'community of practice' (Lave
& Wenger 1991, Eckert & McConnell-Ginet 1998), the
conceptualisation of genre is put at risk by issues relating to mode,
and some of these were addressed in Mod 2: I. Therefore, 'genre' needs
to be characterised at a variety of levels of delicacy by reference to
both linguistic (synoptic) and discursive (dynamic) orientations, which
take into account the development or contestation of the norms of the
group's discursive practices over time. While this view of genre is not
necessarily new (e.g. Miller 1994, Bhatia 2004), the application and
integration of this view in analysing the discourse of communities of
practice - outside academic and commercial discourse communities at
least - has not been developed to any great extent (recent work on
pedagogical contexts e.g. Coffin, Painter and Hewings 2005 an
exception). On the other hand, while not specifically genre-related,
Ho's 2002 work represents an in-depth study of website mediated
communication, what she terms electronic discussion forums (EDF), and
many of the areas of investigation in her study parallel areas of
concern in this thesis. Ho distinguishes between the various forms and
approaches to the study of this type of discourse, and goes on to
examine in detail the nature of the discourse evident in the EDF in
which she was involved. Her work focussed in particular on the
construction of a particular Singaporean identity within the texts in
her study. Using a slightly different perspective, Hård af
Segerstad (2003) analysed in detail the features of email mediated
communication discourse in the context of an overview of several other
forms of CMC discourse. Both works provide detailed surveys on the
literature addressing analysis of CMC texts, and therefore such a
review is not attempted here. However, while both of these works
address areas of similar concern to that of this thesis, they do not
focus on the development of an approach and a model which may be
adapted to the investigation of generic conventions of a wide range of
written interactive texts. As Ho (opcit: 30) observes:
The aim of this thesis therefore is to contribute to this research with
a particular emphasis on the organisation of a social community's
prototypical text-units, where I pay special attention to the ways in
which attitudinal meanings appear at regular intervals and thus help to
signal staging in texts (for which I have adopted the term
‘attitudinal prosodies’). This leads in turn to the
specification of what I am calling its rhetorical organisation
potential.
In describing the fundamental unit of analysis in my study,
the post, I have adopted
the term macro-genre to refer to the 'relatively stable typical forms
of the whole utterance' (Bakhtin 1986: 78), signalling that the actual
'content' of any post may be comprised of any number of core-genres such as narrative, recount, exposition, and so on. In the context of
this thesis, 'macro-genre' refers to texts which incorporate what
Fairclough (2003: 69) refers to as "the socially available resource of
genres in potentially quite complex and creative ways." Core-genres, in
contrast, are resources at an abstract level of linguistic and social
organisation, recognisable by a wider range of community members than
the institutional or community group which develops its local
conventions of practice, and which lead to the development and
negotiation of macro-genres in the sense I am using the term here.
Couture (1986: 82, quoted in Swales 1990: 41) provides one fundamental
definition for the analysis of [macro]genre which accords with the
approach taken in this thesis: "[genre] specifies conditions for
beginning, continuing, and ending a text".
Eggins and Martin (1997: 236) represent the notion of a [core]genre as
"the way the types of meaning in a text co-occur [is] a pattern typical
of a particular genre", and continue "the sequence of functionally
distinct stages or steps through which [a text] unfolds". They note
that "linguistic definitions of genre [identify] 'relatively stable
types' of interactive utterances [and] define genres functionally in
terms of their social purpose" (ibid).
Martin elsewhere (1997) describes the functional parameters of genres
more precisely, and in doing so paradoxically acknowledges that the
boundaries between genre 'types' are not precise, and that genres are
better located in a topology, rather than a typology. What this means,
is that "core genres" can be recognised within texts which are of
completely different types, if, for example, mode is the variable - as
I am proposing here.
In order to locate the recognition criteria for genres in the actions
of participants, rather than the nature of the texts themselves, Swales
proposes the notion of a discourse community which he defines by means
of 6 characteristics (1990: 24-27). The mailing list under
investigation would appear to fit the definition, and with reference to
the notion of the negotiation of macro-genres posited above, his
criterion 4 seems especially relevant: "newly-emergent groupings [need]
to settle down and work out their communicative proceedings before they
can be recognized as discourse communities. If [a new group] 'borrows'
genres from other discourse communities, such borrowings have to be
assimilated."
Bhatia (2004: 149) however, contrasts the notion of discourse community
with that of community
of practice. He makes the distinction between
the texts and genres that enable the community to maintain their
communicative purposes for 'discourse communities', and the values and
practices which hold a community together for 'communities of
practice'. In Module 2: I the term 'community of practice' was adopted
in order to describe the texts analysed as products of the practices of
the email list participants, and for the same reason, in Don 97:
2.2-2.7 the activities of the list were described in terms of
Halliday's (1985: 44ff) levels of context. This is because prototypical
texts need to be considered as a part of the wider social contexts
within which they function as units of meaning.
At the same time, Bhatia's (ibid) criteria problematise the
categorisation of the products of this online group, since these
products appear to be classifiable under the two types of community he
distinguishes and noted above. For this reason, following Bhatia,
(macro)genre users have been re-classed as members of a discourse
community rather than a community of practice, especially since members
of any discourse community can also be members of other discourse
communities by virtue of their control (c.f. Figure 1.1 below) of a
variety of (macro)genres, while being at the same time, members of
different communities of practice depending on their participation and
recognition in that community. Furthermore, with respect to the email
group in this study, while many long-term members are affiliated by the
values and the practices of interaction developed within the group over
time, they each also belong to widely differing
(cultural, ethnic and national) communities and hold widely different
values on many issues. Therefore, the classification of participants as
members of a discourse community acknowledges that it is the products (i.e.'texts') of this group that are the focus of this study, rather
than the
participants themselves, even when their 'textual identity' becomes a
means for distinguishing between them in Chapters 4 and 5[note 2].
These factors have led to the term 'macro-genre' being chosen to refer
to the prototypical text-units of this email-mediated discourse
community as it acknowledges their status as 'generalisation', and
distinguishes them from objects known by such terms as idealisation,
universal template, schema, etc.
In terms of the levels of generalisation outlined by Bhatia (2004) the
term macro-genre which I use in this thesis is differentiated from his
notion of genre colony by its derivation from both analysis of products
(texts), and my participation as a member of the community of
practice/discourse community who uses this means of communication.
Because the identification of genre is contested in the literature, and
complicated by the appearance of genres in a variety of mediated forms,
the next section addresses some of the issues relating to mode and
genre in more detail.
Hasan (1999: 253ff) discusses at great length the problem of identifying the boundaries of stages within any text, and relates this to the interface between register and context as it is conceptualised in SFL. If register is the textual realisation of Context of Situation, then any change in register, whether it be of field, tenor or mode, also signals a shift of context, and hence engenders an internal text boundary. In Mod 2: II, this type of shift was also discussed by reference to Gregory's (1985) conception of texts being articulated by phases, which correspond to changes in context signalled by changes in register, and is supported by Cloran's (1994, 1995, 1999) work on the conceptualisation of context within SFL. For Hasan, one of the problems attending the notion of genre concerns the identification of boundaries or stages in text structure, and relates to the location of what SFL refers to as rhetorical mode, and whether it is related to a specific register variable - field, tenor, or mode.
As stated above, the definition of core-genre adopted by Martin and
others within SFL, incorporates the notion of social purpose - for
example, to persuade, to report, to explain, and so on - but
traditionally, within Systemics, this aspect of a text's functionality
has been subsumed under rhetorical mode, or 'the part language is
playing', along a continuum of ancillary - constitutive (referred to
again in the following section). Thus, rhetorical mode has lately been
considered as helping to construe mode due to
its reference to the
material activity which accompanies the ancillary, whereas Hasan (1999)
argues that the feature 'social purpose' attending rhetorical mode
reinforces her contention that it remain a matter of field.
Martin and
others regard 'social purpose' as helping define a level of discourse
realised by register (i.e. genre in Martin's conception), at a
different level of abstraction. For Hasan, this provides for
problematic contradiction within SFL. At the same time, if genre -
whether core or macro - is conceived of as a level of abstraction
realised by a variety of layers or tracks interrelated to signal shifts
rather than strictly demarcated boundaries - as between stages in a
text - then such contradiction might be seen instead as part of the
normal flexibility of language.
Section 1.4.2 below outlines this notion of over-layering of tracks in
more detail, arguing that such layers can be attended to in turn by the
analyst in order to look at a number of ways that writers organise
their discourse. Over-laying is not necessarily done consciously, but
writers must be somewhat aware that incorporating a degree of
redundancy of coherence markers on several tracks (especially in such a
mode sometimes described as written 'conversation') will
lead to greater 'uptake of illocutionary force'. 'Purpose' or goal in
this view is not located in the intention of the individual writer, but
in the wider social goals that genres have developed to serve. Genre is
therefore an abstraction, incorporating the idea that over-layering
contributes to a another level of text organisation, which is not
merely a combination of sequenced types of registers, and which may
conceivably appear within a number of different 'text-types' as
outlined below. Before returning to the specification of the features
of these tracks or layers in the texts, in the following sections I
discuss distinctions of register, genre and text-type in a little more
detail.
This sub-section addresses the way in which patterns of co-occurrence
of textual, interpersonal and experiential values relate to notions of
register and genre with attention to Hasan's arguments
(in particular 1995 and 1999) that register alone can account for such
patterning.
Hasan’s argument referred to in the previous section,
suggests that the practice within SFL of describing one aspect of a
"contextual configuration" (Hasan 1985: 55ff) by reference to a
continuum which treats verbal action as a matter of [ancillary] v.
