“… Prose discourse—in any of its forms,
quotidian, rhetorical, scholarly—cannot fail to be oriented
toward the ‘already uttered,’ the ‘already
known,’ the ‘common opinion’ and so forth.”
Mikhail Bakhtin, “Discourse in the Novel”
(1981:279)
INTRODUCTION
When speakers voice an opinion on a topic about which there has been
prior discourse or tacit agreement in their opinion community, they are
expected to mark the cultural standing of that opinion.
“Cultural standing” is my label for the location of a view
on a continuum that ranges from highly controversial to completely
taken for granted in the relevant opinion community. Cultural standing
has been insufficiently discussed in pragmatics, sociolinguistics, and
discourse analysis as a constraint on opinion display. Expression of
opinion has been analyzed primarily from three perspectives: how
conversations are structured (conversation analysis; e.g., Pomerantz 1984, Sacks
1987); how social relations of power and solidarity affect the
expression of potentially offensive views (politeness theory; e.g.,
Brown & Levinson 1987); and what evidence
speakers have for their views (Chafe & Nichols
1986). In this article I propose a Bakhtinian perspective (Bakhtin 1981) that looks at how speakers respond
not only to evidence and to the immediate addressee, but also to
previous social commentary on a given topic.
The following thought experiment may provide an intuitive grasp of
the pragmatic principle that we should indicate correctly the perceived
cultural standing of our opinions. Imagine someone who expresses a
controversial or discredited view as if it were the common opinion or
taken for granted – for example, Since the earth is flat, it
follows that … Or imagine the opposite, someone who
expresses what is taken for granted as if it were controversial:
You may think this is crazy, but you could almost say that it is
important to be kind to others. There are probably misanthropic
flat-earth adherents among whom the first proposition is the common
opinion and the second one is not, but that is not the point here.
Imagine hearers who do not think the earth is flat and just take for
granted, rhetorically at least, the value of being kind to others. Such
hearers would judge these speakers to be culturally or mentally
incompetent, or – if the speaker appears to be a properly
enculturated adult in possession of his or her faculties – the
hearers would draw the implicature that they must be speaking
ironically, metaphorically, wishfully, or playfully, because they could
not possibly be serious. This shows the power of the expectation that
speakers should correctly mark the perceived cultural standing of their
opinions.1
In the flat earth example,
clearly part of the problem is the view that is held, not just how it
is worded. But wording makes a difference. If the view had been marked
as controversial by attributing it to another, less responsible version
of the self (In my stranger moments I almost think that
…) or acknowledged as controversial or attributed to another
source (I know it's hard to believe, but I've heard some
authorities really think that….), we would be more inclined
to take S seriously.
More attention to cultural standing would benefit research in both
language and culture. For years, anthropologists tended to focus on
highly sedimented world views – those that are taken for granted
or are the consensual common opinion. Recently, the tendency has been
to focus instead on contested views. Clearly it is necessary to
recognize both: In a given society at a given time, some ideas are up
for grabs and others are more settled. (See Bourdieu 1977:168; Williams 1977:121–7; Comaroff & Comaroff 1991:19–32; Strauss & Quinn 1997: 36–41.) When a view is thought to have
high cultural standing, so that it is generally considered to be the
opinion that most people hold, it can be more powerful than the views
that most people truly do hold, and the views that everyone thinks are
controversial will be uttered only furtively if at all, reinforcing the
view that no one holds them (Noelle-Neumann
1993). Paying attention to markers of cultural standing would be
a way to explore this dynamic in a society at a given time, or to trace
how the standing of a given idea has changed over time.
For sociolinguists and discourse analysts, attention to cultural
standing would bring together many unconnected bodies of research. It
would also shed light on aspects of discourse that have not been
adequately explained by other approaches. In addition to conversation
analysis and politeness theory, important contributions to the study of
how opinions are expressed have been made by analysts looking at
particular genres (e.g., argumentation and focus group conversation),
particular discourse features (e.g., discourse markers like you
know, evidentials, hedging, presuppositions, formulaic speech),
particular topics (e.g., race and immigration), and related functions
(e.g., epistemic modality). In this article, I show that some of these
observations can be integrated in a model of markers of cultural
standing, and others can be enriched by the model.
The article has four main sections. First, I review previous research
to show that while cultural standing analysis owes a great deal to
previous work, it is a different pragmatic constraint than the
interpersonal considerations of politeness analysis or the epistemic
factors marked by evidentials. After the literature review, I outline
the cultural standing model and discuss the kinds of data I draw upon
for evidence. There follows, in the longest section, a detailed
description of cultural standing markers in American English. In the
Conclusion, I discuss the cross-cultural applicability of the model as
well as other ways in which context may affect its application.
RELATED APPROACHES
A review of conversation analysis, politeness theory, and critical
linguists' discussions of epistemic modality will clarify both
what the proposed model owes to previous discussions and what it can
add to them. Other pertinent analyses (e.g., of argumentation or
hedges) will be discussed as they become relevant to particular aspects
of the cultural standing model.
Conversation analysis
Conversation analysts have discovered a preference for agreement in
conversation, at least among English speakers.2
Wierzbicka (2002:240–1)
claims this is an Anglo trait not found in many other European
countries.
If Speaker A offers an assessment and Speaker B
disagrees, B will try to avoid blatant disagreement. These analysts
have perceptively noticed some of the strategies used, such as replying
with a weak affirmation before disagreeing (
Yes but), or
otherwise delaying the potentially offending demurral with pauses or
prefatory comments (
Well …) (
Pomerantz 1984,
Sacks
1987). Another way in which hearers respond to controversial
opinions, they note, is through silence or “next-turn repair
initiators,” which invite the speaker to back down or correct a
questionable statement (
Schegloff, Jefferson &
Sacks 1977,
Pomerantz 1984). However,
conversation analysis is restricted in its ability to describe and
explain markers of cultural standing because of its singleminded focus
on the here-and-now of conversational interchange (as many critics have
noted, e.g.
Hymes 1974:81).
3 Hymes's discussion does not refer to conversation
analysis explicitly. I have taken that interpretation from Duranti
(1997:265).
In the simplified world
of conversation analysis, speakers do not anticipate whether a hearer
will agree or disagree on the basis of previous knowledge about the
hearer or people like him or her. Instead, they react to the
hearer's subtle indications of agreement or disagreement. This
misses the preventive work that speakers do based on the cultural
models they bring to the conversation.
Cultural models are learned, shared mental representations or
schemas: both factual and value-laden assumptions about the world
(Holland & Quinn 1987, D'Andrade & Strauss 1992, Strauss & Quinn 1997). As cultural models
theorists and other theorists of pragmatics have observed, for speakers
these include assumptions about the hearer's beliefs (Sperber & Wilson 1986; D'Andrade 1987:113). Assumptions about
hearers' views could be based on things they have said before.
Frequently, however, assumptions about hearers' opinions are based
on cultural models or schemas of what a person like that would think;
as stereotypes based, for example, on clothing, age, sex, ethnicity,
educational level, or lifestyle. These beliefs undergird the
anticipatory work speakers do in conversation, modifying both how
speakers phrase their assessments and whether they venture an
assessment at all. As described further below, speakers may not offer
an opinion in the first place if they suspect the hearer will view the
statement either as obviously true or as outrageously false (see also
Bourdieu 1977:18).
Politeness theory
Although Penelope Brown and Stephen Levinson's theory of
politeness (1987) makes use of conversation
analysis, they add to it the important element of the social relations of
the speaker and addressee. Drawing on Goffman 1967,
they argue that much of ordinary conversation is motivated by the
speaker's concern for the addressee's “face
wants.” People everywhere, they say, want their actions to be
unimpeded and (the point that is most relevant here) want their
“wants [to] be desirable to at least some others”
(Brown & Levinson 1987:62). To put the
second point more colloquially, they want their identities, values, and
choices to be validated by others. This is relevant to the expression
of opinions. Brown & Levinson note that one possible
“face-threatening act” (FTA) is “contradiction or
disagreement” (p. 66) with the other's opinion. Unless the
speaker (S) can afford to offend the hearer (H) (because S is much more
powerful than H, or S does not need H to do a big favor for him or
her), S will be careful about expressing opinions with which H might
disagree. One way in which speakers avoid the FTA of disagreeing with
their listeners is by discussing only “safe topics,” such
as “the weather … the beauty of gardens, the incompetence
of bureaucracies … and the irritations of having to wait in
line” (112). Drawing on the work of conversation analysis, Brown
& Levinson also mention strategies of softening dissent, as well as
of hedging one's opinions with vague language such as sort of,
kind of, like, in a way (as in I don't know, like I
think…) (112–116).
Brown & Levinson do not discuss the schemas that allow speakers
to anticipate what hearers will find offensive, but they acknowledge
that such anticipation occurs. Thus, they note that “the more S
knows about H, the more close to home will be the safe topics he can
pursue with H” (112). It follows that sticking to conventionally
safe topics is not always the most effective positive politeness
strategy (i.e., one that highlights solidarity between S and H).
Instead, if you and your interlocutors see yourselves as belonging to a
minority opinion community, it enhances solidarity to venture an
opinion that is contested in the wider society but is the common
opinion among people like you.4
This would
be a subtype of Brown & Levinson's positive politeness
strategy 4, “Use in-group identity markers.”
This
is much more effective than talking about safe topics like the weather
as a way of satisfying listeners' desires to be validated in their
opinions.
5 This seems obvious, but it is
missing from Brown & Levinson's analysis. Thus, they give as
examples of acts “that show that S doesn't care about (or is
indifferent to) H's positive face” “raising of
dangerously emotional or divisive topics, e.g., politics, race,
religion, women's liberation” (pp. 66, 67). Whether those
topics are divisive, of course, depends on their standing in S and
H's opinion communities.
Similarly, negative politeness
requires that you avoid comments that you suspect have low cultural
standing in H's opinion community. Of course, if contradicting the
other's opinion is not a serious FTA in S and H's speech
community, then these strategies are less needed, although cultural
standing may still be indicated in other ways.
