Content uploaded by Christoph Unger
Author content
All content in this area was uploaded by Christoph Unger on Aug 26, 2016
Content may be subject to copyright.
cog-prag explanations 03 - printout of 20.04.2002
Cognitive-pragmatic explanations of socio-pragmatic phenomena:
the case of genre
Christoph Unger, SIL International
Paper read at the EPICS I Symposium, Sevilla, Spain, April 10-12 2002
1. Introduction
For the purposes of this paper, I will start with two definitions:
Socio-pragmatics: the study of how social information enters into
communicative behaviour.
Cognitive pragmatics: the study of how the inferential processes in
comprehension work.
A comprehensive pragmatic theory will certainly have to address both these issues.
In the search for such a comprehensive view of pragmatics a question immediately
arises: how are socio-pragmatics and cognitive pragmatics related? One answer given
in the literature is that the two domains are separate and socio-pragmatics controls
cognitive pragmatics in the sense that socio-pragmatic entities are said to provide
deeper explanations for phenomena that cognitive pragmatics can't. This answer is
often given in the literature on genre (e.g. Kitis 1999; Holdcroft 1979; Green 1995;
Goatly 1994; Paltridge 1995; Bex 1992).
Another answer which is suggested in some strands of the literature is that socio-
pragmatic phenomena can indeed be accounted for within cognitive pragmatics, so
that the two domains need not be seen as separate (Jary 1998; Zegarac & Clark
1999).
In this paper I want to discuss the relation between socio-pragmatics and cognitive
pragmatics with particular reference to genre. I will argue that genre is best accounted
1
for in relevance-theoretic terms as providing access to information which helps to
cog-prag explanations 03 - printout of 20.04.2002
fine-tune the addressee's expectations of relevance. This means that the contribution
which genre makes to comprehension falls squarely within the relevance-driven
comprehension procedure and thus within the domain of cognitive pragmatics, so
that the socio-pragmatic phenomenon of genre finds a genuine explanation in cognitive
pragmatic terms. I will then point out that this relevance-theoretic account of genre
extends naturally to other related socio-pragmatic phenomena. This suggests that a
stronger claim may even be made: cognitive pragmatics is related to socio-pragmatics
in such a way that socio-pragmatic phenomena receive truly cognitive explanations.
This argument will be developed as follows: section 2 discusses the socio-pragmatic
nature of genre. In section 3 I will discuss accounts of genre which place genre
recognition outside the scope of the principles governing communication. Section 4
is devoted to my alternative account of genre in relevance theory. In section 5 I
extend this account of genre to related phenomena, i.e. register and culturally determined
differences in ways of speaking.
2. Genre as a socio-cultural phenomenon
There is a common consensus that genre is essentially a social/cultural phenomenon
(see e.g. Eggins & Martin 1997 and references therein; Kitis 1999; Swales 1990;
Paltridge 1995), or in our terms: a socio-pragmatic phenomenon. But what is a
socio-pragmatic phenomenon? An immediate answer could be that it is a set of
assumptions shared by a people group which governs the communicative behaviour
of members of this group. It could be added that a socio-pragmatic phenomenon also
relates communicative behaviour to the structure of cultural institutions. Eggins &
Martin (1997:239-241) point to the example of market transaction or service encounters,
where the structure of the social interaction is reflected in the structure of the
communications involved.
2
Genre certainly fulfils these characteristics:
cog-prag explanations 03 - printout of 20.04.2002
-genre is culture specific. Example: Epic prose such as in Homer's Odyssey
lacking in English.
-genre regulate how to express one's intentions e.g. in business letters,
recommendations etc.
-absence of genre knowledge can cause problems for comprehension. See
example (8) which will be discussed later.
Thus, genre consists of information shared in a group of people, which furthermore
relates communicative behaviour to the structure of cultural institutions. The question
arises: how do such socio-pragmatic phenomena exert this influence on communicative
behaviour? How do they influence the inferential comprehension procedure whose
explanation is the domain of cognitive pragmatics?