[constitutive] mode (1999) would be better seen as part of the system
of field,
which is "concerned with specifying the nature of social
activity" in which both verbal and material actions are implicated
(Hasan 1999: 282). My claim is that the texts in my study are entirely
constitutive of their context, but whether this constitutiveness
relates to their field or their mode appears to be at issue. For me,
this problem is intimately related to the argument over whether we can
classify sets of texts as realising 'genre's or whether, as Hasan
maintains, SFL's conception of rhetorical mode as an aspect of field is
sufficient to account for the development of 'argument', or the
'relatively stable forms of organisation' in texts which might signal
that they were related. She argues that examples of 'verbal action'
such as "defining, explaining, generalising, narrating, lecturing,
persuading…" (opcit: 276) are accounted for by field (rather
than genre), and goes on to state that mode is
better viewed as limited
to matters of contact (opcit: 282), or what I have been referring to as relative interactivity (see discussion Mod 2: I[note 1]). The
definition of
'contact' as a system relating to degree of involvement construing tenor in SFL (see Martin 1992: 528ff, Poynton 1985: 77) is thus also
thrown into question, but has to do with Hasan's conception of
'contact' as applying to "activities which just cannot be performed
with verbal action ALONE: they call for material action" (opcit: 276,
emphasis in original).
In Mod 2: I[note 1], I outlined a set of dimensions along which stretches of
email text(s) could be located in terms of mode, and
argued that the
description offered could be encompassed by the notion of "relative
interactivity". As mentioned above, this term refers to phenomena
similar to Hasan's (1999: 282) use of the term "degree of immediate
contact", so Hasan uses the term ‘contact’ in a
different way from Poynton and myself, and this needs to be made clear
for the purposes of the following discussion.
In Mod 2: I, I argued that, since the context for email list
interaction was in some senses entirely constitutive of itself, i.e.
the written contributions are the context, then the issue of the
constitutive-ancillary dimension was rendered somewhat irrelevant for
the description of mode in that context. This means that the fact that
these texts were originally transmitted and responded to in email, has
nothing to do with their rhetorical mode, and hence is not related to
their field. On the other hand, Hasan maintains that because the
purpose or goal of a speaking or writing act may be described by
reference to rhetorical mode, and since this concerns the nature of the
verbal activity (i.e. field), there is no need to posit a separate
level of discourse called genre which purports to do the same thing. I
return to this point again in section 1.3.2 below when a distinction
between genre and text-type is proposed. Meanwhile, the next section
addresses the notion of contact/familiarity referred to above in
order to clarify its use in descriptions of tenor, rather than mode.
Figure 1.1 below, based on Poynton (1985), outlines a set of tenor dimensions as resources for construing dis/alignment and dis/affiliation in terms of skills, knowledge, common assumptions, shared values and so on.
Figure 1.1: Tenor variables for construing writer - reader alignment and affiliation
For convenience, I have grouped these 'variables' into categories based on dimensions originally proposed by Poynton (opcit), in order to illustrate resources that writers use in negotiating identity - or rather, one of the resources that discourse analysts may use to make accounts of such positioning practices. I present them here because my discussion needs to make reference to a variety of these 'tenor dimensions', both as a means of discussing strategies that construe positioning, and hence the negotiation (over time) of 'identity' pertaining both to writer/speaker and members of the projected audience, and as highlighting the function of tenor in realising the rhetorical organisation potential of texts. These tenor dimensions and their means of activation (such as 'amplification') are discussed and exemplified in more detail in Chapter 3[note 2] where the notion of textual identity is explored. In addtion, the derivation of the macro-genre itself incorporates realisations of tenor, and these can be described by reference to subsystems of appraisal - Engagement, Attitude and Graduation - which are in turn dependent on the interrelation of all lexico-grammatical metafunctions.
The conception of contact as a variable of tenor within SFL can also be
shown to redound with features relating to field. For example, consider
the principle of contraction (Fig 1.1 above), in which
The degree of contact will also determine the relative degree to which
"the realisation of the meanings selected has to be more [or less]
explicit" (Martin 1992: 531). This principle appears to parallel the
principle of 'specialisation' within accounts of field:
‘specialised' lexis is generally a feature of specific fields
of discourse, but such specialisation can only be used when the tenor
of the register - i.e. the positioning of ideal readers - assumes high
involvement: "the higher the degree of involvement […] the
more exclusive the semiosis" (Martin: ibid). Consider an instance when
the topic discussed enters a field where expert [status: authority]
(c.f. above Fig 1.1) is required; in order that use of specialised
lexis does not enact a communicative obstacle, the positioning of ideal
readers must also be one of high [contact: involvement] - in other
words the principle of reciprocity is assumed because speakers need not
explain their specialist terms in detail. This may have the effect of
closing off communication to those not familiar with the jargon or
'exclusive' lexis employed by members of a particular group. In other
words, I am arguing that specialist lexis construes a particular field
(perhaps an expert one) and that such a field implicates a particular
tenor relationship (one of some significant degree of shared knowledge,
that is to say, of involvement). On the other hand, interlocutors could
have a similar knowledge of an ‘expert’ field while
being entirely non-reciprocal in terms the other resources they have
available. In this sense, while ‘specialisation’
relates only to one, relatively narrow aspect of tenor, and not to
other aspects of ‘contact’ or
‘solidarity’, it nevertheless highlights the nature
of the interrelationship between features of register and context.
My representation of tenor variables in Figure 1.1 does not make
reference to the subsystems of CONTACT outlined in Poynton's original
(opcit: 77) for reasons discussed in Mod
2: I[note 1], section 2.1 pertaining
to levels of abstraction and different orders of register. I would
maintain that the
interpersonal discourse semantic is a textual construct, not a
description
of 'material situational setting' as Hasan calls it, nor context of
culture. Such factors influence the production and reception of a text
and its meanings, but cannot be 'read off' any text. This point is
entailed in the next section 1.3.4 when a distinction between 'genre'
and 'text-type' is proposed.
In summary, any description of the tenor of the text also relies on a concurrent description of their realisations in field and mode. Issues of co-positioning and how they may be accounted for present problems for the analyst when looking at lexicogrammatical realisations of tenor, since this process of inter-subjective co-positioning in turn appears to be dependent to some extent on resources associated with both field and mode at the same time, which has been hinted at in the previous discussion. Thus, tenor cannot be analysed independently using the resources typically associated within SFL with the interpersonal metafunction alone. My analysis indicates that the realisation of inter-subjective co-positioning relationships (a matter related to the text’s tenor) is not solely achieved via interpersonal meanings, but that textual, experiential and logico-semantic meanings are also implicated. Halliday has allowed for this idea, and it is supported by arguments made by Thompson, especially1999.
For similar reasons, I maintain that the appraisal framework cannot be
defined as a system of resources pertinent only to interpersonal
metafunctional realisations. Many of the categories of appraisal are
realised by a combination of linguistic resources and strategies (see
for example categories of invoked attitude, Figure 1.2 below) which
represent a range of metafunctional resources. Instead of locating
appraisal within tenor, therefore, it would be better viewed as a
system of classification that describes evaluation in general, a system
located at a level constrained by and ranging across the resources of
register: the
discourse semantic. For this reason, appraisal can be a
valuable means of revealing rhetorical prosodies in any text which are
built up by positioning moves. Such prosodies are one means for tracing
the development of each text's organisation. The proposed prototypical
staging of the texts in
my study have relied to some extent on attempting to track such
prosodies via attitude analysis, particularly invoked attitude.
In this section I outline reasons for proposing that the term text-type be applied to texts defined by reference to the materiality of their mediation, i.e. their production, distribution and reception, leaving the term genre to be determined by features other than mode, or other than by field alone. Just as the register variable field is foregrounded in the construal of genre (as suggested by Hasan's 1999 argument), the register variable mode is foregrounded in the construal of text-type. This does not suggest that a description of genre is exhausted by reference to field, just as a description of text-type is not exhausted by reference to features of mode. Moreover, a description of text-type in the sense I am advocating here refers to features outside the verbal text itself and therefore cannot be synonymous with mode, and in the same way, a description of genre is not synonymous with register, let alone any one aspect of it. On the other hand, it does provide means for a distinction between a genre such as 'hard news report', and the text-type of its 'utterance' such as newspaper, radio bulletin or web page. A song may be recorded in the studio, and then reproduced and distributed on vinyl, CD or tape. Its eventual transmission and reception may have a variety of actual contexts, but typically the text's utterance is going to be constitutive of its verbal action, with channel phonic and medium 'spoken'. Similarly, the core-genre 'exposition' encompasses a great variety of sub-genres, one of which is the academic discourse act - such as lecture, presentation, examination submission or dissertation. The social purpose of exposition may be glossed as persuading or arguing. With this in mind, it can be seen that a variety of text-types may be represented by the rhetorical mode activity of 'persuading', and that the social purpose pertaining to argument and persuasion may be realised by a great variety of generic sub-categories. The distinction proposed here is aimed to obviate the confusion of genre and text-type.
The corollary of my distinction with regard to email texts is that mode is construed by the textual realisation of the actual technological
mediation in which the texts were both created and distributed. Thus,
'email post' and '[various formatted] styles of email post' would be
labels pertinent to text-type, while their 'content', or 'social
purpose', would defined by reference to conventionalised staging -
their argument and social purpose as revealed in the meanings they each
make - as a macro-genre, which in turn, might incorporate as many
culturally-conventional, abstract core-genres as was available to each
email-using discourse community. It can be easily recognised that the
actual textual realisation of such macro-genres would not be
theoretically restricted to email - there is nothing to stop such
macro-genres from appearing in any material form. In other words, the
constraints of the technological mediation tend to promote such textual
realisations in interaction, rather than to produce them.