In certain respects, the cultural standing framework developed here
differs from Brown & Levinson's politeness theory. A minor
problem is that, as Chen 2001 points out,
Brown & Levinson focus on the way speakers protect the
hearer's face, while largely ignoring the way speakers also
protect their own face (Goffman 1959). This
is particularly relevant to the expression of cultural standing. In
marking cultural standing, speakers are not just concerned to avoid
offending the hearer. They are also protecting themselves from the
judgment that they are unaware of or insensitive to group opinion. (See
van Dijk 1987:86 on the importance of
managing self-presentation in talk about ethnic minorities.) Chen 2001 shows that some modifications of Brown
& Levinson's politeness strategies are needed to accommodate
what he calls “self-politeness.” Still, I agree with Chen
that speakers' concern with protecting their own face can be
accommodated by an expanded politeness theory. (Brown & Levinson
raise this issue themselves, 1987:67–8.)
A more serious problem with politeness theory is that speakers are
not always rationally attuned to face wants (whether H's or their
own) in a particular context. Sometimes we are creatures of habit, and
the way we talk about a topic is due to the way we usually hear that
topic discussed, and the way we have discussed it before, rather than
to a strategic assessment of what is advantageous. For example, in the
following excerpt from a series of interviews I conducted on
political-economic issues, compare the way George Gauvin,6
All names of interviewees and focus group
participants are pseudonyms.
a factory worker in his sixties,
talks about “big money” and “the rich” in the
first five pause groups versus the last. The topic was corporate tax
loopholes. (Boldface indicates my emphasis; numbers in parentheses are
pause lengths in seconds; my backchannels and other contextual
information are in double parentheses; unintelligible portions of the
tape are given by empty parentheses.
7 My
transcription conventions are as follows:
):
(1) Big money, boy. ((Mm)) (1.0) Can't beat it. (1.2)
((Mm)) Can't beat big money. (3.0) ((Mm, I know)) The rich get
richer and the poor get poorer. (1.8) It's true. (1.8) ((Mm))
You probably got () big money, I shouldn't talk like
that.
The last line reflects politeness considerations. Suddenly it
occurred to Gauvin that what he had said could have been offensive to
me, so he apologized in a blatant concession to what he guessed to be
my face wants. The blunt wording of the previous lines, by contrast,
was not shaped by his speculation about my opinion but by the way that
idea is commonly expressed in his opinion community. This view is in
fact the common opinion among working-class speakers of his generation
(as I found in my research), and is marked as such by the use of
maxims, clichés, and other kinds of formulaic speech that
express generalizations without any mitigation.
On the whole, however, politeness and cultural standing models are
complementary, in two senses. First, neither model is complete without
the other. Cultural standing considerations affect speakers'
judgments about what would be considered a possible FTA in the
expression of opinions, and negative and positive politeness strategies
for mitigating FTAs, while politeness considerations help explain why
cultural standing is marked in discourse. Second, politeness and
cultural standing models are complementary in the sense of focusing on
different aspects of the context: Politeness is a social
phenomenon, while cultural standing is a cultural phenomenon.
In other words, while politeness markers reflect the interpersonal
relations of the speaker and addressee, cultural standing markers
reflect the background of values and beliefs in a community. Social and
cultural aspects of the context are both important, as will be
discussed further below, but some situations highlight considerations
of politeness and others highlight those of cultural standing. If the
situation is one that highlights considerations of cultural standing,
politeness analysis will give a less incisive analysis of the
discourse, and vice versa.
Bakhtin, critical discourse analysis, and epistemic
modality
More than 60 years ago, Mikhail Bakhtin anticipated the main lines of
my remarks above when he pointed out that there is a difference between
the way the speaker accommodates to the listener and the way the
speaker reacts to previous discourse about the object: “They
differ in their essentials and give rise to varying stylistic effects
in discourse” (Bakhtin 1981:283).8
8Bakhtin called reaction to
previous discourse about the object “the dialogic relationship toward
an alien word within the object” (1981:283).
Anticipation of the listener's response and reaction to previous
discourse are two varieties of what he called “internal
dialogism,” i.e., dialogism that is not part of an obvious dialogue
between two characters.
This key insight, along with Bakhtin's
attention to the way all discourse is “oriented toward the
‘already uttered,’ the ‘already known,’ the
‘common opinion’ and so forth” (
1981:279), is the starting point for the cultural standing
model developed below.
Also key to this analysis is the insight of Bakhtin and many other
theorists that the cultural background of what has been previously
thought and said is not fixed. What was the common opinion or even
taken for granted at one time can be contested at a later time –
or the reverse (e.g., Bourdieu 1977:168;
Williams 1977; Bakhtin 1981:345–8).
The way shifting ideologies are reflected in discourse is the focus
of discussions of modality by critical linguists (e.g., Kress & Hodge 1979, Hodge
& Kress 1988) and critical discourse analysts (e.g., Fairclough 1992:158). Lyons's definition of
epistemic modality is a useful starting point:
Any utterance in which the speaker explicitly qualifies his
commitment to the truth of the proposition expressed by the sentence he
utters, whether this qualification is made explicit in the verbal
component, as in [He may have gone to Paris, Perhaps he went
to Paris, It's possible that he went to Paris] or in the
prosodic or paralinguistic component, is an epistemically modal, or
modalized, utterance. (Lyons 1977:797)
(See also Halliday 1976:192 on modality
defined as “the speaker's assessment of the probability of
what he is saying, or the extent to which he regards it as
self-evident.”) For critical linguists and discourse analysts,
“Modality points to the social construction or contestation of
knowledge-systems. Agreement confers the status of
‘knowledge’, ‘fact’ on the system, or on
aspects of it; lack of agreement casts that status into doubt …
Modality is consequently one of the crucial indicators of political
struggle” (Hodge & Kress 1988:123).
These theorists note that speakers cannot avoid indicating their
commitment to the propositions they state: “In any propositional
utterance, the producer must indicate what Hodge & Kress (1988:123) call a degree of ‘affinity’
with the proposition” (Fairclough
1992:158).
Outside of critical discourse analysis, there has also been
considerable work on epistemic modality in the growing literature on
evidential markings, the ways in which speakers mark the type of
evidence they have for their assertion and the degree of certainty they
give it (e.g., Chafe & Nichols 1986). For
example, Talmy Givón claims that evidentials in KinyaRwanda,
Sherpa, and Ute indicate a universal tendency to distinguish the
following three sorts of propositions:
(a) Propositions which are to be taken for granted,
via the force of diverse conventions, as
unchallengeable by the hearer and thus requiring no
evidentiary justifications by the speaker;
(b) Propositions that are asserted with relative
confidence, are open to challenge by the hearer and thus
require – or admit – evidentiary justification; and,
finally,
(c) Proposition[s] that are asserted with
doubt, as hypotheses, and are thus beneath both
challenge and evidentiary substantiation. They are, in terms of the implicit
communicative contract, “not worth the trouble.” (1982:24)9
I am grateful that
Jane Hill alerted me to Givón's work after this article was
accepted for publication but while there was still time to incorporate it into
my revisions.
The cultural standing model presented below builds on this work by
distinguishing cultural standing from interpersonal (politeness) and
epistemic factors affecting speakers' commitment to an utterance.
Previous work on modalization does not distinguish among these, but it
is useful to do so because modalizers have different pragmatic meanings
depending upon which of these functions (epistemic, politeness,
cultural standing) they serve in a given context. Compare examples
(2–4). In each case, the speaker modalizes, but the way in which
this is to be interpreted differs. (Modalizers are in boldface.)
Epistemic modalization
(2) Discussion about whether our college-age daughter, who
was not present, would join us for a family vacation the following
summer.
CS: Po:ssibly Rachel will come.
In this case I (CS) indicate low commitment to the proposition
because it is hard to predict the summer vacation plans of a
19-year-old a year in advance. I doubted that Rachel would come, but I
could not say for sure because I did not know. I will reserve the term
“epistemic modalization” for the speaker's
qualifications regarding the certainty of the facts.10
For the most part, Labov & Fanshel's (1977:100) classification A-events (known to A, but
not to B), B-events (known to B, but not to A), etc., focuses on
epistemic concerns as well. Their categories of AB-events (known to
both A and B), O-events (known to everyone present), and D-events
(known to be disputable) are relevant to the present discussion, but
their primary concern is with how interlocutors are supposed to respond
to each of these, not how speakers frame them.
Modalizing for politeness
(3) Discussion with a cab driver (CD) about construction
across the street from my house. The driver had an unobstructed view of
the work for several minutes before I (CS) got into the cab.
CD: They're building a new wall.
CS: I thought they were doing something to the
street.
CD: They're building a wall. Or–whatever.
They're building something. God only knows what
they're building.
In this case, the problem is not insufficient knowledge: The cab
driver has no doubt that they are building a wall. The pragmatic
meaning of the modalizers (whatever, something, God only
knows) is “I don't want to get into an argument about
it with a customer.”
Modalizing to express cultural standing
(4) The last sentence of a U.S. News & World
Report article on the health effects of the drug
ecstasy.
Evidence suggests that perhaps some day a
designer drug based on ecstasy may be developed that
could provide potential psychological
benefits—without the scary side effects. (Boyce 2001)11
In
addition to the obvious hedge words highlighted above, a designer
drug based on ecstasy and without the scary side effects
also indicate low speaker commitment to the view that ecstasy itself
has benefits.
One reason for the extensive modalization in (4) is that the facts
are uncertain (epistemic modalization). But another reason is that the
writer must be aware that it is controversial, in the opinion community
represented by U.S. News & World Report readers, to say
that illegal drugs could have benefits. Her modalization reflects her
awareness of the view's weak cultural standing.