One answer is that the maxims or principles which govern the inferential phase of
utterance interpretation are influenced by these social entities. The cognitive
comprehension procedure cannot be completely governed by cognitive principles but
must interact with mechanisms of social cognition. Applied to genre this means that
genre information is claimed to influence the specific content of the maxims or
principles which govern the addressee's inferential recognition of the speaker's
intentions. Thus, genre recognition falls outside the domain of cognitive pragmatics
but interacts with it in subtle ways. (Kitis 1999; Holdcroft 1979; Green 1995) This is
3
illustrated in figure 1:
cog-prag explanations 03 - printout of 20.04.2002
Interpretive Hypothesis formation
Evaluation
Accepted interpretation
Conversational maxims /
Pra
g
matic
p
rinci
p
les
Genre
Socio-pragmatic
information
control
control
Figure 1
Domain of communicative
maxims/principle
= Domain of cognitive
pragmatics
comprehension procedure
Another answer may be that socio-pragmatic information is of a kind which feeds
naturally into the cognitive mechanisms underlying comprehension. The
comprehension procedure is governed by cognitive principles even when socio-
pragmatic information (e.g. genre) plays a role. Such information enters into the
cognitive comprehension procedure by providing contextual assumptions for the
inference processes rather than information for adjusting the principles governing
this process, and falls squarely into the domain of cognitive pragmatics. (Unger
4
2001a) This is illustrated in figure 2:
cog-prag explanations 03 - printout of 20.04.2002
Communicative principle
of relevance
Interpretive Hypothesis formation
Evaluation
Accepted interpretation
context
genre
socio/pragmatic
information
memory
perception
control
comprehension procedure
Domain of pragmatic principle(s)
=Domain of cognitive pragmatics
Figure 2
In the following section I will examine the first mentioned view in more detail.
3. Genre and conversational maxims
Grice (1975; 1989) argued that communication is regulated by a Cooperative Principle
which is further spelled out by more specific maxims of Quantity, Quality, Manner
and Relation:
Cooperative Principle
Make your conversational contribution such as is required, at the stage
at which it occurs, by the accepted purpose of direction of the talk
5
exchange in which you are engaged.
cog-prag explanations 03 - printout of 20.04.2002
Quantity
1. Make your contribution as informative as is required (for the
current purposes of the exchange).
2. Do not make your contribution more informative than is required.
Quality
1. Do not say what you believe to be false.
2. Do not say that for which you lack adequate evidence.
Relation
Be relevant.
Manner
1. Avoid obscurity of expression.
2. Avoid ambiguity.
3. Be brief (avoid unnecessary prolixity).
4. Be orderly.
(Grice 1989, 26-27)
Holdcroft (1979) was among the first to point out that the Cooperative Principle (and
hence by implication the other maxims) may not be operative in the stated form in all
discourse types. In highly constrained discourses, for example, where the participants
have unequal participation rights (e.g. in cross-examinations, interrogations and
interviews), the participants do not obviously cooperate in the usual sense of the
term. Holdcroft suggested a more abstract formulation of the Cooperative Principle
which can be made specific in different ways in different discourse types.
Make your contribution to the discourse such as is required, at the stage at
which it occurs, by the purposes you have in entering into, or which you
have accepted as the purposes of, or which are the generally accepted
purposes of, the discourse in which you are a participant. (Holdcroft 1979,
139)
Kitis (1999) expands on this suggestion. However, she takes the overarching principle
of conversation to be Relevance rather than Cooperation, where she takes Relevance
6
to be a social parameter governing expectations about how the conversation or talk
cog-prag explanations 03 - printout of 20.04.2002
exchange will proceed. These expectations are in turn related to the type of social
event in which the communication act occurs, so that Kitis' supermaxim of Global
Relevance could be paraphrased as be relevant to the goal or purpose of the present
social event type. (Unger 2001, 266).
Social events are mentally represented by means of frames or scripts (Kitis 1999,
650). These scripts, contain information about typical goals or purposes of the respective
events. These typical goals define the discourse type appropriate for the social event.
The discourse type in turn specifies which maxims are operative in it, and how they
differ in relative importance: ‘Global Relevance relates first the discourse type to the
social event (or domain) and only then are (aspects of) other maxims made contingent.’
(p. 652) In this theory, the precise instantiation of the conversational supermaxim is
different in different discourse types, and moreover different sets of maxims are
operative in each discourse type. Understanding the intentions of the speaker thus
depends on a recognition of the discourse type involved, since otherwise the maxims
for utterance interpretation were lacking.