Therefore, to restate the observation made earlier, this thesis regards genre as a function of texturing, which in turn, I argue has as much to do with the resources construing tenor as with those which construe field, and this approach means that the nature of the rhetorical positioning developed within each text is based on the deployment of the resources of the interpersonal as well as the ideational (experiential and logical) metafunction in texturing, and hence organising the argument of the text. In addition, the mode, construed by features of formatting described in Mod 2: I in tandem with lexico-grammatical resources such as thematic development also figure in the texturing of a text, and hence its generic organisation.
Thus, the use of evaluative positioning, and strategies relating to
signalling alignment (related to Goffman's 'footing' discussed in more
detail below 1.4.1) are relevant for the analysis of the rhetorical
organisation of any text, and in hypothesising from examples the
rhetorical organisation potential of such texts in general -
specifically, the prototypical macro-genres associated with particular
institutionalised activity sequences (Martin 1992: 537ff) - from any
community of practice studied. In Mod 1[note 3], for example, I argued that
resources usually seen as construing that part of register known as field - the representation of social actors and their relative agency
as construed by the resources of the transitivity system in SFL -
should also be considered as contributing to the positioning of
audience and the organisation of the text's arguments, and hence the tenor of the text. From the perspective of the metafunctional resources
which realise field, tenor and mode, it might seem as if the textual
and logical metafunctions would be more heavily implicated in creating
texture - and hence in construing the mode and text-type of any text,
but this paper takes the position that while mode is a defining
feature of text-type, and that the resources of the textual and logical
metafunctional construe mode, interpersonal and experiential meanings
are just as defining of the texture of any text, and that texture in
this sense is more constitutive of 'genre' than of 'text-type'.
One recent study of the textuality of email messages (Gruber, opcit)
used the notion of 'genre' in order to present a clear description of
the nature of 'scholarly e-mail messages' in terms of their use of
theme and intertextuality. This indicates that Gruber also regards
activity-type and field as foregrounded in the construal of genre.
Although the article remarked on the interpersonal metafunctional
features of the texts, within the scope of his report it was necessary
to focus on a limited range features in order to demonstrate its
claims, and so it provided little support for my own contention
regarding the significance of the resources of tenor. However, as
stated earlier, I prefer to reserve the term genre, or core-genre, for
more abstract texturing conventions, and to employ the label macro-genre for institutionalised and typical arrangements of text
events via the use of 'rhetorical organisation potential' which is
realised in a prototypical macro-genre.
The notion of genre adopted by Gruber
(op cit) owes its conceptualisation to Fairclough (1992) and differs slightly
from that proposed by Martin (1992) although the two are related. Like Gruber,
the approach that I use is dependent on the insights of both Critical Discourse
Analysis (CDA) including the work of Fairclough, as well as SFL, particularly
the work of J. R. Martin and Ruqaiya Hasan - despite their differences over
the definition of genre. Another prominent genre analyst, Bhatia (2004: 60),
conceives of what I am calling core-genre as generic
values, preferring this term due to what he claims as a lack of 'any
specific textual sequencing' of rhetorical acts. I maintain that such core-genres
do have recognisable sequencing, and that many of them are incorporated into
the macro-generic organisation of email posts as interdiscursive elements, their
inclusion signalling an assumption that the audience will recognise their core-generic
status (see for example texts [jvs84.20/rob] and [jvs228.56/stan] which employ narrative and limerick sequences respectively
In the following section, the relationship of 'generic staging' to
local norms or conventions is discussed, and the issue of the
identification of the boundaries of text-units is addressed once more,
with reference to the theoretical perspectives already introduced above.
In Don 1997 I proposed that this email-mediated group of interactants
or discourse community (the term I favoured then was derived from
Hymes' 1974 notion of speech community) could be described and
interpreted using a set of 'frames of coherence', which I saw as cues
or signals operating at a variety of discursive levels in the texts
which realised its context. This thesis takes up the same notion of
frames of coherence, and uses it to describe a model developed by
analysis and participant-observation which identifies the typical
patterns of generic staging within the specific written community under
focus. I propose that this model also has a more general application
for investigating interactive written discourse, and for developing
prototypes for other discourse communities. The framework provides a
means for accounting for the social practices of a specific 'speech
community' (and other related email-mediated groups) characterised by
their development of a set of norms or conventions, and the means by
which members signal alignment, or footing in Goffman's terms (1981:
128ff):
In the above quotation, Goffman makes reference to a liminal role,
which is involved in signalling higher level interaction. In the
framework I have developed, it is the notion of layers at different
levels of abstraction to which the term 'frames of coherence' refers.
At each layer, framing operates via signals or cues which are used by
writers to indicate, and by readers or analysts to interpret the
changes in footing which may indicate boundary conditions of the
generic stages of a typical contribution to interaction. As indicated
above, as well as the notion of footing, framing also refers to another
insight of Goffman's (1974: 210ff ), that 'directional cues', or
metalinguistic signals (Hanks 2000:177), take place in a variety of
'tracks', 'channels' or 'layers' in any text. It is in the complex
over-laying of these tracks that the framing and hence organisation of
argument and positioning of any post/text is accomplished. For this
reason, part of the analysis of texts in this study concentrated on
what I have termed "markers" at Layer Three (see below 1.5.4). These
markers subsumed those elements in a discourse which signal conjunction
or comment on the unfolding of the discourse in any way.
Each group, whether it is conceived of as a community of practice or
discourse community, develops its own conventions for framing their
contributions at each layer or track of interaction, and participation
in a written speech community mediated by email is therefore no
different in this respect than in any other mode of interaction. In
other words, as argued earlier, once text-type is held constant, the
matter of speech genre in Bakhtin's (opcit) terms, becomes more
relevant for determining how the conventions of any group have
developed in order to legitimate their practices, and control identity.
For this reason, parts of the thesis used findings based on the
main list in the study, for comparison with findings using excerpts
from the interaction of another similar list.
The specific written community which provided the main source of data
for this study, was introduced in Modules 1 and 2: the textual
activities of a specific email discussion list and its members. In one
sense, this thesis was conceived of as a case study of some of the
norms of interaction which have become apparent during the course of
the list’s history, and which ‘realise’
its context. However, the underlying aim of the research was to develop
a reliable and sturdy model with which to approach other types of text,
and similarly account for their interactive, dynamic orientation, or
the nature of their dialogicality.
Identification of text-units, or units of analysis, is one of the
fundamental issues in text analysis - and most especially in the texts
created in this mode of interaction due mainly to the multi-logical
nature of participation. (c.f. Mod 2: I [note 1]).
As indicated above, it is the nature of the signalling of (sub)text
boundaries which indicate the staging of any text and hence its
internal organisation - as well its relationship to other texts. What constitutes the boundary of any text/speech event
(e.g. turns, stages, or phases) in this mode, I characterise as a
matter of both level of abstraction, and the signalling or reading (or
both) of sequencing in a variety of ways. By ‘level of
abstraction’ I mean that I determine boundary conditions by
means of observing ‘interference patterns’ when
features observed at a variety of Layers are
overlaid. For example, units may be formed by the types of relation
outlined in Hoey (2001 referring to Winter's work), such as sequence
(or logical), and matching relations. One way in which such matching
relations are signalled is through negation.
Hunston's (1989) analysis of her corpus of scientific articles addresses the same issues as those tackled here, namely that evaluation and rhetorical positioning are the main elements which help organise a text: "…evaluation organises all the meanings in the text into a coherent entity" (opcit: 71). Her thesis involves analysis of moves which signal text structure by looking at features on three main dimensions: Status, Value, and Relevance. Because the texts she examined are members of the core-genre [expository: analytic] in Martin's (1987) terms, the three dimensions she identifies could also be applied to the texts examined here since many of them incorporate an expository core-generic orientation. However, because the register, and in particular the field and mode of discourse of our samples differ, the nature of the evaluative strategies differ also. The use of the appraisal framework for this thesis also means that the perspective on the use of evaluation differs. While appraisal (see Mod 2, Part II ) looks at types of attitude and means of engagement separately, Hunston's model (opcit) groups such indicators of each dimension together. Furthermore, because the targets of evaluation in Hunston's texts were non-animate, certain types of attitude under appraisal - such as those pertaining to human behaviour and emotion (Judgement and Affect) would not be relevant for her texts. Many of the evaluative resources in Hunston's texts would be classified as subcategories of Appreciation under appraisal, while identification of features via sub-dimensions of Status, Value, and Relevance would be attended to as systems of Engagement under appraisal. At the same time, my study examined linguistic resources which also appear in Hunston's framework such as markers, and the notion of prospection . These resources help signal both evaluative orientation, and text unit boundaries.
Some of the means for determining text boundaries at what I am calling
'Layer 1' (c.f. 1.5.2 below) were presented and discussed in Module 2:
I. The framework presented here relies on this earlier discussion, in
particular the description of 4 fundamental styles of text formatting,
and the 3 dimensions of analysis constrained or enabled by the
technological mediation of the texts (Mod 2: I, section 3.3 [note 1]). In this
sense, frames of coherence can be seen most clearly as functioning to
re-contextualise each new contribution to the discussion, realised as a
post to the list. The 4 fundamental styles or text-types of response
outlined in Mod 2 are reproduced and glossed for convenience here:
In addition, there is one further style of text-type whose defining
features overlap somewhat with those at the next Layer (2) 'down' (see
section 1.5 below). This text-type is not formatted as a response to
any previous contribution (although in theory all posts are responses
to previous contributions) and is classed as:
As the description suggests, this style most often realises what at
Layer 2 is termed an Initiation.