As the last example indicates, sometimes a speaker's commitment
to a proposition is affected by a combination of epistemic, politeness,
and cultural standing considerations. Different situations bring
different combinations of these considerations to the fore, and it is
important to think about which are most significant in a given
instance. Academic journal articles, for example, are a genre in which
the cultural standing of the author's views in the relevant
community of scholars is a highly salient constraint, at least as
salient as epistemic concerns and more so than interpersonal ones. Even
when most concerned about possible FTAs (e.g., thinking about the
opinions of the editor or potential reviewers), scholars anticipate a
response on the basis of schemas about how controversial or accepted
their views are in various scholarly communities. The choice of a
politeness or cultural standing analysis of academic writing makes a
difference when one is analyzing the significance of intensifiers such
as really, actually, obviously, surely and indeed in
this context. Hyland 1998 and Myers 1989 argue that, in academic articles,
intensifiers are “positive-politeness devices, enabling writers
to assume shared ground with their readers and stress common group
membership” (Hyland 1998:353).
The cultural standing model presented below, however, gives a very
different interpretation of intensifiers. In a context of opinion
display, intensifiers are markers of debate because the speaker
intensifies to strengthen the argument in the face of anticipated
criticism (see Schiffrin 1985:40). That seems
a better analysis of the examples of intensifiers in academic writing
that Hyland gives (e.g., This brings us into conflict with
Currie's account, for static images SURELY cannot trigger
our capacity to recognize movement, quoted in Hyland 1998:350).
Similarly, hedges are very common in academic writing not only
because we are expected to be careful about acknowledging the limits of
what can be concluded from our data (epistemic concerns) and because we
want to avoid personal offense (politeness concerns), but also because
we are expected to acknowledge that there are competing views in the
literature (cultural standing concerns, as Hyland also recognizes;
e.g., 1998:35112
Hyland's insightful discussion shows awareness of cultural
standing issues: “It is clear in the following extracts, for example,
what propositions the writers consider to be established knowledge, and what
they regard as more contentious” (1998:351);
and “By employing markers of evidentiality with inclusive pronouns, for
instance, writers are able to construct a shared context with their readers
and draw on assumed beliefs specific to their particular social group”
(1998:367). Lacking a general model of cultural
standing, however, he relies more heavily on politeness theory than his data
warrant.
; see also
Salager-Meyer et al.
1996:164).
A similar caution should be raised about Givón's analysis
of evidentials. While the evidentiary justification needed for
propositions that are “taken for granted” and those that
are “asserted with relative confidence” apply as well to a
cultural standing analysis, there is a significant divergence between
propositions he describes as “asserted with doubt, as
hypotheses” (because they have weak epistemic justification) and
those with weak cultural standing. Propositions asserted with doubt
because they have weak epistemic justification are “beneath both
challenge and evidentiary substantiation” because speakers
realize they are “ ‘not worth the trouble’,”
according to Givón (1982:24). Quite
the opposite happens in the case of a proposition that is controversial
because of its weak cultural standing: As I will show below, in such
cases speakers engage in extensive preventive measures because they
know the statement may well be challenged.
Fundamentally, discussions of evidentials (expression of
speaker's evidence) and modalization (expression of speaker's
commitment or detachment) do not account for markers of cultural
standing (expression of speaker's view of the acceptability of an
idea in the salient opinion community) because these are different
phenomena. Speakers can simultaneously express high personal commitment
and low cultural standing, or the reverse, as in this so-called
improvement in the standard of living (quoted in Stubbs 1996:208).13
Thus, although Stubbs's definition of modality in
terms of commitment or detachment, which “concerns whether a
proposition is presented as true, false, self-evident, a matter of objective
fact or of subjective opinion, shared knowledge, taken for granted or
debatable, controversial, precise or vague, contradictory to what others
have said and so on” (1996:204–5)
overlaps with the notion of cultural standing, they are not the same.
In this example,
so-called simultaneously indicates high cultural
standing (it is widely said that there has been an improvement) and low
speaker adherence.
THE CULTURAL STANDING MODEL
Some of the main points of my model of cultural standing are as
follows. Cultural standing is a continuum, ranging from the highly
controversial to what is so widely accepted that it is not even
acknowledged as a proposition about which one could have alternative
opinions. This model was greatly influenced by Pierre Bourdieu's
discussion of the difference between dogma (the “universe
of discourse [or argument]”) and doxa (the
“universe of the undiscussed [undisputed]”;
1977:168). He further divides the realm
of dogma into orthodoxy (the dominant ideology) and heterodoxy (views
that contest the reigning orthodoxy). It is important to keep in mind
the distinction Bourdieu makes between the reigning orthodoxy, which is
widely believed in an explicit way, and what is doxic, which is so much
taken for granted that it does not need to be spoken, because in
cultural analysis this distinction is frequently elided.14
Givón's category of the
“taken for granted,” for example, seems to include both the
orthodox view and beliefs that are doxic. This confusion occurs
frequently, despite the fact that the distinction between what is
explicitly believed and what is tacitly held has a long history in
anthropology, going back to Boas 1938 and
Kluckhohn 1941, 1943.
Bourdieu's categories could be
ordered as follows, moving from views with the lowest cultural standing
(most controversial) on the left to those with the highest cultural
standing (most accepted) on the right
15 One reviewer of this article questioned how well
Bourdieu's categories lined up with mine, noting that in
Bourdieu's account, “The doxic gets disrupted by heterodoxy,
and then orthodoxy swings in in counterreaction.” Bourdieu does
discuss this kind of historical progression (e.g., 1977: 168–9). However, historical order need not
replicate cultural standing order, even if this sort of historical
progression were normally the case, which is not clear.
:
The present discussion modifies Bourdieu's framework by
recognizing four major points, instead of three, along the continuum of
cultural standing. Between Bourdieu's categories of heterodoxy
(which I label the controversial opinion) and orthodoxy (which
I label the common opinion), I add a zone of contested but not
so highly controversial opinions, which I call the debatable
opinion. This is the realm in which it is acknowledged that there
are legitimate alternative points of view, corresponding to
Givón's propositions that are “open to
challenge.” The extra category is necessary because opinions at
these different points are expressed in distinctive ways.16
Dan Segal has pointed out (personal
communication, February 2000) that there is a fifth option: views that
are so far from the cultural realm of possibilities that they are
“off the chart.” That is useful to keep in mind, but I will
not discuss them here because by their very nature they are not
discussed, and hence lack standard markers in discourse.
Another way in which this cultural standing model differs from
Bourdieu's is that in his scheme these are sharply defined
categories: Doxic beliefs are unsaid, while heterodox and orthodox ones
are said. No such hard and fast distinction will be made here. An
examination of discourse reveals a gradual continuum rather than a
sharp boundary between what is said and what is unsaid. It is possible
to bring the unsaid to awareness and express it, but in the normal
course of events that rarely happens (Strauss &
Quinn 1997:46).
Cultural standing is relative in that what is controversial in one
opinion community could be the common opinion in another. An opinion
community is any social group, of any size, in which opinions are
discussed (or, at the taken-for-granted end, in which assumptions are
shared). It can be a face-to-face group like a family, social clique,
or local community, or a far-flung one in which various media –
electronic chat rooms, newspapers, music, movies, and television
– mediate the discussion. Opinion communities are formed by prior
discussion of a topic, and familiarity with this prior discussion is
part of the expected competence of members of that community.17
It follows that for a great many topics,
children would not be part of an opinion community.
Not all opinions have a cultural standing in a given social group. To
have a cultural standing, a proposition has to have been the topic of
prior discourse or agreement. The requirement of prior discourse rules
out matters of only passing interest. Many opinions (e.g.,
That's a nice shirt. What a beautiful sunset.)
concern matters that are not topics of general discussion; hence, their
expression is governed by purely referential considerations and
politeness more than by cultural standing. Similarly, examples (2) and
(3) above fall into the category of propositions without cultural
standing, because there was no prior discourse about those topics with
which we, the speakers, had to contend. Note, however, that a matter
need not be a weighty national issue to be important to some group of
people and become a topic of discussion. The issue of whether our
daughter would join us for a family vacation (ex. 2) had not been
previously discussed, but if we had continued to discuss it because it
was important to us, then it would have acquired a standing in the
opinion community constituted by our family.
Complicating this analysis is the fact that the way S marks the
cultural standing of an opinion depends not on the established cultural
standing of that opinion in S's or H's opinion communities
but on S's assumptions about its cultural standing. As Perelman
& Olbrechts-Tyteca note, “The reference group…may be
considered in different ways. Sometimes one thinks of the …
opinion of those regarded as spokesmen for this common opinion or of
what is commonly considered to be the opinion of the spokesmen”
(1969:72).
Speakers may overgeneralize and assume that all members of a sex,
nationality, ethnicity, or generation share the opinions of the most
vocal representatives of those groups. If there is no widespread
awareness of what the state of discussion has been, speakers who
confidently express a point of view as if it were the common opinion
can lead everyone else to treat that view as the common opinion. By the
same token, in a small opinion community only one person disputing a
view that was previously unquestioned may be enough to disrupt the
assumption that it is the common opinion. However, that tactic can
backfire if the speaker does not have an assured enough position in the
group to carry this off. Perelman & Obrechts-Tyteca further point
out:
[The reference group] is highly unstable. Indeed,
if certain individuals diverge in behavior from what is regarded as
normal, their conduct may modify the norm … But if a person
deviates beyond certain limits, he will be excluded from the group, and
the reference group will be modified. (1969:72)
A further complication is that the way speakers mark cultural
standing depends on which opinion community is salient for them as they
are forming their utterance. There are several possibilities: the
opinion communities that shaped S's own opinion, the opinion
communities to which S and H both seem to belong, H's opinion
communities (if those are different from S's), the opinion of most
people, or the opinion of the most influential people in society at
large. Salience is dependent on situation and topic (e.g., shifting the
topic from gender inequality to religion may shift the salient opinion
community). Although this might seem to be a problem for the model, in
practice it is not difficult to tell, for a given utterance, which of
these opinion communities was most salient. This issue is taken up
again in the Conclusion.