Green (1995) turns to discourse type considerations in accounting for the distribution
of scalar implicatures. The idea is that the current goals and plans of the interlocutors
influence what inferences are licensed by the (quantity) maxims. Different types of
exchanges may have different plans and goals. In some cases, of course, the quantity
maxims might interact in a way that leads to the generation of scalar implicatures.
This is not so in others, and in these cases scalar implicatures will not be entertained.
In other words: different discourse types regulate indirectly how the conversational
maxims interact to generate implicatures, by specifying the nature of communicative
cooperation with respect to the requirements of the current talk exchange. (Unger
2001:269.) While on this account Grice's Cooperative Principle and maxims are in
force as originally formulated, the specific force of the Cooperative Principle is seen
7
as influenced by genre.
cog-prag explanations 03 - printout of 20.04.2002
Even though there are clear difference in the accounts of Holdcroft, Kitis and Green,
they do share the following architectural feature: a process of genre recognition must
precede the regular comprehension procedure because the genre determines the
specifics of the maxims or principles operative in this comprehension act. Genre and
other socio-pragmatic information is assigned a special place in communication,
outside the realm of cognitive pragmatics proper.
It is Holdcroft himself who pointed to a serious problem of accounts with this
property:
The claim that acceptance of [the Cooperative Principle] and of its generalized
version can involve the acceptance of different sets of maxims in different
discourse-types, and even within different discourses of the same type carried
on with rather different ends in view, is plausible, but frustrating. (Holdcroft
(979, 141)
What leads to this frustration is the following: if different maxims can operate on
different texts of the same type, then genre recognition does not provide a reliable
source for determining which interpretation of the communication maxims is to be
used. All explanation for utterance interpretation is then lost, both for discourses
following the standard type and for ones that deviate from it.
To illustrate this problem, consider two examples from Green (1995) of the same
genre, inquiry, which nevertheless lead to different effects with regard to scalar
implicatures:
(1) Suppose first of all that the hearer H has asked S, “How many children
does Nigel have?,” that H is to be understood as wanting an answer in
terms of an integer (rather than an answer of the form ‘between n and
m children’), and that no one has subsequently rejected the question or
changed the subject. The speakers are therewith engaged in an inquiry,
and hence being as informative as is required involves giving a complete
answer to the question that drives the inquiry if they can. In such a
case, if S does not make a stronger claim than that Nigel has fourteen
8
children (such as that Nigel has fifteen children), then the auditor may
cog-prag explanations 03 - printout of 20.04.2002
infer that S takes himself to be in no position to do so, and thus that S
knows that the stronger claim does not hold, or at least that S does not
know that the stronger claim holds. (Green 1995, 107)
Green's next example is also a case of inquiry, but no scalar implicature results
from a weaker statement in this case:
(2) Suppose on the other hand that S and H are jointly inquiring into the
question whether every adult male in the Shifflett clan has at least
fourteen children. Here once again the speakers are engaged in an
inquiry, but one that is driven by a different question from the one that
drove the conversation in our last case. In response to the question
whether every adult male in the Shifflett clan has at least fourteen
children S may say, “Nigel has fourteen children, and so does Jed and
so does Uriah.” The claim that Nigel has fifteen children is
conversationally relevant, since it straightforwardly implies that Nigel
has fourteen. Yet from the fact that S does not make this stronger claim
we cannot infer that he does not take himself to be in a position to do
so. (Green 1995, 108)
Thus the conversation type (characterized by the aim of the interlocutors) alone does
not determines the current requirements of the talk exchange, but other factors also
come into play.
Kitis (1999, 658) suggests that given an adequate discourse typology which is
properly based on social domains these problems will disappear. However, she also
contends that no discourse typology could cover every discourse:
One … must be aware that this approach offers only an initial and rather
limited navigational orientation as events are not fully, or even correctly,
represented in fossilized, static, abstracted, generalized and socially
normalized models, but are also dynamically adapted to individualized
cognitive interfaces … (Kitis 1999, 663)
But if every discourse can in principle follow 'individualized cognitive interfaces'
then Holdcroft's problem is obviously not remedied. I conclude that accounts of
9
genre which make the cognitive pragmatic comprehension procedure contingent on
cog-prag explanations 03 - printout of 20.04.2002
genre recognition as a separate process suffer from serious defects.