What I have called frames of coherence (or 'relevance') for example,
are, of course, intrinsically related to the topics chosen by
contributors for argument or discussion. In a coherent, or easily
readable text, writers will most often put the main point or
‘topic sentence’ in initial position, especially in
paragraphs (see for example Eden & Mitchell 1986). Martin
(1992:437) calls such topic sentences
‘macro-themes’. Similarly every sentence will have
a Theme , and the sequencing of Themes, or ‘Method of
Development’ in a text is one of the means by which a
coherent text may be built up. Whether themes are unmarked (i.e.
realise grammatical Subject and thus are cast as responsible for the
argument of the clause), and what actual semantic relationship obtains
between these tokens in the texts, highlights an important facet of the
interrelationship between the interpersonal, the textual and the
experiential metafunctions in the topics which are taken up by other
interactants (see also Thompson 1998, and Hunston 1989). For example,
Themes are generally presented as given knowledge, and will have some
relevance to the construction of tenor, or relationship of writer to
audience by its assumptions as to what needs explaining or what can be
taken as read.
In terms of the experiential metafunction for example, the nominal
group, as outlined in Mod 1[note 3], carries a lot of semantic weight as well
as responsibility for instantiating frames of coherence via identity
chains as cohesive devices (see e.g. Martin 1992: 417ff, Cloran 1999,
Hasan 1999). In the next section, therefore, a short discussion of the
role of nominal groups is presented.
Nominal groups typically function as Participants in the clause, as
well as acting to make claims or assumptions via their use in
qualifiers and circumstances. In very dense texts, nominal groups may
realise grammatical metaphors by which means verbal processes become
unavailable for argument. Such nominal groups instantiate the Thing as
Head noun, and thus may realise the Agent or Goal of a Material Process
clause, but the same nominal group may also include textual information
in the cohesive links they realise via both reference and deixis.
Interpersonal information is also carried by the nominal group when
naming and referring to other interactants by role ascription, for
example, or via evaluative lexis in either classifiers or epithets.
While some of these issues were raised in Mod 1, the potential
strategic use of nominalisations needs to be addressed again briefly
here, since it is implicated in the rhetorical development of arguments
in context, and especially in the reading of invoked attitudes
regarding ambiguous or underspecified targets of attitude, something
which I rely on as a given in Chapters 4 and 5 below. The following
extracted sentence will serve to illustrate some of these points:
In my short time on the Net, almost a year now, the most common impulse
for members of a group is[ to curb, to reign in, to conform the behavior of
the newb to an habituated standard]. [wvn10.2/sally19] In this clause complex, the topical theme the most common impulse for
members of a group is signalled by 'presuming reference' (c.f. Martin
1992: 100ff), through the use of the definite article, and impulse is
given the status of Head. The members of a group, occupying a less
prominent position in the clause as qualifier of the Head noun, are not
argued as experiencing such 'impulses'; this is presumed as given. To
deny the assertion in the sentence would entail keeping the topical
theme intact and arguing against the attributive process: it isn't to
curb… Such a topical theme would also then be available to
be referred to by means of a pronominal in the rest of the text. In
contrast, the 'presenting referent' (as distinct from 'presuming'), an
habituated standard, appears in the rheme, and therefore in the New
position. However, its location as part of an embedded clause, also
makes its actual relevance unavailable for argument. The existence of an habituated standard is presented as if its identity is retrievable
intertextually via this grammatical placement. The marked
circumstantial theme functions here to draw attention to the writer's
claim to having valid experience in the matter by foregrounding her
time of observation. Furthermore, by referring to the newb as a generic
category of listmember, and via the definite article, such participants
are ascribed a role with certain attributes which are presumed as
understood by readers. It is interesting to note here the function of
the graders most and common. If the nominal group is 'unpacked', these
epithets can be observed to expand the dialogic space, and thus
function under Engagement (Mod 2: II: 2.3) to help soften slightly the
provoked negative judgement of the presumed members of a group. The
clause complex needs to be unpacked as:
This rendering indicates that the impulses themselves are represented
as separate from those who experience them, and that there exist other
impulses - that in fact, the Actor, or rather the Carrier in the
attributive clause, is one of many other impulses. The purpose of the
statement is to classify impulses only, not the actions of members of a
group. On the other hand, if the nominal Head impulse were also to be
'unpacked' and taken back to a form in which the transitivity of
'Agents' and 'Affected' were made clear:
then the functionality of the definite article in singling out only one of many impulses is lost, and commonly will imply that this is the main activity which members of a group carry out.
My point here is that the grammar of this particular clause complex is 'motivated' (not necessarily consciously) as a strategy for expressing a negative judgement of generic participants as target - including in this case implied audience members - in a less than direct form. Having the judgement rendered in this form can function at particular junctures of an argument to present unargued, or presumed propositions, on which the writer may base further arguments.
In the texts examined in this study, strategies involving invoked (or
implied) judgements of ambiguous targets most often operate on both
interactive and autonomous planes at once, and most often occur as
summaries of sections of text for strategic reasons such as leaving
interpretation more 'open'. Invoked attitudes have therefore been taken
to signal boundary conditions of text segments as argued in Mod 2: II:
1.4. Two extracts below serve
to illustrate this point. Both examples represent the final sentence(s)
of a post, and in both the invoked attitude depends not only on
retrospective (intra-textual) reference, but inter- and extra-textual
knowledge as well. At the same time, as is common in these texts, the
final or pre-closing move incorporates a change in orientation to the
future, and in Ex 1.2 (and for this writer in general ), outside the discourse itself to the writer's material situational
setting.
What is good for the goose ought
to be good for the gander, and thus postmodernism is as critically a constituted
fiction of social structure as the loading
dock ethos to
which I must report in a few minutes.[tvs47-/simon14] Is there really such a need to determine "who's better?" Some of my best
friends are ... oh, never mind. Shannah, are you drawing these lines because I've _agreed_ with you a couple times lately? [sft22.8/stan3] Such types of closing move could be seen as involving what Sinclair
(1987, cited in Hunston 1989:99) calls a change in posture which is
signalled by changes in attribution and tense. These examples certainly
involve the appearance of the writer and (relative to registerial
context) a change to present or future time. In the case of Ex 1.3, the
final interrogative sentence could be classed as encapsulating as well
as prospecting - something that Sinclair (1993:12) claims is not
generally a function of questions. In the examples above the appearance
of the writer as Addresser, and the indicators of orientation to future
time are underlined. Ex 1.2 also provides another example of presuming
reference realised by a complex nominal group (highlighted in green)
qualified by an embedded clause. At this point, what these examples
serve to illustrate is the interrelationship of lexicogrammatical
resources being deployed at certain points in the text where rhetorical
"posture" (what Goffman opcit refers to as
‘footing’) shifts. In turn, such changes serve as
one important indication of the articulation of text organisation
identified in this study.
The problem of incorporating a dynamic analytic perspective to texts of
this type, i.e. what are essentially monologic in creation, and yet
allow for dialogic-type responses within the same context of
interaction, can be illustrated by discussions and analyses which
attempt to capture and account for the patterns of unfolding discourse.
Cloran (1999, following Hasan 1985) for example, argues that a
'rhetorical unit' (RU) can be classified according to its location on a
continuum involving mode. The continuum has, at one extreme, language
as ancillary to the social activity taking place, and as constitutive
of the activity at the other extreme (c.f. above 1.2.1). It has already
been pointed out (Mod 2: I: section 1, and above 1.2.2) that the use of
language in email list interaction is entirely constitutive of its
context of situation. Yet, to ignore its material situational setting
(MSS: Hasan 1996: 39), i.e. its technological mediation, would render
any explanation of its context of situation completely inadequate. This
relates to the research problems encountered in this study, in which
the tension that obtains between the two perspectives of text analysis,
synoptic versus dynamic, is largely tied to a notion that the texts
produced in any social activity can be classed on a continuum between
written or spoken mediums. The analytic problems that this raises can
be illustrated by reference to Cloran's (op cit: 199) discussion of
RU's which can be classed by reference to their location at the
ancillary end of the mode continuum. She observes that the 'rhetorical
configuration' of RU's produced at this end of the continuum may be
recognised by:
[the fact that](a) the central entities are the interactants
themselves, and (b) the events referred to are occurring concurrently
with the moment of speaking or will occur immediately as a consequence
of the message.
Comparing what Cloran says above with the common features of texts
produced in email interaction - a written, reflective mode using
language as constitutive of the social process - central entities are
similarly found to refer to interactants themselves, usually as I, you,
we, or in actual forms of direct address (c.f. above Ex 1.3).
Additionally, the events referred to in electronically-mediated texts
can be viewed as "occurring immediately as a consequence of the
message" if the "moment of speaking" is instead viewed as the
utterance, i.e. the moment of reading, rather than the moment of
writing, since response is considered to take place when the text is
read by a conscious human participant . Furthermore, the use of
deictics such as here, usually taken to indicate a concrete situational
context involving close material - at least visual or temporal
-proximity, is also quite high in these texts. In Table 1.1 below for
example, the frequency of lexical items you, I, we and here in the
thread "JVS" (n = 16,969), and "ALL" texts (n = 53,377) is compared
with those in the cobuild corpus, and standardised to 1,000 words.