Why does the cultural standing of an opinion affect the way people
talk about it? There are probably various general reasons for this.
Opinion norms are like behavioral norms. Members of a community do not
automatically follow them, but they are expected to know them, and if
they deviate, to acknowledge that somehow. Not to do so is to signal
that one does not know or care about the group's opinion. Thus, at
the controversial and debatable end of the continuum, marking cultural
standing is important for management of self-presentation. If the
cultural standing of a view is different for speaker and hearer, or the
same for speaker and hearer but different from the common opinion in
the larger society, acknowledging this fact is necessary for positive
or negative politeness. The way a speaker marks cultural standing for a
particular addressee is a prime rhetorical means of creating a
“subject position” – that is, of representing self
and other as certain kinds of people (e.g., people who take certain
stances for granted). Finally, in situations where speakers and hearers
share the same views, and share the views of their wider opinion
communities as well, further comment would be pointless. Thus, at the
taken-for-granted end, the second half of Grice's Maxim of
Quantity is in effect: “Make your contribution as informative as
is required for the current purposes of the exchange; do not make your
contribution more informative than is required.” Whether cultural
standing markers are a deliberate signal or simply embedded in habitual
phrasing, they can still serve as an index of cultural
standing, and hence be useful for cultural analysis.
DISCOURSE FOR OPINION DISPLAY
The majority of examples used below to illustrate the model of
cultural standing are drawn from genres the primary purpose of which is
opinion display (Myers 1998:103). Some
examples of these in speech are opinion statements elicited through
interviews, focus groups, debates at meetings, talk radio, political
speeches, “soapbox oration” (McIlvenny
1996), and assessments in casual conversation. Written genres
that have the primary purpose of opinion display include letters to the
editor, editorials, op-ed essays, on-line bulletin board postings,
interpretations of research in academic prose, and other argumentative
writing. Opinions are embedded as well in other genres that do not have
opinion display as their primary function: such as gossip and
narratives, where they are part of the narrative evaluation as well as
the audience's commentary; expository prose, as in textbooks;
opinion surveys, the wording of which betrays what is taken for granted
and what is disputed at the time; jokes, many of which are funny only
if the joke teller and audience share certain opinions; and even
fiction, where the way characters comment or do not comment on various
issues reflects their cultural standing in the author's and
intended audience's opinion communities. The effect of genre on
markers of cultural standing is discussed in the Conclusion.
Many of the examples below were drawn from discussions that I
elicited in semistructured interviews and focus groups or that I
encountered in various media (talk radio, on-line bulletin boards, and
newspaper editorials) in the mid-1990s on the topic of welfare reform
in the United States – that is, changes to the system of cash aid
to poor people.18
The primary data drawn
on in this article are taped semistructured interviews about the
welfare system that I conducted in 1995 with a diverse sample of 16
Rhode Island residents, as well as interviews I conducted in 1985 and
1990 with Rhode Islanders about issues of class mobility and corporate
behavior, and one of many interviews I conducted in 2000 for similar
research in North Carolina; a 1994 focus group on the topic of welfare
I moderated with nine white women ranging in age from 39 to 56; 60
on-line message board postings collected on 26 June 1995 from America
On Line's “Access Point” message board on welfare
reform; a tape recording of the Mary Ann Sorrentino radio talk show, 2
August 1993; and letters to the editor on the topic of welfare reform.
Examples taken from meetings or chance encounters were based on my
notes, rather than tape recordings. In those cases, I have asked for
the speakers' permission to use their words, if I thought they
could be identified.
Radical revisions to federal welfare
legislation were being debated at that time, and the topic was
frequently in the news. The common opinion in the country at that time,
as revealed by opinion polls and evident from casual conversation, was
that the welfare system needed to be changed because it rewarded
laziness and out-of-wedlock births, especially among racial minorities
and immigrants. Also common, but perceived as more debatable, was the
view that it is important to provide education and positive incentives
for work. More controversial opinions were that of the Marxist left
that it is good to keep people out of the workforce because that drives
up wages, and that of the far right that government aid for poor people
should be eliminated.
19 See Fairclough 2000 for analysis of welfare reform
discourse in Britain during the same period.
Additional
examples for analysis were drawn opportunistically from comments on
other topics under debate nationally or locally.
INDICATORS OF CULTURAL STANDING
Many of the linguistic and paralinguistic features discussed below
have been noticed by other discourse analysts. The purpose of the
following discussion is to draw together these scattered observations
into an overall model of cultural standing, as well as to propose some
additional markers that have not been previously discussed, to my
knowledge. Several caveats need to be kept in mind about these
indicators. First, all of these should be taken as only prima facie
indices that need to be judged in context: Differences between speakers
and genres can lead to different results. (For example, some speakers
could have a more hesitant and others a more confident style.) It could
also happen that an utterance mixes markers from different cultural
standing categories. This could happen, for example, if the cultural
standing of the view is not clear (e.g., the speaker sees it as
somewhere between debatable and the common opinion). In that case,
however, the view should be expressed with markers from adjacent
categories. Markers from nonadjacent categories should be mixed in an
utterance only if they mark different propositions, or if the speaker
changes his or her mind about the cultural standing of the view while
talking (perhaps as a result of nonverbal signals from the addressee),
in which case there should be other signs of a change of mind. Also,
each of these categories covers a continuum; for example,
“controversial opinions” covers a range from the highly
controversial to the somewhat controversial. As was stated previously,
judgments of standing are always relative to a particular opinion
community, and what counts is the judgment of the speaker about what is
controversial, debatable, and so on, not ours as outside analysts.
Finally, the following description of cultural standing markers is
based on analysis of American English, and other sorts of markers may
be used in other dialects and languages (see Conclusion).
Controversial opinions
When speakers believe their opinions to be controversial, they will
try to soften their statements in various ways to make them seem less
out of step with the common opinion. This is commonly done by (i)
self-censoring, or using euphemisms or other indirect language; (ii)
denial; (iii) hedges; (iv) attribution to others, or use of the
impersonal second or third person; (v) lamination; (vi) apologies,
requests for permission, preemptive concession, and other verbal
acknowledgments of uneasiness; (vii) self-initiated repairs; and (viii)
pauses, disfluency, hesitation, and other prosodic and paralinguistic
signs of discomfort. Teun van Dijk (1987:91–6) names several of these strategies
(use of indirect language, denial, mitigation, attribution to others,
self correction, avoidance, concession) as typical “moves”
in expression of ethnic prejudice in both the Netherlands and the
United States. There is also overlap with negative politeness and
off-record discourse strategies (indirect language, giving hints,
hedging, use of impersonal pronouns, apologies; see Brown & Levinson 1987), markers of dispreferred
seconds in conversation analysis (e.g., pauses, fillers, and apologies;
Sacks 1987), and discussion of difficult
topics in psychotherapy (indirect language, mitigation, hesitation, and
disfluency; Labov & Fanshel 1977).
Self-censoring or using euphemisms or other indirect
language
Sometimes S will simply refrain from directly stating a controversial
opinion. We can guess these censored views by seeing what S or speakers
like S say in other contexts, and by decoding the hints in vague or
allusive language (Labov & Fanshel 1977:287;
van Dijk 1987: 95, 100–1).
(5) CS had asked Daniel Collins, a factory worker, what he
thought about socialism.
DC: I'd rather not go into that…It might get into
different things.
CS: ((Long pause to see if DC would say more)) I remember you
saying last time that there's no freedom of speech; people call you a
radical.
CS: Because, actually, what you were talking about with the
government owning the utilities and so on sounds like democratic socialism in
Europe.
DC: ((Nods))20
Transcript based on
notes. Ironically, the tape had ended before this
point.
It seemed clear that Collins did not want to be tape-recorded saying
that he is a socialist.
Denial
As van Dijk points out, “One of the most stereotypical moves
used in prejudiced talk is Apparent Denial, which usually contains a
general denial of (one's own) negative opinions about ethnic
groups, followed by a negative opinion: ‘I am not a racist, but
…,’ or ‘I have nothing against foreigners, but
…’ ” (1987:91; see also Hewitt & Stokes 1975 on disclaimers, especially their
discussion of “credentialing,” and Overstreet
& Yule 2001).21
Credentialing
remarks like I am not an x but … seem to function the
same way as lamination, discussed below: If you recognize that your
behavior seems problematic, you must be more open-minded than your
comments would otherwise make you appear.
Whether or not a
speaker really believes what he denies believing is irrelevant in a
cultural standing analysis. Either way, such denials are a strong
indicator of a controversial position:
(6) Newspaper interview with chief health director in
Washington, DC, about delays in testing postal workers for inhalation
anthrax (Garvey & Rosenblatt 2001:A15).
“No one was put aside because they weren't
important or didn't work on Capitol Hill,” said
Walks.
Hedges
Hedges are notoriously hard to define (e.g., Markkanen
& Schröder 1997). George Lakoff defined hedges as
“words whose job is to make things fuzzier or less fuzzy”
(1972:195). Since then, the concept has expanded
beyond Lakoff's concern with the fuzziness of category membership in
particular. I will use “hedge” for words (e.g.,
probably), phrases (It is possible that), syntax (e.g., use
of the subjunctive modal auxiliaries could or might, or of
double negatives that are to be interpreted as weaker positives, such as
not unhealthy), and rhetorical strategies (e.g.,
understatement; van Dijk 1987:95) that are
conventionally used to modalize.22
This
makes hedges one type of modalizer. Phrases that serve to mitigate in
the particular context but do not normally have that function (e.g.,
a designer drug based on ecstasy and without the scary
side effects in [4]) are not hedges by that definition.
(See Markkanen & Schröder 1997:7 on the
question of how hedges relates to modalizers.) I can see using hedge
as a verb (hedged, hedging) in a broader way that is
synonymous with modalizing, as Markkanen & Schröder propose (1989, discussed in Markkanen & Schröder, 1997:5).