4. Genre and relevance-guided comprehension processes
According to relevance theory (Sperber & Wilson 1995), utterance interpretation is
based upon the tendency of human cognition to allocate processing resources to that
information which promises to maximize cognitive efficiency: to gain the greatest
(positive) cognitive effects for the least processing effort. This tendency is called the
cognitive principle of relevance. Since ostensive communication claims the addressee's
attention and thereby demands the investment of processing effort, the cognitive
principle of relevance predicts that a communicator can hope for successful
communication only if the stimulus chosen claims to be at least relevant enough for
the addressee to be worth the processing effort involved. Moreover, since human
cognition tends to maximize relevance, it is in the communicator's interests that the
level of relevance of the stimulus for the addressee is more than mere adequacy.
Thus, ostensive communication is characterized by a communicative principle of
relevance:
Every act of ostensive communication conveys the presumption that it is at
least relevant enough to be worth the addressee's attention and that it is also
the most relevant one the communicator could have made given her abilities
and preferences. (Adapted from Sperber & Wilson 1995)
This presumption is a member of the set of assumptions communicated by an ostensive
stimulus. It is a claim that calls for verification: is the stimulus indeed as relevant as
presumed? It is furthermore a general claim that licenses the addressee to expect a
certain level of relevance. These expectations relate more specifically to the stimulus
in question and are therefore more specific than the presumption of relevance on
which they are based. Since the most relevant stimulus is one that minimizes processing
effort (by the definition of the cognitive concept of relevance), the most obvious
procedure for testing the presumption of optimal relevance would be to access the
10
most easily accessible interpretation testing whether the level of relevance achieved
cog-prag explanations 03 - printout of 20.04.2002
by this satisfies the addressee's expectations of relevance. If not, the next easily
accessible interpretation is tested, and so on following a path of least effort. The first
interpretation which satisfies the addressee's expectations of relevance is the one
intended by the communicator.
To illustrate this relevance-theoretic comprehension procedure, consider the following
example:
(3) A: Did you have a good trip?
B: We were caught in a traffic jam near Mainz.
(Unger 2001a, 142 example (9))
B’s utterance is an indirect answer to A’s question. Indirect answers typically raise
quite specific expectations of the relevance of the response which can be satisfied
through the recovery of a small number of strongly communicated implicatures
(Sperber & Wilson 1995, 194-195). The following table presents a relevance-theoretic
analysis of this example. The left-hand column paraphrases the mental representations
entertained by A in the course of the interpretation process. The right-hand column
comments on the source and status of these representations:
11
cog-prag explanations 03 - printout of 20.04.2002
(4) (Unger 2001a, 142)
(a) B has said "We were caught in
a traffic jam near Mainz." Decoding of B's utterance.
(b) Information about whether B
had a good trip would be relevant
to A.
Assumption made mutually manifest by
A's utterance.
(c) B's utterance will be optimally
relevant to A. Presumption of optimal relevance,
communicated on the basis of the
communicative principle of relevance.
(d) B's utterance will be optimally
relevant to A by providing
information as to whether B had a
good trip.
Expectation of relevance created by A's
adopting the presumption of optimal
relevance and enriching it in the light
of the mutual cognitive environment.
(e) B and his passengers were
caught in a traffic jam near Mainz
on the occasion of the trip in
question.
Development of the logical form of B's
utterance. Accepted as the explicature
of this utterance.
(f) Being caught in a traffic jam is
an unpleasant experience. Encyclopaedic information. Accepted as
an implicated premise.
(g) If one had an unpleasant
experience on the road, one did
not have a good trip.
Encyclopaedic information. Accepted as
an implicated premise.
(h) B had an unpleasant
experience on the road. Follows from the explicature of B's
utterance and
(f)
.Im
p
licated conclusion.
(i) B did not have a good trip. Conclusion from (g) and (h) which
satisfies the expectation of relevance
(d) raised by B's utterance. Accepted as
an implicated conclusion.
The assumption in (4b) is highly accessible. It gives details about the kind of information
which A would find relevant if true. It is easy for A to combine this assumption with
(c4) (which is the presumption of optimal relevance communicated by the utterance
on principle) and form the belief in (4d), which amounts to a specific expectation of
relevance. The inference process illustrated in (4e-i) provides cognitive effects of the
expected kind and hence validate the presumption of relevance.