Except for the British spoken corpus (brspok), you has a higher
frequency in the representative text sample JVS than for any of the
other corpora, and this is repeated for the items I and here. These
features tend to demonstrate that, as might be expected, the
lexico-grammar of the textual metafunction alone cannot be relied upon
exclusively in this context of situation to match up with previous
correlations linking it to mode, and for me, points to a context that
entails a higher degree of relative interactivity (c.f. Mod 2: I )
[[Table 1.1: Comparison of some deictic markers in selected corpora
3 highest frequencies in compared corpora for each lexical item: 1; 2; 3]]
Hasan (1996: 46) notes that the differentiation between monologic and
dialogic modes will be the most salient feature determining whether the
negotiation of context - its norms or conventions appropriate for
social activity in any community - is possible or likely, via what she
terms ‘process-sharing’ (see discussion Mod 2: I).
Her use of the term ‘dialogic’ in this sense refers
to interactants having immediate feedback possible in their context of
situation. As discussed in Mod 2: I, the context of situation
engendered by an email list, as an asynchronic mode of interaction,
does not make such immediate feedback possible, but it does allow
delayed feedback and it is this type of interactivity which was
concurrently investigated in this study. A continuum extending between
most institutionalised and most individuated extends the
subcategorization of context in Hasan's (op cit) model, and accounts
for the ways in which convergent and redundant coding, especially in
material situational settings (MSS), acts to reduce ambiguity, or as
Hasan notes, reduces the probability of individual negotiation over
what is 'norm-al'. By ‘convergent coding’ Hasan
refers to the means by which interactants use the available resources
both linguistic and material, to increase the likelihood that their
meanings will be understood and acted upon. When interaction is
face-to-face, the possibilities for redundant or convergent coding are
increased, with the consequence that the oft-remarked-upon lack of
visual and auditory cues in email interaction tends to result in a
context which is ripe for misunderstandings. In order to reduce this
possibility, interactants may resort to several mode-related avenues
for dis-ambiguating their verbal behaviour, as discussed in detail in
Mod 2: I. The nature of the institutionalisation of such methods of
rendering the relevance of the verbal behaviour more (or less)
transparent is related to the 'layers' (tracks, or channels) of
meaning-making introduced here, which interactants employ dependent on
what is available in specific MSSs. It is these devices - of
formatting, cohesive harmony, and evaluation - that comprise what I
refer to as frames of coherence, similar to what Hasan (op cit: 46)
refers to as frames of relevance, and which act to co-articulate the
boundary conditions of text-events, and thus realise, or instantiate at
least, what I am calling the rhetorical organisation potential (ROP) of
these texts. Because I see such ROP as negotiated and developed over
time, as open to change, and as a type of generalised abstraction taken
from a sample of representative texts in this study, I prefer to see
the actual posts as ‘instantiations’ of the ROP
rather than ‘realisations’ of something which is
already ‘there’ in the email list to begin with.
The following section presents a summary of what constitutes the layers
of analysis I introduced above, together with a description of certain
textual features which are salient at each layer, after which, in
section 1.6, this approach is illustrated by reference to a text (post)
which the resulting model is able to accommodate despite the post's
unconventional framing - or what is better termed its atypicality in
terms of its appearance in this context of interaction.
In effect then, layer describes the levels of analysis that are
attended to in order to characterise the organisation of text-events in
this activity mode. It is these recurrent text-events that I have
generalised as a model I call 'rhetorical organisation potential'. In
chapter 2, I go on to describe the nature of the rhetorical
organisation potential of these texts in detail, together with a
description of the method used to derive the model from an analysis of
representative texts.
At the first, or 'outer' layer of analysis, relevant features include the obvious interface and writer-determined formatting features outlined in Mod 2: I: section 3.3. The labels for the four fundamental styles or text-types prevalent in most lists are reproduced above (1.4.2) , and these make reference to the gross means by which posters construct their whole texts as contributions to an ongoing conversation via the use of graphic means such as signalling the quotation of parts of previous posts. In this sense, Layer 1 analysis treats whole posts as complete objects after the fact. This level of analysis is synoptic to the degree that it attends almost completely to the expression plane, and how the disposition of quoting, spacing, and other punctuation within the body of the post - as well as the technological interface itself - helps to frame the actual content of the text in order to cue meanings. It is concerned to discover how the writer has set apart his or her turn(s) within the post itself.
At this level also, the subject line, and the contents of the header in
general, form one of the first framing devices that the reader has
available. The subject line in particular frames the content of the
post as part of an ongoing series of contributions - in the first
instance by the technological insertion of "Re:" when the post is made
in response to another on the list, as well as when the writer chooses
particular words deemed relevant to the content. Other information in
the header can also be considered to frame the content in this sense,
since it tells the recipient who the writer was and at what time (and
place) the message was sent . However, in determining whether a post is
a Response of whatever type, Layer 2 features are more salient,
although formatting signals their presence. In this sense, the subject
line at Layer 1 is an empty value: it is part of the outer layer and
remarkable if absent, but the actual content only becomes relevant when
other features at Layer 2 are not present (c.f. next section). Features
relevant for analysis at this layer are obviously related to many of
those outlined under Dimension II of the cline of relative
interactivity introduced in Mod 2: I: 3.6.
[Eden & Mitchell paragraphing for the reader?]
This layer is concerned with the relationship of the post to previous
posts (or parts thereof), and with the signals used to indicate an
orientation to the context of the ongoing interaction. This level of
analysis has a dynamic orientation to the degree that it is concerned
to describe the means by which writers indicate and readers might
perceive how the post can be re-contextualised. Rather than treating
the whole post as a bounded object, this level of analysis looks at
intertextuality in both the content and the expression planes. As
mentioned earlier (1.1.2 and 1.3.3), the use of quoted material from
previous posts can be viewed at one level as part of the manifest
intertextuality of the text. The means by which this material is set
apart from, or integrated with the turns constructed by the writer can
be indicated by both formatting and by discursive features within the
text, as will be illustrated below and in more detail in Chapters 2 and
5. Writers may make reference to previous contributions in a variety of
ways to indicate the relevance of their own contribution, but beyond
this, they may also indicate their stance in relation to the proposals
and propositions made in previous contributions as well. This issue was
also introduced and discussed briefly in Mod 2: II: 5.
Briefly, from a dialogic perspective and the writer/Addresser's point
of view, all their contributions are responses to some previous text,
and in the context of an email list, my observation is that most
contributions indicate some relevance, and hence their responsiveness,
to some topic addressed previously on the list. Viewed from the
perspective of the reader/Addressee, all contributions engender some
response on their part , and this is the default. However, these
responses are not always made 'overtly', i.e. in writing, and so,
within the category of 'overt response', the first entry condition for
purposes of analysis is 'written & posted response'. In order
to determine whether such overt responses are directly responding to
some previous contribution (as distinct from responses to general ideas
and topics familiar to listmembers), elements appearing in the subject
line and in the body of the post are taken into account. When these
elements do not appear, then the post is classified, at the next entry
point, as an 'initiation'. At the same time, however, due to individual
posting styles, or the interface used by some listmembers, etc, these
so-called 'direct responses' may also be made with new subject lines,
and in these cases, relevance is usually indicated only within the body
of the post. For the purposes of this study, and the description of the
typical or conventional post, this method of response is classed as
'marked'.
Lexicogrammatical features which are relevant for this layer of
analysis are those of IDENTITY, such as repetition, replacement,
re-statement and reference (i.e. phoricity: cohesive collocation,
lexical co-reference, synonymy; (non-manifest) intertextuality: assumed
knowledge, etc), NEGOTIATION, and the nature of exchanges (c.f. Martin
1992: Chs 2 & 3), and TAXIS, i.e. expansion (elaborate, extend,
enhance: c.f. Halliday 1987; Martin 1995). In terms of NEGOTIATION,
what is attended to at this level are features which relate to the
argument of the clause, and whether the negative or positive evaluation
of previous statements is taken up in the response. In other words,
whether the overt responses made in posts are congruent in terms of
MOOD . The default position in this case is not 'support' (negative or
positive) but 'non-support', which includes the response
"ignore/silence" as mentioned above (where 'ignore' relates to direct
elicitations, and 'silence' to statements of fact of opinion. See Ch
2:#). In order for a response to be classed as a reply in the
Goffmanian sense outlined in Module 2: II: section ## the text needs to
represent some from of reciprocal and congruent response. Between these
two classes of Response are a variety of responses which address the
propositional content of the previous quoted contribution, but provide
a challenge by supplying meta-commentary, by dismissing the value,
status, or relevance (Hunston 1989) of the content, or diverting the
topic in other ways. As an example, in the excerpt below, the response
takes up (via [support: confront]) the position by arguing with it, and
thereby makes a reply despite the disagreement:
Date: Sun, 16 Nov 1997 19:49:06 -0800
From: "T- M. S-" <email>
Subject: Re: Wide-talkers v. narrow-talkers
At 5:29 AM -0800 11/15/97, RW- wrote:
>Yes. If we have to get serious, although you don't have to be
snotty about it...
bullshit. he does have to be snotty about it. obviously.
as you clearly noticed and signaled to us by denying it.[wvn46.15/ter] In contrast, in the following excerpt, the turn which follows the
quoted material does not address its positioning, and would therefore
be classed as a response but not a reply:
Date: Mon, 4 Feb 2002 12:56:00 -0800
From: "T-- M. S--" <email >
Subject: Re: Excuse me but I couldn't resist!