In spoken and written English, a common
function of hedges is to soften controversial statements (on mitigation, see
Labov & Fanshel
1977: 287 and van Dijk
1987:95–6, 100–1).
23 There is insufficient recognition of the function of hedges to
mark cultural standing. See, for example, Channell
1994, which lists ten functions of vague language, but omits low
cultural standing.
A single hedge word could indicate that the
speaker sees the view as debatable rather than controversial, but multiple
hedges, or hedges in conjunction with the other indicators listed here, are
markers of controversial opinions.
(7) The last sentence of a U.S. News & World
Report article on the health effects of the drug ecstasy (Boyce 2001) (see ex. 4).
Evidence suggests that perhaps some day a
designer drug based on ecstasy may be developed that could
provide potential psychological benefits—without the scary side
effects.
(8) Comment after approximately 40 minutes of focus group
conversation among middle-aged white women, in which no one mentioned
racial discrimination as a cause of poverty.
CS: How about, some people would say discrimination against
minorities?
((Unresponsive body language from the group))
African-Americans and other groups that would account for poverty?
(1.0)
What does anybody think about that? (1.9)
Judy: I think that's (.8) probably pretty
sure.
(9) Interview with Nancy Goodall. She had stated that the
media and government deliberately present a negative view of welfare
recipients, and I asked why.
((Sigh)) Well (2.9) if you want to get into a real paranoid view
you could almost say that (2.2) um (2.8) well, you've probably
read Big Brother, 1984 ((Um-hm)) and, um, if you can divert
people's focus (2.4) from problems that perhaps would be a
little more difficult to address (1.3), um, you can divert
them.
Attribution to others, or the impersonal second or
third person, or unnamed agents
Myers notices that in focus group conversation, sometimes a
“view is presented as one school of thought – a
possible view” rather than the speaker's own (1998:104). Myers proposes that presenting
a view that is not necessarily the speaker's own is a way of making
sure that a focus group fulfills its function of getting a variety of views
out for discussion. Speakers also do this when they pronounce a view that
they suspect is controversial (van Dijk 1987:98).
It is a standard rhetorical ploy to attribute one's ideas to someone
respected who can lend weight to the controversial proposal. However,
sometimes a view is attributed to someone with no particular authority
who is named as the source so that S can test the waters, putting a
view out for H's reaction without aligning herself with it (Goffman 1974:512). An example of this seems to
occur in the following conversation I held with a flight attendant (FA)
after I noticed she was reading an article in People magazine
about a high school Spanish teacher who mocked a boy in his class with
pierced ears.
(10) FA: His Spanish teacher said, “There are
only two types of men who pierce their ears. ((lowers voice)) Fags
((resuming normal volume)) and pirates. And since I don't see any
water around here…”
CS: ((Silence, no encouragement))
FA: Which I think was totally out of line.
((A few minutes later, she repeated the same line to another
flight attendant in a conspiratorial tone, and both of them
laughed.))
Alternatively, attribution can be to an unnamed agent (It could
be said that) or to the impersonal second or third person (You
might say, One could say; cf. Brown & Levinson 1987:194ff). Notice also the typical use of the
modal auxiliaries might or could in the potential
subjunctive mood in these constructions24
The unstated agent of It is said seems to be a
weighty authority that does not need to be named, whereas the unstated
agent of It could be said seems purely
hypothetical.
:
(11) One could conclude, if one was in a
very bad mood, that it is not in the interests of affluent feminists to
see the wages of working-class women improve. (Ehrenreich 2000)
(12) Interview with Nancy Goodall (see ex. 9).
((Sigh)) Well (2.9) if you want to get into a real paranoid
view you could almost say that …
Lamination
Erving Goffman points out that there can be a layering, or
“lamination,” of selves in discourse (1974:516–23; see also Hill
& Zepeda 1992, Balaban 2000). One
common way of modalizing controversial views is for S to signal that, to
use Goffman's terms, the “principal” (originator) of a
certain view is a less responsible version of one's self and, since
the speaker is aware of this, the principal can be distinguished from the
encompassing “addressing self” (“the one …
currently responsible and accessible to the listener,” 1974:520). Diminished responsibility for one's
views can be explained in any number of ways (e.g., crankiness, intoxication,
youthful error, temporary insanity). This is the principle that if you can
recognize that your behavior is crazy, then you are not really crazy (p.
521).
(13) One could conclude, if one was in a very bad
mood, that it is not in the interests of affluent feminists to see
the wages of working-class women improve. (Ehrenreich 2000)
(14) Interview with Nancy Goodall (see ex. 9).
((Sigh)) Well (2.9) if you want to get into a real paranoid
view you could almost say that …
(15) Prelude to a comment at a faculty meeting, following
presentation of an exciting campus building plan.
Maybe I'm getting old and my back hurts so
I'm going to say this in an inflammatory way but [we
shouldn't be considering a new blade of grass or new brick if we
can't make sure that we have financial aid for all the students who
want to be here].
Preemptive concessions, apologies, requests for
permission, and other verbal acknowledgments of uneasiness
Related to lamination (in which the speaker admits being crazy, for
example) is preemptive concession, in which the speaker allows that
others might think the idea to be expressed is crazy, simplistic,
wrongheaded, and so on.
(16) Interview with Joan Morse. I had asked how her ideal
system would work.
Well the obvious um (1.7) I heard somebody
say–this is very simplistic and don't throw anything at
me–but the solution to the homeless is to send them
home.
(17) Prelude to suggestion in a faculty committee
meeting.
I'm just trying this out, okay? Go ahead and shoot
me down.
(18) Prelude to comment at a faculty meeting.
This is going to be unpopular.
Sometimes speakers apologize or ritualistically ask the
listener's permission before using controversial language.
Interestingly, the verbal formulae used are the same as or similar to
the ones used in connection with taboo words. These apologies or
requests are cultural standing markers if, as in (19) and (20) below,
what is really controversial is not the words so much as the ideas
expressed by them.
(19) Discussion by Ann Fagan Ginger, Executive Director,
Meiklejohn Civil Liberties Institute, on a radio program, The
Lawyer's Guild, of a proposal to use military tribunals to
try prisoners in the war on terrorism.
We have, if you will
excuse the expression, men [in positions of power
and influence, showing poor judgment].
(20) Radio interview, Studs Terkel, quoted in an article
defending humanism (Ruddick 2001:B8).
We're facing a certain challenging moment as though it
were a test for us, a test of our intelligence as well as our, may I
use the word, humanity.
Speakers may otherwise verbally indicate their unease with expressing
a controversial view. These are all examples of the more general
practice, noticed by conversation analysts, of postponing disagreement,
such as the use of Well followed by a long pause in (9) (Sacks 1987; see also van Dijk 1987:98 on “avoidance moves”).
(21) Anonymous bioethicist, quoted in the New York
Times, 2 Dec. 2001, 4:4.25
The
article makes it clear that this view is controversial: “After a
few nervous glances, the diners went around the table, each offering a
similar confession. Few would say so in public,
however.”
I don't know how to say this, but in my heart of
hearts, I don't think cloning is inherently wrong.
Self-initiated repair
Listed above are a number of ways in which speakers indicate their
awareness that their views are controversial before they voice them.
Sometimes, instead, the apology or concession comes as a self-initiated
repair (Schegloff et al. 1977) right after a
speaker makes a comment that he belatedly realizes is controversial.
Van Dijk notes, “Corrections … are interview features that
are typical for the spontaneous nature of ‘delicate’ talk, as in
‘maybe ‘civilized’ isn't the right word’”
(1987:96). Self-initiated repairs do not always
indicate a controversial opinion; the analyst's background knowledge is
needed to judge if they do.
(22) Interview with Mathew Healey. He was discussing whether
the United States should have two official languages.
If there are more Spanish people in the country than there are,
you know, more Spanish-speaking people than Americans,
then–oh that was riddled with political incorrectness ((CS
laughs))–but uh (.9) with more English-, more Spanish-speaking
than English-speaking, then okay, it would make sense to have
a bilingual country.
Pauses, disfluency, and other prosodic and
paralinguistic hesitation markers
Similarly, when speakers are preparing to state controversial views,
very often they hesitate, stumble over their words, whisper, stretch
out their words26
Yamada (1997:90) gives this as an example of a
characteristically Japanese technique for signaling reluctant
disagreement, but it is used by Americans as well, as in example
(24).
or rush them together, and otherwise nonverbally signal
their awareness that their view is problematic (
Labov & Fanshel 1977, esp. chap. 9; Pomerantz
1984:70ff.; see
Hill
n.d. on disfluency). In (23) below, the speaker is a woman who moved
to Burlington, North Carolina, as an adult and sees herself as different
from most of the people there. She probably guessed that I would share her
dim view of “rednecks,” so her hesitation is more likely due
to the controversial nature of her opinions in her opinion community (her
friends are much more conservative than she is) than to concern about
offending me.
(23) Interview with Bess Phillips for background information
about Burlington, North Carolina.
I think uh (1.6) I don't know if you can accept
it. (.5) ((More quietly)) but I think a lot of (.7) a lot
of people (3.4) ((Rolls eyes, grimaces, whispers)) are kind of
redneck. You know?
(24) Interview with Nancy Goodall.
I'd pro::bably be happy with an overall ban of
handguns.
For another example, see Nancy Goodall's comments in (9), which
in addition to all the other markers of a controversial opinion already
noted, also include a sigh, verbal fillers, and long pauses.
Debatable opinion
Unlike a controversial opinion, a view that the speaker sees as a
debatable opinion is expressed with more openness and less
hesitation. Unlike the common opinion, however, it is still uttered
defensively and shows the signs of dispute in the prior social
discourse. Deborah Schiffrin's analysis of “rhetorical
argument,” “a discourse genre in which the speaker
establishes … a position that is potentially
disputable”(1985:40)—is particularly
apropos here. Unlike “oppositional argument,” where the opponent
is present (1985:41), in rhetorical argumentation
an opponent is imagined. Schiffrin (1985:39–40) discusses each of the following ways
in which orientation toward an imagined opponent gives this discourse its
characteristic features: (i) there is a statement of position and support;
(ii) views are qualified as the speaker's own beliefs; and (iii) there
is comparison with alternative beliefs. Sometimes there is (iv) use of
intensifiers.