12
The sequence of presentation of the inference processes in this table does not imply
cog-prag explanations 03 - printout of 20.04.2002
serial succession of these processes. On the contrary, relevance theory claims that
content, context and cognitive effects are adjusted in parallel (Sperber & Wilson
1998). There are of course inferences which logically depend on a previous one, and
this constrains the extent of the parallel processing somewhat.
Expectations of relevance can be quite specific about the content and/or level of
relevance. They can also be more specific in another respect. Consider the following
examples:
(5) Mike to Sally: What do you think are the advantages and disadvantages
for us to move to Berlin?
Whatever Sally's answer might be, Mike will expect that it won't consist of just one
simple utterance. Rather, he expects that Sally will produce a complex stimulus
consisting of various utterances that will achieve relevance as a unit. Similar in the
following example:
(6) a master may explain to his apprentice how to exchange the shock
absorbers of a car, doing the action as he speaks. (Unger 2001a)
In this case, likewise, the addressee expects a complex stimulus which extends over
time and consists of sequences of utterances as well as of simultaneous non-linguistic
actions.
The point of these examples is that complex stimuli raise expectations about their
complexity, i.e. about how they will unfold over time. Texts/discourses can be regarded
as complex stimuli in this sense, that is, as a unit of communicative behaviour and
not a linguistic one.
We have seen in example (3) that highly accessible contextual information can be
accessed early in the interpretation process to fine-tune expectations of relevance.
Unger (2001) argues that genre information, too, can be easily accessed for fine-tuning
13
expectations of relevance in complex stimuli. To explain this, we must consider
cog-prag explanations 03 - printout of 20.04.2002
Sperber's (1996) account of culture in terms of an epidemiology of representations.
Cultural knowledge is a pool of mental representations widely distributed in a group.
Often, these representations are quite stable within the group in the sense of being
shared and transmitted over several generations. What makes a mental representation
into one that is successful in this way? The answer has to do with the human
tendency to pay attention to the most relevant information (in the technical sense). If
(a set of) representation(s) is useful for establishing the relevance of an input, it is
itself strengthened and becomes easier for subsequent access. This in turn means that
these representations are more readily accessed for processing further input. The
more a (set of) representation(s) is productive in cognitive processing of inputs in
this way, the more beneficial it will be for the individuals of this group to make sure
others share these assumptions. The most effective way to do this is to communicate
them. This in turn increases the accessibility of these assumptions. Thus, cultural
knowledge is information which is relatively easy to access on the grounds of the
role it plays in the search for relevance.
Recall now the relevance-theoretic comprehension procedure again. The first step is
for the hearer to adopt particular expectations of relevance. In the analysis of example
(3) in (4) we saw that highly accessible information can be used in this step. Genre is
cultural knowledge. We have just seen that cultural knowledge is typically highly
accessible. Hence it can be expected that this kind of information can be used for
fine-tuning of expectations of relevance.
Finally, let us consider in more detail the ways in which genre information may
fine-tune expectations of relevance of complex stimuli. Recall that expectations can
be rather general in that they relate just to the level of cognitive effects, or they can
be more specific, relating to the type of effects to be expected. In the case of
complex stimuli, expectations of relevance can also provide specific expectations as
14
to how the text will unfold.
cog-prag explanations 03 - printout of 20.04.2002
Now let us look at some particular genres:
(7) Seminar motion to establish a working group for research in translation
Whereas, the SIL International Translation Department sponsored a
Relevance Theory Update Seminar at Horsleys Green in November
2000,
And whereas, the participants unanimously expressed a desire for
the development of an ongoing Working Group for Cross-Disciplinary
Research in Translation,
And whereas, ongoing formal investigation and development of
translation theory is an imperative for SIL,
And whereas, the Translation Department has expressed a desire to
assume corporate leadership for this venture,
MOVED to request the Academic Services Division of SIL Academic
Affairs to endorse the genesis of this group, and to support its
development through the Translation Department.
Moved: NN
Seconded: NN
Unanimously carried: (all participants present).
(Unger 2001b; used by permission of the Int. Translation Department
of SIL International)
This is an example of the motion genre. Here, the genre raises very specific expectations
as to what utterances will be used when: whereas…moved to…moved by. This is a
case of a highly conventionalized genre. Similar ones are business letters, legal
contracts (Klinge 1998) or letters of recommendation of an employer for an employee
in German. In the latter case an elaborate system of conventional expression has
developed.