At 9:57 AM -0500 2/3/02, D-- M. H-- wrote:
>I must admit I am angry about the actions of al Qeeda.
One interpretation (pattern of understanding) of any "anger" felt
"about" some event that does not occur in one's immediate physical
presence is that it is (made possible by) previously felt but not
resolved rage, usually left over from infancy. [gen02.18/ter] In the response given in the example above, the arguability of the main
clause has been changed, with the Subject of this second clause being
nominalised as 'anger', while the Subject of the first framing clause
quoted has become an (unstated source's) interpretation of 'cause of
anger' in general. This means that the writer of the response has
adopted a "Speaker" role (as distinct from an Addresser who is
responding to the Addressee) in which he is the 'animator' but does not
acknowledge his status as 'principal' in Goffman's terms, instead
extravocalising, or sourcing the attribution to another authority: one
interpretation (c.f. Mod 2: I, section 3.6.1). The writer here gives
information about 'anger' as a disembodied emotion rather than as a
process, a process felt by a conscious 'emoter'. He does not respond
'in kind'. To put it in terms of Berry's (1981) framework, an offer of
information in a K1 position (an "A event"), has been responded to by
yet another K1 move: instead of accepting the information given (and
extending, elaborating or enhancing on the propositional content), it
has been treated as a request for information - as a K2 move, or "B
event". A more congruent realisation of the exchange would allow a
change of Subject, but with the process maintained or negated via
polarity. For example: "Yes, many people are angry about…"
or "There's no need [for you] to be angry about…", or even
"I understand your anger, but…". These would at least
maintain a number of co-referents in an identity chain apart from the
semantic domain which includes angry and anger. From the perspective of
the transitivity, [to be] angry is in the first instance the Attribute
of a conscious Carrier, whereas in the response, anger has taken the
Carrier role and been given attributes of its own.
The system of Response types is outlined in detail/represented visually
in Chapter 6
The content of the subject line in the Header is again implicated in
analysis at this layer and acts as a signal of relevance, indicating
whether the contribution will address the content of another
contribution, or initiate a new topic.
The contribution of "non-manifest" intertextuality - of references to
assumed knowledge, metaphor, shared allusions and so on - is also
relevant to this layer of analysis. Expectations that participants will
be able to retrieve such references is dependent on what Halliday and
Matthiessen (1999) amongst others have termed phylogenesis: the
development of a semiotic system(s) within a community. Because the
list conversation, if regarded as ‘whole text’ has
involved large changes in speaking subject (Bakhtin op cit), and even
changes over time in the mix of interactants, the term logogenesis
cited earlier would not be applicable here. The relevance of the
development of conventions for positioning others through evaluation
especially was introduced briefly in section 1.3.1 above, as it is
particularly salient for the discourse organisation of the group in the
study. This is revealed by the extensive use participants make of a
wide spectrum of strategies for invoking attitude (in contrast to, but
not to the exclusion of inscribing attitude: see Mod 2: II. #), and not
only limited to invocations which rely on intertextual allusions.
Because such positioning strategies can help account for the
construction of poster identity within the group, this aspect of the
system of Appraisal will be discussed in more detail in Chapter 4
below, with reference to the analysis of attitude in the texts. The
analysis of attitude and the disposition of attitudinal prosodies is,
at the same time, used as the fundamental signal of rhetorical text
organisation in this thesis, and is a focus of Layer 3 description
outlined in the next section. Layer 2 analysis is discussed again in
the latter part of Chapter 5 where the possibility that poster-gender
influences response patterns is presented.
This layer again treats the post as finished object, but is concerned to also identify intra-text signals of coherence and the development of an argument via logogenesis (c.f. Halliday and Matthiessen 1999, Martin and Rose 2003). In this sense, the features which signal the framing at this level are treated both synoptically and 'dynamically' to the extent that it views rhetorical positioning as unfolding with the co-text. From another perspective, this means that analysis at this level is also concerned to identify text development by paying attention to work being done on the interactive and autonomous planes of discourse, and takes into account what Sinclair (1993) glosses as deictic and logical acts. In Mod 2: II I also argued that one of the main ways that the development of the argument may be identified, is through an appraisal analysis which takes into account not only the dispersion of attitudes themselves, but also the sources and targets of evaluation as they appear in sequence throughout a text. These, in combination with strategies which appraisal classes under engagement (see Mod 2: II: 2.3), trace the evaluative positioning strategies of the writer, and the writer's construction of the ideal reader whether a named Addressee or the audience of 'onlookers'. As outlined in Mod 2: II, section 2.3.2, anyone, from named Addressee through to unknown 'eavesdropper' may be the real readers who may take up or resist the positioning in overt responses, and future research on the actual relationship between positioning and nature of responses is planned using the framework introduced in this thesis, as proposed in Module 2: II: section 6. In the context of this thesis, however, it is the means by which positioning strategies articulate the relationship between writer and projected audience members (stance of the writer), and the negotiation of 'identity' (deemed the social purpose of the texts in this study) - which in turn depend on the generic conventions available in the local ROP - which is the focus of this layer of framing.
My argument, similar to those proposed by Hunston 1989 and Hood 2004,
is that evaluative peaks, or prosodies, are one of the main means by
which an argument is developed and by which it can be analysed.
Evaluative prosodies are obviously related to the positioning evident
in any text, but my own position is that these prosodies are also
theoretically and practically inseparable from those aspects of the
text's coherence which have traditionally (under SFL approaches) been
treated as realised by textual, logical, and experiential metafunctions
in the lexicogrammar. These aspects of a text's coherence I view as
essential scaffolding for the development of the coherence in any text
and for the analysis of its generic structure potential, but there have
been few studies to date which focus on the interpersonal aspects of a
text and its tenor in order to arrive at descriptions of genre
organisation potential via rhetorical text units. Thompson (1999)
discusses in detail the implications for construal of context of
situation in SFL terms, and makes similar observations on the lack of
absolute hook-up between the lexico-grammar of the metafunctions and
the realisation of aspects of the context of situation, Field, Tenor
and Mode. He notes (opcit:106) that Halliday has "consistently stressed
that correlations between contextual parameters and metafunctions are a
matter of tendency and statistical probability, not of determination".
In addition, the focus on what I am calling "markers" in helping to
construe attitudinal positioning and argumentation parallels Thompson
and Zhou's (2000) investigation of conjuncts with attitude (p.124), and
their "emphasis on seeing texture and structure as created by
interactive negotiation between writer and reader, rather than simply
as the reflection of objective logical relations between propositions"
(p.140), and these features of the texts are attended to at Layer 3..
What is attended to at Layer 3 may consist of any discourse elements
which are viewed as useful in tracing the means by which a
text’s coherence, specifically its argumentative purpose, is
constructed. Layer 1 provides the formatted scaffolding that a writer
uses as indicators of the main sections of their post, while Layer 2
attends to indicators writers use to signal a post’s context
as part of an ongoing ‘conversation’. At Layer 3,
the organisation of the argument within the main stages of the post
were examined to discover what linguistic strategies are typically
employed to organise this argument. At the level of the paragraph,
initial or orienting clauses and clause complexes were examined to
discover what means writers use to orient readers, and similarly, what
writers choose to conclude their paragraphs. Longer stretches of prose,
termed ‘stages’ in this model, were also treated in
this way, and the model outlined in Chapter 2 describes, for example,
Opening and Closing Framers whose internal strategies realise their
organisation.
In this study, a variety of linguistic systems were noted ‘on
the fly’ using a dtd as analytic approach, as will be
described in more detail in the next Chapter. However, while such
systems as lexical cohesion (repetition and substitution), collocation,
and Thematic progression were observed in passing, the main features
attended to in this study are focussed on a very large group of
discourse markers, since they were observed to have been used by
writers in signalling the development of the argument and the
articulation of strategies which form arguments within stages. By
‘strategies’, I refer to discourse patterns
subsumed under the twin categories of matching and logical relations
such as assessment-basis, purpose-means, problem-response, and
hypothetical-real, for which markers such as conjunctions, adjuncts,
and disjuncts are essential.
It is at this Layer also that the dispersion of attitude, especially
that of invoked attitudes, was determined as significant in signalling
argument organisation. It needs to be noted that attitude was analysed
using a separate dtd, and that both analyses were compared in order to
observe interference patterns created from an overlay of both.
Specific features of the discourse which are implicated for the framing at this level can be exemplified by meta-discursive devices, such as intra- and extra-vocalisation, and other features of engagement implicated for construing provoked appraisal, as well as intertextual reference, or evoked appraisal, which was outlined in detail in Mod 2: II: section 3.3.3 (inter alia). My argument in Module 2 was that rhetorical strategies of provoked and evoked appraisal are a primary means of constructing overall arguments, particularly at certain junctures where an Addresser, mindful of his/her audience, implies (provokes or evokes) judgements regarding the values, positions or (potential) behaviour of audience members in order to maintain affiliation, or to ward off disalignment. By invoking attitude through the use of ambiguity of target, source, or evaluative lexis, the writer can make judgements without specifically putting alignment with audience members as risk, but at the same time, these instances of ambiguity appear to be located at regular intervals in the texts, and serve to both articulate the progress of the argument and highlight the writer's assumptions regarding audience and their possible ideological/axiological differences at the same time. Specifically, the proposal set out in Mod 2: II, was that instances of provoked and evoked appraisal in these texts provide one of the most useful ways of approaching the identification of textual boundaries or phase shifts. It was contended there that such instances of invoked attitude tended to be employed at the close of phases of argument, especially in the pre-closing sections of posts themselves. These findings will be demonstrated in more detail in Chapters 2 & 3, 4.