Statement of position and support
Giving support for a position is a strong indicator that it has
questionable cultural standing. If the view were widely shared, it
would not need defense.27
Schiffrin
suggests that “coordinating conjunctions (and,
so, but) often introduce the position, and
subordinating conjunctions (because, like) often
introduce the support” (1985:39; see also
Schiffrin 1987).
(25) Newspaper story by a local columnist sympathetic to
protesters who were opposed to Rhode Island's support for an
expensive convention center at a time when cuts were being made in
public assistance (Kerr 1993:E1).
The
Convention Center has to be questioned and challenged.
[POSITION]
It simply sucks up too much public money in the
midst of hard times not to be. [SUPPORT]
(26) Talk radio discussion of welfare. The caller is
concerned about new regulations that will give people who are sick or
disabled only three months of temporary welfare coverage, even if the
disability coverage is not yet available.
I'm terrified of the suicides that I know is going to come
down the line [SUPPORT] with this, I believe, irresponsible
decision. [POSITION]
View qualified as the speaker's own belief
If a position seems to be debatable, speakers often do not flatly
assert it as true; instead, they frame it as their opinion using
phrases like I think, I believe, to me, to my way of thinking,
and so on.28
Schiffrin gives the
following explanation of the special place of beliefs: “Beliefs,
opinion, judgments, and feelings differ from assertions and statements
of fact because they are representations of internal, cognitive states
that are available for neither observation nor verification. Even when
presented in a conversation, then, they remain the special
informational preserve of the speaker. Note, however, a paradox:
Because beliefs are the speaker's informational property, the
speaker's right to maintain them cannot be denied, but because
they are unavailable for proof, neither can others' rights to
doubt their validity be denied” (1985:40).
I would modify this statement a little. “Assertions and statements of
fact” are also presentations of beliefs – the difference is not
whether a belief is being presented, but what speakers see as its cultural
standing. Thus, to gain the right of incorrigibility it is not enough
to utter a belief. It has to be explicitly marked as the
speaker's belief by saying I believe, I
think, and so on. No one can doubt that is what you believe, even
if they disagree with the assertion.
Note, however, that
unlike what they do when offering controversial positions, speakers are
more likely to “own” the view by using the first person and
present tense (
I believe, I feel, and
I think) rather
than impersonal second person, third person, or agentless constructions
in the potential mood (
you could say, one might say, or
it
could be said). (See also Goffman
1974:
531,
1981:285; Stubbs
1983:186–7.) Note that self-attribution versus
other-attribution does not always differentiate controversial from debatable
opinions; see (8) and (23) for examples in which
I think is used to
qualify a position that the speaker otherwise treats as controversial, or
perhaps somewhere between controversial and debatable.
(27) Comment in focus group of middle-aged white
women.
Elaine: I think our welfare system could offer more
incentives to people to work. I think that's the
problem.
(28) Comment by caller in talk-radio discussion of
welfare.
I'm terrified of the suicides that I know is going to
come down the line with this, I believe, irresponsible
decision.
(29) Comment in focus group of middle-aged white
women.
Suzanne: I'm a child of the sixties and I
really feel almost personally responsible
for bringing on the age of “drugs and sex and rock n'
roll,” because we blew away every value that was, we
left nothing standing, and now I have teenage
children and this is our fault.
(30) Comment about welfare recipients in focus group of
middle-aged white women.
Betty: Someone said something about pride
earlier on. That to me is a big beef too.
Comparison with alternative beliefs
When speakers see their views as debatable, they are aware of
competing views that are also considered legitimate in the opinion
community. Sometimes those competing views are mentioned explicitly, as
in the following example:
(31) Letter to the editor (James Haddad, Providence
Journal, 10/30/93).
In the [newspaper] story, Senator York expressed
outrage over the recent cuts in Rhode Island's General Public
Assistance program. To quote Senator York: “Government should be able
to help those least able to take care of
themselves.”
This rhetoric has been repeatedly uttered by liberals since
the days of President Johnson's Great Society program. Unfortunately,
history has shown that most government programs designed to “help
those least able to take care of themselves” have failed
miserably.
Sometimes opposing views are indicated more indirectly. One way in
which competing views are alluded to is through negation. As
Kress & Hodge point out, if someone says, “There isn't a
tiger in that room,” the speaker “(and we as hearers) must
have thought of a tiger being there first, before the proposition has been
understood in its negative form” (1979:145;
see also Perelman & Olbrechts-Tyteca 1969:155;
Labov & Fanshel 1977:104).
Alternatively, the proposed view could be put as a question,
a negative response to which would give the competing view, as in (32):
(32) Cover of a mainstream magazine, Parade, an
insert in many U.S. Sunday newspapers (Parade, 19 May
2002).
Attorney General John Ashcroft is taking firm steps to thwart
terrorism.
Understandably, his efforts provoke a critical question:
Has America's Top Cop Gone Too Far?
Another indirect way of presenting competing views is by
representing the voice of the critic (where
“voice” is used here in Bakhtin's [1981] sense to mean the discourse
characteristic of a speaker; see also Hill
1995). In (33), Mathew Healey, a defender of traditional nuclear
families, carries on an argument with the borrowed voice of a critic
(“Okay, well define what a family is”), conceding some of
the critic's points (“with all of its problems”):
(33) Interview with Mathew Healey.
And I really think family structure is important. And you can
be attacked, Okay, well define what a family is. What a family
was and had worked for however many thousands of years, with
all of its problems.
Contrastive stress (selective emphasis in speaking, font
changes in writing) can be used to highlight ideas the speaker feels
are different from what has been said already (Schiffrin 1985):
(34) On-line bulletin board discussion of welfare
(capitalization in the original).
Welfare has become a way of life. It can't be salvaged,
instead it should be ABOLISHED with a new system for TEMPORARY help to
replace it.
As Fairclough 2000 points out, even
modifiers to standard terminology can subtly index debate.
Fairclough gives the example of a British government welfare reform
proposal in which reference to “paid work” “is an
implicit acknowledgement that there are other understandings (and
discourses) of work” that would include unpaid labor (Fairclough 2000:181). Shifting ideas of what should
count as work are also indexed when mothers are asked “Do you
work outside the home?” instead of the standard phrasing,
“Do you work?”
Intensifiers
As was discussed in the first part of this article, in situations of
opinion display intensifiers (e.g., really, surely, or
indeed) mark either a debatable position (as Schiffrin 1985:40 points out) or support for a debatable
position. A word like really does two things: It reaffirms the
speaker's support for the position while acknowledging that others
would not find that position real or true. This makes it appropriate
for the in-between cultural standing of debatable opinions. Notice that
intensifiers can be combined with hedges, as in (36), which has both
the intensifier really and the hedge almost: Both are
indicators of a debatable position. (See the previous discussion of
hedges; multiple hedges are the hallmark of controversial opinions; a
single hedge may indicate a debatable opinion.)
(35) Interview with Mathew Healey (see ex. 33).
And I really think family structure is
important.
(36) Comment in focus group (see ex. 29).
… and I really feel almost personally
responsible for bringing on the age of “drugs and sex and rock
n' roll…”
Common Opinion
Common opinions are recognized as opinions to which alternatives
exist, so they are expressed, but typically (i) in short sentences with
no qualification or support. Frequently they are conveyed with (ii)
clichés, maxims, and semi-formulaic expressions (Perelman & Olbrechts-Tyteca 1969, Schiffrin 1987), and (iii) no prosodic or
paralinguistic signs of hesitation. Sometimes they are given as (iv)
the answer to a rhetorical question (Schiffrin
1987), (v) marked with you know (Schiffrin 1987), or (vi) expressed with taboo words
or other colorful, emphatic language and emphatic stress. People who
disagree with a common opinion may still invoke these cultural truths
to support their own views. In such cases, there is typically either
(vii) an explicit concession to some part of the common opinion, or
(viii) an implicit concession indicated by borrowing and resignifying
terms that represent the discourse of the common opinion.
Short sentences with no qualification or support
Unlike debatable opinions, for which speakers will offer support,
views judged to be the common opinion are usually given in short,
unqualified sentences with no support. Indeed, the common opinion is
likely to be used as support for more debatable matters of opinion:
(37) Radio talk show discussion of welfare. The caller is
defending some forms of charity, such as children's toy funds. The
host says that is different from welfare.
Host: It's different
because we give private money specifically for that purpose.
Because we're not a heartless society.
(38) Discussion in focus group of middle-aged white women
about whether their children will learn to deal with the temptations of
sex and drugs. Suzanne had expressed confidence that her kids will be
okay.
Barbara: But it's scary.=
Suzanne: =It's awful. I know what's out
there. My parents didn't have a clue.
Barbara: I wouldn't be their age again for a million
dollars.29
My parents didn't have
a clue and I wouldn't be their age again for a million
dollars, if taken to refer just to the speaker's experience,
are not the common opinion. Note, however, that they are expressed in short,
unqualified sentences, probably because the speakers assume their reactions
are shared by the other women there.
(39) Interview with George Gauvin.
Big money, boy. ((Mm)) (1.0) Can't beat
it. (1.2) ((Mm)) Can't beat big money. (3.0) ((Mm, I
know)) The rich get richer and the poor get poorer. (1.8)
It's true.