Some genre may be a little less detailed in the expectations they raise. Instead of
raising expectation about utterances or words to use in a particular sequence with a
particular meaning, the expectations may be regarding the sort of thing to be expected.
In a linguistic article, for example, the expectations will be that one finds an abstract,
15
an introduction, the main argument, a conclusion and a bibliography. Other examples
cog-prag explanations 03 - printout of 20.04.2002
of genres where the expectations about the unfolding of the text may be more or less
precise in this sense include sonnets, short stories, weather forecasts, news bulletins,
and advertisements (Unger 2001a, 295).
Yet another way in which genre may contribute expectations of relevance is exemplified
in (8):
(8) Isaiah 5:1-7 (Working translation of Unger (2001a); The Hebrew text
with interlinear glosses is presented in the appendix.)
(1a) I will sing to my friend/lover a song of my friend to his vineyard.
(1b) My friend had a vineyard on a fruitful hilltop (2a) and he digged it
up and removed its stones and planted the best species of grapes in it
(2b) and he built a tower in its midst, and also a winevat he hewed into
it. (2c) And he waited for the vineyard to bring good grapes, but it
brought wild (bitter) ones. (3) And now, inhabitants of Jerusalem and
people of Judah, judge between me and my vineyard. (4a) What else
was there to do for my vineyard which I haven’t done, (4b) why did I
expect good grapes and it brought forth sour ones? (5a) And now I will
let you know what I will do to my vineyard. (5b) I will remove its
hedge and it will become firewood. (5c) I will tear down its walls and
it will become a trampled land. (6a) I will bring about its peril, it will
not be pruned and not be hewn, and in it shall grow thorns and thornbushes.
(6b) And I will command the clouds to not let rain fall on it (or: …the
clouds to refrain from letting it rain over it) (7) For the house Israel is
the vineyard of the Lord of hosts, and the people of Judah are the
planting of his delight, and he waited for justice, but there was only
injustice (or bloodshed), for righteousness, but there were only cries.
To understand this text it is crucial to recognize that it is an instance of the Hebrew
love-song genre. This genre makes use of typical symbols (or allegories): orchard or
vineyard typically symbolizes the female partner, the gardener or owner of the vineyard
or orchard the male partner, and the vineyard-owner’s action of toiling and caring for
the vineyard is a symbol for the male partners’ expressions of love to his partner.
This genre information has important consequences for comprehension. Among these
are the following: the assumption that the text is going to be a love-song causes the
16
reader to assign the reference of the first person singular in (5:1a) to a fictitious
cog-prag explanations 03 - printout of 20.04.2002
female speaker rather than the prophet Isaiah. Also, the reader is encouraged to seek
the relevance of the text in analogies between what is said about the vineyard and a
human love relationship. These analogies will allow the reader to appreciate the
emotional impact of the text. (A detailed analysis of this text along these lines is
given in Unger 2001a:pp.) In other words: genre information for Hebrew love songs
gives the addressee clues as to which cognitive effects certain symbolic expressions
will have, i.e. how to interpret them. The expectations are not about form but about
content.
On this account, genre information enters the comprehension procedure as integral
part of the parallel adjustment of context, content and contextual implications which
is governed by the communicative principle of relevance. The socio-pragmatic
phenomenon of genre receives an explanation in cognitive pragmatic terms. This
architecture is different from the approaches discussed in the previous section and
does not share their problems.
Another point that should be noticed about this relevance-theoretic account of genre
is that it motivates the existence of genre: since communication is driven by the
search for optimal relevance, communicators do well to rely on contextual assumptions
which help the addressee to adjust their expectations of relevance of complex
stimuli. By the same token, it is advantageous that genre concepts should emerge.
Lastly, I should point out that it doesn't have to be genre information that contributes
to the fine-tuning of expectations of relevance, as was seen in (4). Indeed, genre does
not always enter into comprehension. Also, expectation of relevance are constantly
adjusted over time during the on-line processing of discourse. In this process, genre
information may give way to other assumptions for this purpose. We can see for
example in example () that the text in later parts talk about divorce rather than love,
and that the vineyard- and vineyard owner symbolism come to carry a different
17
meaning than in the beginning. Genre is not a straight-jacket for discourse fixed once
cog-prag explanations 03 - printout of 20.04.2002
for the whole unit.