The idea that writers use both opening and closing sections of texts
for highlighting significant meanings in their texts, and that readers
expect this disposition of meanings is not new. Eden & Mitchell
(1986: 418) for example note that: “Readers expect to find at
each paragraphs’ peripheral points something which merits
special attention”. One of the main approaches in this thesis
is to make observations of features at Layer 3, overlaid with those at
Layer 1, and thus describe the common or conventional ways that signals
of framing are co-articulated: to discover what typically constitutes
the peripheral points of texts and parts thereof.
Figure 1.2 below sets out the main types of invoked attitudes
identified at the time of writing, including examples of how these
types of provoked and evoked attitude may be typically realised. In
Martin and White's then classificatory system (c.f. below),
subcategories of invoked attitude are all subsumed under evocation, but
my own analysis suggests that strategies covered by 'flagged' and
'provoked' attitude are better seen as having a closer relationship to
each other. This is because, for example, grammatical metaphor and
instances of nominalisation as discussed above (1.4.2.1) are certainly
deployed as resources in the evocation of attitude, but are located in
Martin & White’s schema, on the furthest end of the
cline of more experiential, less 'saturated' evaluative meanings which
are highly dependent on other co-textual signals. In contrast, my
approach is to view the provocation of socially shared attitude as
encoded in highly evaluative lexical metaphors located at the other end
of such a cline. In the middle are strategies which involve such
"markers" or flags of attitude as have been discussed variously under
headings such as engagement, conjuncts with attitude, and matching
relations.Figure 1.2: Typology of invoked attitudeThe typology
represented above shows for example, that flagged attitudes refer to
such resources of co-textual signalling as those subsumed under
Engagement. In Chapter 3, one of the means for signalling presumed
dis/alignment with interlocutors is reported in more detail: the use of
'not', 'no' and other signals of negation which not only organise the
argument on the autonomous plane, but also act to position readers in a
variety of ways which set up or provoke the reading of evaluative
attitude in texts. This type of 'flagging' of attitudinal positioning
is also addressed in Chapter 3.
In terms of the category [evoke: afford] 'identity chaining' (see for
example Cloran 1999, Hasan 1999, Martin 1992) provides a further means
for tracing evaluation together with its targets and sources, and the
coherence of the texts, but identity chaining is in any case,
concurrently investigated in any detailed appraisal analysis. As
investigated in Module 1, the analysis of texts concentrating on the
nominal group, transitivity, and the realisation of social actors can
also contribute to the tracing of such 'afforded' attitude, especially
when their reading is dependent on signals that are outside the scope
of the primary text.
This view of staging or organisation can be likened to the move
structure
framework used by Dudley-Evans, Swales, etc, for the analysis of
academic discourse. Lewin et
al (2001) refer to these types of moves as 'functions' (op cit: 17) and
they cite a study by Dubois (1997)
in which moves are classified as 'rhetorical functions'.
The following section provides an example of one of the more
un-conventional texts as a means of demonstrating what the model
attends to when identifying boundary signals which are then described
using a dtd as template. Chapter Two goes on to describe the
conventional staging of posts in terms of this dtd and closes with a
number of example posts which are typical of the macro-genre and the 4
fundamental text-types developed within this speech community (see
above 1.3). Although the data on which I base my findings consist of
files of almost every post contributed to the discussion list dating
from October 1995 up until the present day, texts which were used for
the close analysis which informs this study and the resultant model
were chosen as representative of the interaction and as prototypical
texts. My model is therefore necessarily based on a selection of the
products of the list's activities, and it attempts to account for the
dynamic creation of meaning in the context of unfolding of the
discourse and within the text-units identified. The selection of what
counts as 'representative', and the accounting for aspects of the
meaning-making in text events is also based on my own active
participation in the group. The significance of active participation in
this type of case study was argued in Mod 2: I: section 4.1, where my
own assumptions on the nature of group membership and culture were
acknowledged as motivating the methodology adopted. This methodology
will be outlined in detail in chapter 2, in which the working
assumptions, the text selection, and the framework itself will be
presented and illustrated.
What follows is one text taken from the concluding stages of the thread
JVS ('jerry versus steve': Chapter 2 below outlines a description of
text tagging systems). The post appears to be organised somewhat
unconventionally from the point of view of the macro-generic stages
typical of the list interaction, in that it is almost completely
comprised of a culturally recognisable core-genre [limerick]. However,
in the context of the ongoing interaction, and despite the lack of
obvious co-textual framing at what I am calling Layer 1, the
positioning of both Addresser and especially the evaluative target
Addressee, would be immediately understandable for any participant in
that conversation, the putative Addressees. In this sense, its status
as a coherent contribution to that thread can be accounted for using
the resources of reference and evaluative positioning at Layers 2 and
3, and the most obvious of the resources used are outlined below. In
addition, some of the resources by which identity construction is
effected via such intertextual reference and evaluative positioning are
again discussed with reference to this text in Chapter 4. As an example
of the 'prototypical' text, [jvs234.56/steve33] reproduced here as Ex
1.6, needs to be classed as an extreme example of the macro-genre,
while at the same time, it constitutes a relatively simple example due
to its conventional core-generic staging. It therefore provides a good
introduction and test case for the presentation of the model in detail
in Chapter 2. Meanwhile, the sections which follow demonstrate how the
layers of framing act to provide the levels of abstraction on which the
model is based.
Date: Mon, 7 Jun 1999 02:00:13 -0800
From: spr@email
Subject: There goes rhymin Simon...
1.There once was a list, analytic
With Simon, Kaylene, and a CritiC
A couple o' bards
A trickster (not cards)
And Ray in his Caddie.. or Buick?
2.To spice up this bozo-filled mix
Add 12-steppers, pomos, and cliques
MBTIs
Gals versus guys
Aussies and bikers and pricks
3.Small wonder that tempers start flaring
When feelings find overdue airing
Content alone
Is dry as a bone
But affect's a burden for... sharing
4.Emotion, a curious thing
To our own we invariably cling
When instead it's not ours
It must come from Mars
Flung by a shit-stirrer king
5.Inflation, projection, denial
Can all turn discussion to trial
It's hard to be sanguine
When yer 'squirrels' they are hangin
And your humor is soaking in bile
6.The couching of feelings in theory
Makes some of us itchy and leery
Straightforward gripe
Trumps prettified snipe
And leads to clear vision, not bleary
7.My message I'll sum up discreetly
In verses so softly and sweetly:
Hiding one's rage
On CRT page
Says the very same thing, but effetely.
***
8.Biker T-shirt: "I AM the man from Nantucket."
Stan
------- Ex 1.6 above can easily be seen as divided into 3 main stages as
indicated by its overt formatting, which in turn is partly a feature of
the interface, partly a feature of the signalling of a
culturally-recognised core genre, limerick, in which attitude can be
'provoked' rather than inscribed, via expectation of amusing rhyming
'twist' in the final lines of each stanza.
At its outermost 'layer' an email post can be divided into two main
stages as provided by the interface, or technological mode: the header
and the body, as already discussed in section 1.5.2 above. This may be
easily observed in the example text above.
The body of this text is formatted as 2 stages (more specifically 3
stages if the typical closing framer stage is taken into account - in
this case consisting solely of the handle); the first or main turn, and
a preclosing stage or short turn which is marked as separate from the
rest of the text by means of white space ('carriage return') as well as
a short line of asterisks. The main turn of the body is divided into
stanzas conventional for this core-genre (or ‘activity
sequence’ Martin 1992: #; Lemke 1995a: 86), and signalled
explicitly by the separation of each stanza (termed part for the
generalised labels used in the framework outlined in more detail
Chapter 2 below) by a line of white space. Finally, the end of the post
is comprised of a closing framer, again signalled by separation via the
use of white space, and in this case comprising the first name (or
handle ) of the writer. At this Layer, the post is marked in terms of
the local list-current conventions mainly through its lack of an
opening framer. At the same time, this style of text can be classed as
an example of the "non-indicated" style (c.f. 1.4.2 above) rather than
an "initiation", because of its obvious relationship to the ongoing
context of discussion, as evidenced in other Layers.
There are several indicators that this text can be re-contextualised as part
of an interactive conversation - mainly reference to list-current identities,
together with actual naming of other posters/contributors, as well as the evaluation
of these named and referred-to contributors. The subject line used, There goes
rhyming Simon . . ., indicates that the post has not been made in direct response
to any other, due to its lack of the marker ‘Re:’ and this would
usually indicate it has been made as an initiation. However, as noted above,
reference to rhyming Simon in the subject line would alert participants that
it was made in response to an earlier contribution by the poster identity Simon,
who similarly posted a limerick previously ([tvs188.50/simon19b]). A typical
Reply makes its initial evaluative positioning clear by maintaining the topic
in some way, and this is not obviously the case with this post (c.f. Chapters
2 and 5 below, and Mod 2: II. section 6 for discussion of typical Response and
Reply openings). At the same time, the text's main evaluative target, although
not referred to by name, is easily retrieved by list participants due to the
convergence of several thematic strands (Lemke 1995a: see discussion Chapter
4 below) and therefore this post can be classed as part of the ongoing thread,
and as a Response to a specific previous contribution, rather than a Reply.
The presence of so-called thematic strands, related to topic maintenance, is
therefore definitive for distinguishing between a Response and an Initiation.