Clichés, maxims, and semi-formulaic expressions
Clichés and maxims are formulaic expressions that usually
express the common opinion in some opinion community. Perelman &
Olbrechts-Tyteca define a cliché as “an object of
agreement regularly expressed in a certain way, a repeated formula of a
stereotyped character” (1969:165), and they
say about maxims that these “condense the wisdom of the nations”
(1969:165). The difference between them is that
while a cliché is usually popular for a short time and is marked as the
outlook of the moment, maxims (aphorisms, proverbs, adages) are meant to sound
timeless and carry more weight (1969:166). Maxims
are unqualified generalizations expressed in complete sentences, often in the
simple present tense (see Stubbs 1983:120–1 on
use of the simple present tense for “eternal truths”). Sometimes a
formulaic expression is combined with expressions like they say
(“a quotative expression which itself conveys general
consensus,” Schiffrin 1987:275). Some
examples of maxims I recorded in interviews about welfare or read in
on-line discussions of the topic are the following: You make your
bed, you sleep in it; If you play, you have to pay;
It is better to teach a person how to fish than to just give
fish. Not all of the common wisdom expressed in maxims is
unsympathetic to welfare recipients. On the other side are Walk a
mile in my shoes and There but for the grace of God go I.
(Also, from a different context, The rich get richer and the poor
get poorer, [39].)30
These
examples show that there can be a variety of conflicting “common
opinions” on a topic.
The following is an example of a cliché:
(40) Interview with Mathew Healey.
Now we all have to go to graduate school ((laughs)) because
we're not competitive and then we'll finish graduate school and
find out that we're overqualified. ((both laugh)) You know,
“Would you like fries with that?”
Sometimes the common opinion is expressed in what could be called
semi-formulaic expressions, which are typical ways of wording a certain
point of view that are not frozen enough to be called clichés.
The speakers in (41) and (42) are both working-class women talking
about welfare recipients:
(41) Interview with Carol Russo.
[H]ere are these people that sit around on their
butts.
(42) Interview with Marlene Randall.
[T]hese young girls …sitting on their
butt.
No prosodic or paralinguistic signs of hesitation
In contrast to the expression of controversial views, speakers do not
hesitate if they believe their views to be the common opinion. See, for
example, the latching of Suzanne's comment on Barbara's in
(38) above. Notice also the paucity of verbal fillers and absence of
disfluency in the form of false starts, repeated words, and
self-interruptions in these examples. There are long pauses in (39);
however, unlike the controversial statements in (8), (9), (16), and
(23), in (39) the pauses come between clauses rather than within them,
suggesting a slow speaking style rather than hesitation.
Answer to a rhetorical question
It makes sense that a rhetorical question-answer pair would express
the common opinion. After all, the question would not be considered
rhetorical if the answer were in doubt (Schiffrin 1987:278; van Dijk
1987:108).
(43) Comment in focus group of middle-aged white
women.
Rosie: There could be a man and a woman who have the same
degree, same knowledge, and everything else, and guess who gets the
most money? It's the man.
“You know”
Schiffrin found, in her interviews, that you know sometimes
“marks the general consensual truths which speakers assume their
hearers share through their co-membership in the same culture, society,
or group” (1987:274).31
Schiffrin credits T. Labov 1980 with a
similar point. There is a large literature on you know. (See Tree & Schrock 2002 for a summary that is
consistent with the point developed here.) By contrast, Macaulay's
analysis of Scottish speakers finds you know use varying
idiosyncratically and seeming to function in “the rhythmic
organization of utterances, particularly when it is used at the end of
an utterance” (2002:17).
She notes
this is more likely if
you know is spoken with falling rather than
rising intonation (
1987:291). (Rising intonation
turns
you know into a marker of sympathetic circularity, a way for
speakers to make sure hearers are with them; Bernstein
1971:111.) Of course,
you know can be used in
this way only if the view it marks
has a cultural standing. There is
additional support for this interpretation of
you know in (44) and
(45) below: In (44)
you know prefaces a cliché, and in (45)
you know comes before
like they say and a semi-formulaic
expression (
it's Saturday night, they're going to go out.
And raise hell.):
(44) Interview with Mathew Healey (see ex. 40).
You know, “Would you like fries with
that?”
(45) Interview with Carol Russo. She is talking about the
irresponsible behavior of welfare recipients.
You know, like they say, Well, it's Saturday
night, they're going to go out. And raise hell.
Taboo or other colorful emphatic language and emphatic
stress
Unlike controversial statements, which are carefully worded to avoid
offense, when speakers express what they believe to be the common
opinion they may feel so confident of the view's acceptability
that they strengthen it with colorful or taboo language, such as Carol
Russo's and Marlene Randall's use of butt in (41)
and (42). Emphasis can also be indicated vocally by pitch and stress,
and in writing with an exclamation point.32
The emphatic stress used in the common opinion needs to be
distinguished from the contrastive stress of the debatable opinion.
Contrastive stress is more selective within a clause or sentence,
picking out only the particular words or phrases that are highlighted
as different from other opinions.
(46) Letter to the editor (M.J. Douglas, Providence
Journal, 1 Sept, 1993).
I was disgusted to read the
article on Juana Rivera's efforts to give her daughter an
expensive coming-of-age party … She should be sent back to
wherever she came from, and good riddance! This state and this
country will never be solvent until the leeches are removed
from the welfare system.
(47) (Quoted in Pomerantz
1984:72).33
Pomerantz does not
comment on the use of by god and god damn
here.
a:n uh by god I can' even send my kid
tuh public school b'cuz they're so god damn
lousy
It should be noted that in the United States, this use of
“plain English” indexes a particular sort of common
opinion: the common sense of the average person rather than the
circumspect rhetoric of elites.
View is conceded
It is typical for those who want to defend a position that is
debatable or controversial to start with a concession to some part of
the opposing common opinion (see van Dijk 1998:39 on “apparent concessions”).
When opponents make this concession, it is a strong indicator that they
see the propositions conceded to be widely shared.34
As Perelman & Olbrechts-Tyteca point out, “Such
expressions as ‘although,’ ‘in spite of,’ and
‘doubtless’ indicate that one is making certain concessions
but their main function, depending particularly on their position in
the phrase, is to show the degree of importance attached to what is
conceded” (1969:157).
(48) Student newspaper op-ed essay defending the existence
of God (Plaxico 1998).
Recent and exciting
developments in science are telling us more every day about the
intricacies of our minds, particularly the elaborate nature of the
biological processes and chemical reactions that get our bodies through
each day. I do not dispute these discoveries, nor do I ignore them
in light of my faith.
Borrowing and resignifying terms that represent the
discourse of a common opinion
Opponents of the common opinion may implicitly acknowledge its
widespread acceptance by borrowing and resignifying its key terms. For
example, liberals acknowledge the common opinion that welfare is bad by
borrowing the word welfare and resignifying it in the term
corporate welfare, where it still connotes unjustified
government aid, only in this case tax breaks and other financial aid to
large corporations rather than cash assistance to needy individuals.
Example (49) below shows exactly this liberal resignification of terms
from conservative discourses, while (50) shows a conservative
resignification of the liberal epithet racist:
(49) Column opposing California state bailouts of bankrupt
utilities (Skelton 2001:A3, Los Angeles
Times).
Moreover, the utilities have parent companies they've
been supporting handsomely in recent years. It's now time for these
folks to practice some self-responsibility, some family
values. Parents should help their children in need—not just
run them out to survive off government welfare.
(50) Headline of ad placed in several U.S. college
newspapers opposing reparations for slavery (David Horowitz, quoted in
Brownstein 2001:A48).
Ten Reasons Why Reparations for Slavery is a Bad
Idea—and Racist Too
These may be examples of what Fairclough (1989:171) calls “destabilizing” of
discourse conventions.
Taken for granted
Like all the other categories, taken for granted represents
a range on the continuum of cultural standing, not a single point. At
the end closer to the common opinion, propositions are expressed but
are (i) presupposed rather than asserted. With increasing unanimity and
movement toward what the speaker considers “facts” rather
than “opinions,” what is taken for granted is not expressed
at all, because it does not need to be. At this far end of the cultural
standing continuum, what is taken for granted really does go without
saying, unlike the common opinion, which we sometimes paradoxically say
with the preface, It goes without saying, but I'll say it
anyway. What is unsaid because it is taken for granted can still
be inferred by the analyst because it is (ii) implicit in what is
said.
Presupposed rather than asserted
Linguists and philosophers of language have long defined
presupposition in a way that includes cultural standing. For
instance,
We shall take the view that the notion of presupposition
required in discourse analysis is pragmatic presupposition, that is,
“defined in terms of assumptions the speaker makes about what the
hearer is likely to accept without challenge” (Givón 1979a:50). The notion of assumed “common
ground” is also involved in such a characterization of
presupposition and can be found in this definition by Stalnaker (1978:321):Presuppositions are what is taken by the
speaker to be the common ground of the participants in the
conversation.
Presupposition-triggers – words or constructions that indicate
what the speaker posits as assumed common ground – include the
following: definite descriptions (e.g., the worldwide
recession of 2001 presupposes there was a worldwide recession in
2001); factive verbs (I realized that we were
under attack presupposes that we were under attack); iteratives
(Once again we have an invasion presupposes that
there was a previous invasion); possessives (The
governor's intransigence has jeopardized the budget
process presupposes that the governor is guilty of intransigence);
and cleft constructions (It wasn't the tax cut that caused the
recession presupposes that something caused the recession)
(Karttunen n.d., in Levinson 1983:181; Simpson
1993:125).35
Although subordinate
clauses are also presupposed in the way most linguists define this term,
this construction is a less reliable marker of cultural
taken-for-granteds.
Presuppositions are rife in ordinary
conversation, although not always for propositions that have a cultural
standing. However, if what is presupposed in an utterance does have cultural
standing, that standing is likely to be high – either taken for granted
or almost so. As Labov & Fanshel note, “The more that the speakers
share as common knowledge, the more they are likely to refer to such
propositions rather than assert them” (
1977:122). As (51) shows, propositions that are taken for
granted can be embedded in utterances in which the primary assertion is framed
as controversial or debatable:
(51) Focus-group conversation among middle-aged white women.