5. Beyond genre: extending the relevance-theoretic account of genre to other socio-
pragmatic phenomena
The above relevance-theoretic account of genre was based on the idea that genre
information is typically easily accessible contextual information by virtue of its
epidemiological distribution in a people group. Notice that cultural information other
than genre information has this property as well. Hence it can be expected that
cultural information other than genre information can be used for fine-tuning
expectations of relevance as well. In this section I want to discuss two such cases:
register and cultural preferences of expression.
5. 1 Register
The concept of register was introduced to describe variations in linguistic expression
conditioned by the social setting of the communicative act. For example, in a lecture
about pragmatic theory I may use the term "ostensive-inferential communication." In
a conversation about the same topic, on the other hand, I may prefer to use the
expression "communication by obviously intentional behaviour" instead, the latter
being more in line with everyday vocabulary. The idea of the notion of register is
that in the field "technical communication", the mode "prepared oral communication
with limited audience feedback" and the tenor "technically precise expression
preferred", technical terms (such as "ostensive-inferential") are more readily expected
then in conversations about the same topic (which could classified as field "technical
communication", mode "spontaneous oral communication with feedback" and tenor
"down-to-earth explanation"), and vice versa. In other words: the linguistic choices
available to the communicator and expected by the addressee are narrowed down in
18
different social settings in different ways.
cog-prag explanations 03 - printout of 20.04.2002
Register is seen as differing from genre in the following ways: register specifies a
range of linguistic choices (the available "repertoire") which are preferred in the
given communicative setting. Genre, on the other hand, specifies expectations about
discourse organization: the schematic structure. (In the relevance-theoretic account,
genre can also specify expectations about content or interpretive richness of texts,
see above.)
In relevance-theoretic terms, register can easily be account for as providing information
about the communicator's preferences. Such information may be associated with a
cultural concept which is not specific to a certain type of communication. Thus, the
information "the communicator has a preference for using technical vocabulary" can
be stored as an encyclopaedic entry in the concept ACADEMIC ENDEAVOUR.
In this way, register information is easily accessible for fine-tuning expectations of
relevance in the same way as explained for genre. The way in which register and
genre enter into comprehension are the same. This sheds light on the intuition that
register and genre are closely related concepts. The differences between register and
genre are purely descriptive: register information is about communicator's preferences
and may be stored under general cultural concepts, whereas genre information is
about all kinds of expectations of relevance in complex stimuli and presumably
stored under genre concepts. This sheds light on the difficulty in distinguishing
between the two, as evidenced in the literature. From the perspective of explaining
communication (i.e. that of pragmatic theory), the difference does not matter. It may
matter for ethnomethodological (socio-pragmatic) description, though.
5. 2 Cultural preferences of expression
An other socio-pragmatic phenomenon can be found in culturally defined preferences
of expression. An example of this can be the differences in the usage of preparatory
19
questions in Behdînî-Kurdish on the one hand, and in English and German on the
cog-prag explanations 03 - printout of 20.04.2002
other (Unger 1994). In Behdînî-Kurdish, preparatory questions such as in (9) are
used to a much greater extent than in German or English:
(9) Isn’t there Bêgova? Half of it’s inhabitants belong to the Berwarî tribe
and the other half to the Gulî tribe. (Unger 1994; 2001a)
How can this difference is degrees of usage be explained? I suggested that an explanation
might run as follows: there is a cultural maxim Don’t show superiority to others
which is much more prominent in Kurdish culture than in German or English culture.
Thus, it is highly relevant for Kurdish speakers to often communicate compliance
with this cultural norm, much more so than for English or German speakers. One
way of doing this efficiently is by using the question form, since (sincere) questions
typically show that the communicator is not superior in knowledge to his addressee.
In contexts where the speaker can be justified in assuming that the audience can
easily infer the speaker’s intentions behind the preparatory question (i.e. when they
are not likely to mistake the preparatory question for a real one), the use of such a
question may achieve too things simultaneously, thus saving the audience unnecessary
processing effort: it gives evidence of the speaker's informative intention as well as
of his intention of communicating compliance to social norms. The audience is
thereby also saved processing effort.