In determining this post’s status as part of an ongoing thread (despite
its lack of maintained subject line in the Header), the first two stanzas -
by means of intertextual reference - both orient the readers and claim their
affiliation with the use of referents such as Simon, Kaylene, CritiC, a trickster,
Ray, bozo-filled mix, 12-steppers, MBTIs, Aussies, and bikers. The actual topic
of the post begins at stanza three with the presumed reference to tempers start
flaring. The topic of ‘hidden anger’ on the part of one of the participants
has been maintained throughout the thread. Chapter 4 discusses the implications
of this for (invoked) evaluative positioning, but the point here is to note
that at Layer 2 such referents in the text indicate its membership in a thread:
a chain of responses to a similar topic.
Classification of a post as Response or Reply also depends on the
evaluative stance taken by the respondent to the positioning in the
responded-to post(s). As stated earlier, this is usually dependent on
whether the Response realizes congruent mood to take up or argue with
the earlier positioning, and/or provides evidence of topic expansion
(enhancement, elaboration, extension). If it is agreed that the post
being discussed here is a response to a limerick contributed earlier by
the poster identity Simon, then it should also show specific reference
to that post. In this case, there is no room for congruent argument
within a typical exchange, and so because the genre [limerick] does not
use the resources of NEGOTIATION, the post has been classed as a
Response rather than a Reply. On the other hand, it does take up the
positioning to some degree, as evidenced in the following excerpt from
the earlier contribution:
There once was a psych, analytic,
A Freudian internet critic,
His cold common sense,
And a sly arrogance,
For some was far too acidic.
Our Stan who likes object relations,
And long Harley biker vacations,
Says to us, Netdynam,
"Yo group, here I am,
But I'm not here to fill expectations." The acceptance of the positioning seems to be most evident in the writer ("Our
Stan") taking up the positions made for him in the above excerpt: he responds
by making acidic comments on the members of the group, Netdynam, expands on
his [status: authority] as psych, analytic by using the terms inflation, projection
and denial, and closes by reference to his identity as a biker.
At this Layer, the "content" of the main stages of the text are
observed, in order to derive stage intra-organisation labels in terms
of their "expression". In the case of Ex 6.1 the main stage consists
entirely of a limerick, and as such, the
‘paragraphing’ is constrained by the form. As well,
the evaluative prosodies are somewhat constrained by this form also, as
discussed further below. The ‘pre-closing’ stage
employs intertextual reference to the core genre and relies for effect
on the audience’s familiarity with the content and form of
other limericks, as well as the situational context – the
thread - in which this is a coherent contribution, and in which he is
claiming ‘victory’ though his prowess at genre
manipulation .
The use of a core-genre, limerick, as the entire content of the main
stage of the post constrains the choices for argument organisation
employed, since the constraints of stanzas and rhyming override other
needs for markers and common prose signals. At the same time, it allows
the writer to signal that the content must be read against the
culturally assumed background of ‘play’, and the
expectation that the last line of each stanza must provide some
evaluative or semi-surprising quip. Within this main stage of limerick,
several sub-stages or stanzas may be observed. Stanzas 1 and 2 provide
the orientation, by outlining the ‘setting’: a
description of the email group discussion. The
‘marker’ in this case could be said to rely on an
intertextual signal of the genre ‘fairy tale’:
There once was…
The actual topic of the post is introduced in stanza 3, as noted
earlier, and this is where the report of Affect is first introduced:
tempers start flaring; as an action with no specific Emoter (in
Appraisal terms). The possible negative value attached to this Affect
is ensured with the Circumstance which follows: feelings find overdue
airing. As noted previously, the target of this negative evaluation of
a situation is not stated, but readers familiar with the thread will
have no trouble retrieving the intended target in the context of the
writer’s previous claims regarding his behaviour. Stanzas 3
and 4 fall naturally together, while stanzas 5 and 6 extend the
negative evaluation of the target. The final stanza (7) is a claim,
framed as a pronouncement, which 'sums up' the negative evaluative
stance of the writer towards the target. Each stanza, as expected,
closes with a negative appraisal of the target or a comparative
positive appraisal of actions which the unnamed target does not
display.
In summary, the model divides the body of this post into two stages:
turn and closing framer, the main stage being further sub-divided into
2 sub-stages: turn-parts and a pre-closing reframer. The turn-parts is
comprised of an opening and a continuing, and the continuing section is
also divided into parts - in this case, realised as stanzas.
Such stages are usually derived in the model by looking at a combination of
signals of boundary conditions. The ones I have taken into account in this study
are changes in orientation such as tense, attribution, addressivity (this chapter:
contact eldon), change in topic, or changes in evaluative positioning. A loose
group of textual and interpersonal markers were also tracked and treated as
'framing cues' of this type.
This section focuses on an example of evaluative prosody, what could
also be referred to as 'changes in evaluative positioning' over longer
stretches of text. It is a term referring to accumulated values in a
text, and so relies on signals of cohesion and reference to both mark
the boundaries of such phases, and to signal that series of evaluative
positions are linked within them. Many such linked positionings can be
seen to provoke an overall attitude (c.f. Fig 1.2 above).
As observed in the previous section, Stanzas 3 and 4 are both directed
towards a similar topic – that of feelings and emotions in
general. The stanzas which follow are then linked to it by the
rhetorical strategy of expansion . Stanza 4 begins with the Theme and
Subject emotion, which is then appraised as curious. The following 2
stanzas, 5 and 6, are grouped as a subsection, since they expand on the
topic of emotion in more specific terms yet also refer to general
evaluations of ways in which ‘emotion’ may be
approached. Specifically, in stanza 5, types of emotional reaction are
evaluated by ascribing to them the means to turn discussion to trial,
with the example of an unnamed possessor of humour which is negatively
judged as soaking in bile. Stanza 6 changes theme while retaining
semantic domain with the couching of feelings in theory. This
nominalised activity has the power to cause negative [appreciation:
reaction]: itchy, and negative [affect: dissatisfaction]: leery; in
some of us. This in turn represents a strategy which claims affiliation
with readers, and at the same time wards off the resistance of those
who may enjoy ‘couching feelings in theory’ by
means of dialogic expansion. Stanza 6 also contains the highest
concentration of negative Appraisal of the unnamed target, via both
positive and negative Appreciation functioning as tokens of [judgement:
capacity: negative]. This stanza is the last but one - and in these
texts, one of the common patterns observed is that the last but one
unit is often the bearer of the evaluative nexus or peak. The
pre-closing or final unit in this pattern is then given the function of
'coda', realized in general as an orientation to some future, usually
by any or all of textual, ideational or interpersonal prospection
(explain Ch 2?),
In this text, this function is not borne entirely by the final stanza
of the limerick which begins with my message as theme, and claims to
sum up what has gone before: the final stanza here has the function of
"sum-valuating" the larger unit, and this is then followed by the
pre-closing unit, which characteristically changes orientation, here
represented by
8. Biker T-shirt: "I AM the man from Nantucket." The final attitudinal term in the final Stanza 7, effetely, brings
together some of the other themes of jokey masculinity which have been
evident in the thread up until this point, and once more evaluates the
target with negative [judgement: capacity].
[Thus the post can be seen as having an organisation at Layer 1
represented by a limerick and as comprised of 3 main stages. The stage
encompassed by the Turn may be further described as
‘organised’ via its semantic prosodies which refer
to intertextually-shared identities as the targets of evaluation. These
evaluative phases in the body of the text are signalled by both
inscribed and invoked attitudes.]
This chapter has argued for the approach used in this thesis for
analysing a set of representative texts in order to propose a typical
or conventional macro-genre for the posts produced by members of a
specific electronic mailing list. The approach was designed to look at
a variety of discourse signals writers use in organising their posts
and the arguments they contain. The approach was designed to
investigate the organisation of posts by viewing each post as a text
employing these signals, or framing devices, at 3 distinct levels of
analysis. The levels of analysis were termed Layers, and these were
described as being integrated within each post. Layer 1 attended to the
‘gross’ formatting features of the texts, such as
the use of paragraphing and other formatting features available in the
medium. Given the use of paragraphs and other formatted signals of
stages within the posts, the approach is described as setting out to
examine how arguments are organised within these overtly signalled
stages. Specifically, the approach uses a methodology outlined in more
detail in the next Chapter, in which orienting and concluding sections
of stages are observed, and the strategies for organisation of argument
using contextual markers is described. Layer 3 was introduced as
attending to these discourse functions, and in addition, as
encompassing the insights provided by attitude analysis which was
presented in Module 2: II. The overall purpose of this study was to
propose a model which I have termed the rhetorical organisation
potential (ROP) of these types of texts and which is set out in Chapter
2 to follow.
Chapter 2 presents and explains the dtd through which the analysis was
conducted and refined. In addition, the set of representative texts
used for the study, and the methodological approach adopted is
introduced and defended in further detail. The resultant typical
macro-generic text organisation is discussed through examples of the 4
typical formatting styles first outlined in Mod 2: I.
----------
A.Don. revised February 2006, January 2008.
[note 1] At present Module 2: Part 1 is not available online. Anyone interested in a copy in Word format, please contact me.
[note 2] Chapter 5 of the thesis dealing with textual identity is available from the thesis download page
[note 3] The paper which is here referred to as Module 1 (The Representation of Social Actors) suggests a framework for analysis based on Hasan's notion of a cline of activity and transitivity analysis teamed with van Leeuwen's categorisation of 'social actors' represented in texts. It uses the same two texts analysed under appraisal in Mod 2: II.