The topic is causes of poverty.
I think the breakdown of the
family is the big factor. [PRESUPPOSED: FAMILIES
HAVE BROKEN DOWN; ALSO, EVERYONE KNOWS WHAT IS MEANT BY
“FAMILY,” AND THERE IS A SINGLE BIG FACTOR]
(52) Question from National Opinion Research Center, General
Social Survey.
Are we spending too much, too little, or about the
right amount on halting the rising crime rate?
[PRESUPPOSED: THE CRIME RATE IS RISING]36
In 1984, this question was replaced with the more neutral
wording, “Are we spending too much, too little, or about the
right amount on law enforcement?” A question about spending on
“solving the problems of the big cities” was changed to one
about spending on “assistance to big cities.” A very
interesting study of changes in cultural standing could be made by
examining the wording of questions in national opinion
surveys.
(53) Interview with Mathew Healey (see ex. 33).
What a family was and had worked for however many
thousands of years, with all of its problems. [PRESUPPOSED:
THE TRADITIONAL FAMILY HAS HAD PROBLEMS]
Unsaid but implicit in what is said
What is unsaid but implicit in what is said is different from what is
presupposed in the technical sense given above, which is said but in a
way that backgrounds it. Some ideas are left unsaid because they are so
controversial that they are censored. What is censored in one context,
however, will be spoken in another. This differentiates it from what is
unsaid because it is so taken for granted that it does not need to be
said – indeed, may not even be recognized, as anthropologists and
others have noticed for a long time.37
This does not exhaust the types of unsaids. There is also
what is unsaid because it is not part of the cultural realm of
possibilities (“off-the-chart” views, see n. 16), and the
unsaid of what is not easily voiced because it is known only in a
fragmentary, disconnected, or experiential way (Strauss & Quinn 1997:39).
Unsaid
taken-for-granted beliefs include the basic, shared schemas of
culturally constructed objects and events that have been studied by
cultural models researchers (e.g.,
Hutchins
1980,
Holland & Quinn 1987,
D'Andrade & Strauss 1992). Thus, when
Mathew Healey says
You know, ‘Would you like fries with
that?’ (40), that is sufficient to convey the unsaid,
taken-for-granted schemas he can assume we share about fast food jobs
and low-wage service-sector work more generally, and the joke,
circulating at the time, that this will be the only sort of work
college graduates will be able to find. His prior remark,
Now we
all have to go to graduate school because we're not competitive
and then we'll finish graduate school and find out that we're
overqualified, also leaves unsaid various schemas about graduate
training, unforgiving market economies, and a typical
upper-middle-class U.S. life course in which it is taken for granted
that one is competing for the sorts of jobs for which graduate training
might give one an edge. Such discussions of human behavior rely as well
on implicit person concepts (in this case, that people make career
choices based on economic rewards; see
Strauss
2000 for discussion of conflicting person concepts implicit in
the way Americans talk about class). Denials (e.g.,
No one was put
aside because they weren't important or didn't work on
Capitol Hill [6] and
we're not a heartless
society [37]) rely on the implicit assumption that what
is being denied is bad. When speakers give evidence to support a
position, they implicitly assume that this kind of evidence can be
trusted. When they discuss one topic in connection with another,
without any explanation for the connection, they take for granted the
cultural models that explain the connection. Thus, in the student op-ed
essay defending religious beliefs (48), the writer takes pains to
affirm her belief in neurobiology and physiology, though without
explaining why these are relevant. In the university opinion community
for which she is writing, the belief that the findings of science have
some bearing on whether religious claims are to be believed is just
taken for granted. (See
D'Andrade n.d. and
Quinn n.d. for detailed discussions of how to
extract implicit cultural models from interviews.)
CONCLUSION
There are various reasons why application of this model to discourse
is not as straightforward as the previous discussion makes it appear.
First, speakers may deliberately inflate or deflate the cultural
standing of their views. In one conversation I witnessed, the speaker
stated a view that was highly debatable, if not controversial, in a
completely unmodalized way (Where is Bin Laden? Why haven't we
captured him yet? If Clinton was still president, we'd
have him by now!). In that context, it was clear that he
did so to needle his addressee, whom he knew to have a low opinion of
the former president. Such deliberate failures to mark cultural
standing appropriately are not a problem for the model. Like
Grice's conversational maxims, they are intended to create
implicatures – in this case, about the playfully competitive key
of the exchange.
Another complication is that cultural standing is always relative to
the opinion community that is most salient for a speaker in a given
situation. This depends in part on the kind of communicative event that
is occurring. When offending the addressee has potentially serious
consequences, speakers are likely to use the markers that are
appropriate given what they guess to be the view's standing in the
hearer's opinion community. In those situations and in ones where
speakers have time to prepare (e.g., written utterances, rehearsed
speeches), it is rare to find slip-ups of the sort illustrated in (1),
in which S first expressed standing in his own community without regard
to what he thought was mine. When the larger ends of the conversation
are to express, strengthen, or reinforce intimacy, S's and
H's shared opinion community (i.e., the opinion community S thinks
they share) will be paramount. If the speaker does not particularly
care what the addressee thinks (as is sometimes true of participants in
social science interviews), the relevant opinion community is more
likely to be the general public. An interesting topic for future
research is what happens when two or more opinion communities are
salient for the speaker. I have found that speakers have particular
ways of indicating that a view is the common opinion in the larger
society but not in the opinion community they share with the addressee,
a special variety of the common opinion best termed the
conventional wisdom. For example, Nancy Goodall recited some
of the conventional wisdom about welfare mothers in a sing-song voice.
Her sing-song intonation framed the views as common while mocking them
as trite (Strauss n.d.). Other conventional
wisdom markers are scare quotes (both in print and in the “quote”
finger gesture; see Fairclough 1989:89) and
expressions like so-called and supposedly.38
As Dan Segal has pointed out to me, other
dimensions intersect with cultural standing – for example, whether the
view should be treated seriously or not (personal communication, March
2002).
In the majority of the examples given above, there were not great
differences in authority between speaker and addressee with respect to
the topic at hand. Would the presence of such differences (e.g.,
teachers lecturing to students, political leaders addressing the
public, parents informing their children) affect the way cultural
standing was marked? The cultural standing model presented above
predicts that if there is no potential danger to S's appearance as
knowledgeable about the cultural standing of the views espoused (e.g.,
parents do not usually worry that their preschool children will know
that their values are controversial in the wider society), then S would
be less likely to mark controversial or disputable views in the ways
indicated above. As Sperber & Wilson note, “The communicator
may be in a position of such authority over her audience that the
success of her informative intention is mutually manifest in advance.
Journalists, professors, religious or political leaders assume, alas
often on good grounds, that what they communicate automatically becomes
mutually manifest” (1986:63).
There seem to be genre differences as well, and these pose a greater
problem for the model. For example, my preliminary investigation of
letters to the editor and on-line bulletin boards suggests that, in
this context, controversial or debatable opinions are often stated
without mitigation, just like the common opinion. This may be so
because the context is by definition one for expressing debatable and
controversial views, so no further markers of that are necessary.
Still, it may be necessary to modify the scheme presented above to
adapt it to different genres and communicative events.
Finally, there are various ways in which this model might not be
universal. For example, some languages may have linguistic resources,
or discourse practices, absent in English that could be used as
cultural standing markers. Given the overlap between discourse markers
of epistemic modality and cultural standing, it may turn out, for
example, that the evidential morphemes present in some languages can
also serve as cultural standing markers. Similarly, discourse practices
in U.S. speech communities – such as use of the first person to
preface debatable opinions (I think, in my opinion) –
might be avoided in other speech communities (e.g., in Japan; Yamada
1997:90; see also Bernstein 1971:98 on greater middle-class than working-class
use of I think in his British sample, and Besnier 1992:167 on the way knowledge is always framed as
shared rather than personal on Nukulaelae Atoll). Or there may be
greater use, as Goffman comments in comparing “folk
communities” with middle-class Americans, of “adages,
sayings, little homilies, and the like” to convey “the
wisdom of the people” (1974:522). In many
societies, the ability to express one's opinion indirectly, in highly
allusive language, is greatly valued (see Hobart
1975, Keenan 1975, Strathern 1975, and Sherzer 1989
on Balinese, Malagasy, Melpa, and Cuna oratory, respectively). We might find
that a close attention to expression of controversial opinions in such
societies could lead to much greater detail under the general heading
of “self-censoring or using euphemisms or other indirect
language.”
It would not be surprising if points along the continuum were spaced
differently in different speech communities. For example, in speech
communities that avoid conflict, a view may not need to be very
controversial to be marked as controversial, while among speech
communities that have a high tolerance for conflict and debate (e.g.,
Israel; Blum-Kulka 1982, cited in Wierzbicka 2002:242), the scale could be skewed the
other way, with views that are controversial being treated as if they
were simply debatable, and debatable views as the common opinion. There
could be historical shifts, as well, in expectations about how cultural
standing is to be marked (Salager-Meyer et al.
1996).
Such cross-cultural differences in the way cultural standing is
marked do not negate the potential universal that a cultural standing
continuum will be recognized and marked in some way
everywhere. For example, Hobart explains that in the south Balinese
village council meetings he observed, an address was expected to
include nyelasang, “a restatement of common knowledge or
a repetition of universally accepted values on the subject under
debate,” before ngèdèngang pemineh pedidi,
“a brief statement of the current speaker's
opinion” (1975:75). Furthermore,
“Speeches which rely heavily on the promulgation of social ideals
may well propound contrasting or contradictory views,” (1975:75n), a strategy that recalls my observation,
“It is typical for those who want to defend a position that is
debatable or controversial to start with a concession to some part of the
opposing common opinion.” Acknowledging that one example from another
society is not sufficient to establish the point, and aware of potential
skepticism from my readers, I will close with a claim I mark as debatable:
It is my view that in every speech community there will be some way of
marking cultural standing.39