6. Conclusion
In this paper I discussed two types of pragmatic accounts of genre: one in which
genre the contribution of genre to comprehension is seen as separate from the cognitive-
pragmatic comprehension procedure, and one where genre recognition is an integral
part of it. I have argued that the former meets severe problems whereas the latter
provides a cognitive pragmatic explanation of the role of genre in communication.
This account naturally extends to other socio-pragmatic phenomena such as register
and cultural differences in ways of expression. Given that genre is a prototypical
20
socio-pragmatic category and that my relevance-theoretic account extends to other
cog-prag explanations 03 - printout of 20.04.2002
socio-pragmatic categories as well, my conclusion is that socio-pragmatic phenomena
in general receive a cognitive pragmatic explanation in relevance-theoretic pragmatics.
In the introduction I said that pragmatic theory surely needs to deal both with socio-
pragmatics and cognitive pragmatics. That relevance theory opens up genuine cognitive
explanations for socio-pragmatic phenomena shows that this theory is well suited to
give a fairly comprehensive account of pragmatics. This means that relevance-theoretic
research can fruitfully interact with socio-pragmatic or ethnomethodological enquiries
in a new level of cross-disciplinary interaction.
7. References
Bex, A. R. 1992: ‘Genre as context,’ Journal of literary semantics. 21, 1-16.
Downing, A. 1996: ‘Register and/or genre?’ In Vazquez, I. & A. Hornero (eds)
Current issues in genre theory. Zaragoza: MIRA editores. 11-27.
Goatly, A. 1994: ‘Register and the redemption of relevance theory,’ Pragmatics.
4(2), 139-182.
Grice, H. P. 1989: Studies in the way of words. Cambridge MA: Harvard University
Press.
Green, M. S. 1995: ‘Quantity, volubility, and some varieties of discourse,’
Linguistics and Philosophy. 18, 83-112.
Eggins, S. & J. R. Martin 1997: ‘Genres and registers of discourse,’ In T. A. van
Dijk (ed.) Discourse as structure and process. London: SAGE publications.
230-256.
Holdcroft, D. 1979: ‘Speech acts and conversation - I,’ The Philosophical Quarterly.
29, 125-141.
Jary, M. 1998: ‘Relevance theory and the communication of politeness,’ Journal of
21
Pragmatics. 30, 1-19.
cog-prag explanations 03 - printout of 20.04.2002
Kitis, E. 1999: ‘On relevance again: From philosophy of language across
“Pragmatics and power” to global relevance,’ Journal of Pragmatics. 31, 643-
667.
Paltridge, B. 1995: ‘Working with genre: a pragmatic perspective,’ Journal of
Pragmatics. 24, 393-406.
Sperber, D. 1996: Explaining culture. Oxford: Blackwell.
Sperber, D. & D. Wilson 1995: Relevance. first edition 1986, Oxford: Blackwell.
Sperber, D. & D. Wilson 1998: ‘The mapping between the mental and the public
lexicon,’ In Carruthers, P. & J. Boucher (eds) 1998: Thought and language.
Oxford: Oxford University Press., 184-200.
Swales, J. M. 1990: Genre analysis: English in academic and research settings.
Cambridge: Cambridge University Press.
Unger, C. 1994: ‘Aspects of the Dialect of Behdinan,’ Paper read at the Kurdistan
Forum of the School of Oriental and African Studies, London.
Unger, C. 2001a: On the cognitive role of genre: a relevance-theoretic perspective.
University of London PhD thesis.
Unger, C. 2001b: 'Genre and translation.' Revista Alicantina de Estudios Ingleses 14.
Special issue edited by José Mateo Martinez & Francisco Yus devoted to
Pragmatics and Translation. 297-321.
Vazquez Orta, I. 1996: ‘Register, genre and linguistic choice,’ In Vazquez, I. & A.
Hornero (eds) Current issues in genre theory. Zaragoza: MIRA editores. 29-50.
Zegarac, V. & B. Clark 1999a: ‘Phatic interpretations and phatic communication,’
Journal of Linguistics. 35(2), 321-346.
Author’s address
Christoph Unger, SIL International
Mozartstr. 26
35625 Huettenberg
22
Germany
cog-prag explanations 03 - printout of 20.04.2002
23
Email: christoph-kuelvi_unger@sil.org
