Reaction Paper
Summary of Style Features | 181
Seven
Summary of Style Features
The foregoing analysis has shown some of the linguistic devices thatmake up conversational style. The devices that have been illustrated and discussed are, necessarily, only some of the complex network of
conversational strategies that operate in interaction. As Pittenger et al.
(1960: 242) note under the function they call reinforcement, “The wise
working assumption, then, is always that no matter how many possible
contributing factors we have itemized, there may still be others that we
have overlooked.” Furthermore, the devices that have been discussed
are not discrete phenomena but rather dimensions along which con-
versational mechanisms operate. Following is a summary of the dimen-
sions that have been examined:
1. Relative personal focus of topic
2. Paralinguistic features (absolute use and use of marked shifts)
a. loudness
b. pitch
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
182 | Conversational Style
c. pauses
d. voice quality and tone
3. Expectation that enthusiasm be overtly demonstrated, for
example, through
a. quickness of response
b. paralinguistic features
c. free offer of related material
d. use of questions (information, echo, etc.)
4. Use of questions, including
a. echo questions as back-channel
b. information questions
5. Pacing
a. cooperative versus obstructive overlap
b. timing of contribution, relative to previous contribution
c. rate of speech
d. floor-getting devices (increased amplitude, repetition of words)
6. Use of repetition, for example
a. to finish another’s statement or add to another’s line of
argument
b. to incorporate another’s offer into original statement or
argument
7. Topic cohesion (and tolerance for diffuse topics)
8. Tolerance for noise versus silence
9. Laughter (when, how much)
These and other characteristic ways of saying things are used in the
production of specific conversational devices, for example:
1. Machine-gun questions
2. Mutual revelation/personal statements
3. Use of ethnically marked or otherwise in-group–associated
expressions
4. Story rounds
5. Ironic or humorous routines
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
Summary of Style Features | 183
The tendency to use such devices based on these dimensions is
neither precisely predictable nor random. There are patterns or co-
occurrence expectations according to which certain devices tend to clus-
ter and by which signals in one channel are associated with certain
signals in other channels. For example, fast rate of speech; frequent
overlap and latching; use of frequent questions; use of high amplitude
and high and low pitch, as well as contrasts of these to yield exagger-
ated contours; tolerance for noise rather than silence—all of these were
seen to co-occur. (Although this is beyond the scope of the present
study, I would hypothesize that these linguistic signals are correlated
with such nonverbal factors as use of broad facial expressions and ges-
tures, as well as relatively close kinesic proximity and frequent touch-
ing during talk.)
Based on their use of these and other devices, the six people gath-
ered for Thanksgiving dinner had styles that were unique in some ways
and in others resembled those of other people present to a greater or
lesser degree. In some senses, Steve and I shared styles; for example, we
tended to talk a lot, and we used much overlap, latching, quick expres-
sive responses, and fast, clipped questions. Peter talked less, but he also
used overlap, free offer of opinions and thoughts, personal topics, and
quick abrupt questions in ways similar to Steve and me. We three told
the most stories, and we told nearly all the stories that occurred in
rounds. Thus, Peter, Steve, and I seemed to share stylistic strategies,
resulting in what I term a high-involvement style. In contrast, the sty-
listic strategies that characterized the speech of Chad, David, and Sally
resulted in what I term a high-considerateness style.
Yet style is not a matter of polar distinctions. Any device can be used
to varying degrees, and each person’s style is made up of a unique com-
bination of devices. Although Steve and I shared pacing strategies, his use
of humor was more frequent and more extreme than mine. He often
initiated comic routines, whereas I often built on others’ humor. Steve
told more stories, and a greater percentage of his talk was devoted to
narrative. He was also more likely to initiate stories unrelated to prior
talk. Peter shared many strategies with Steve and me, yet his sense of
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
184 | Conversational Style
humor was strikingly different. Peter tended to maintain a serious de-
meanor and deliver ironic lines in mock sober tones; Steve laughed more
and marked ironic statements with exaggerated intonation contours.
Peter used expressive paralinguistic features in narratives and plain talk.
David, whose pacing devices were very different from those of Steve, Peter,
and me, exercised a form of humor which resembled Steve’s in many ways.
Whereas Chad, at first glance, seemed to be using pacing devices similar
to David’s, it turned out that he did use fast pacing and overlap when the
topic was objective rather than personal. He never volunteered personal
information, whereas David did; he never contributed stories to rounds,
as David did; in fact, he rarely offered stories at all unless he was asked.
Chad also used humor much less than David.
Sally, in many ways, was the one whose style was most different.
Her voice was the softest; she talked the least. When David and Chad
told stories, there was evidence that they did not get to the point in
the way that Steve, Peter, and I expected, but when Sally told stories,
there was evidence that we could not tell what the point was, nor could
David or Chad. Sally’s talk showed a relatively high percentage of
humor, but the humor was often of a different sort (content rather
than style irony).
In other words, each person used a unique mix of conversational
devices that constituted individual style. When people’s devices matched,
communication between or among them was smooth. When they
differed, communication showed signs of disruption or outright
misunderstanding.
Because the present analysis is based on the talk of only six people,
it is impossible to draw conclusions about the cultural determination
of their styles. Nonetheless it is equally impossible to ignore the fact
that those whose styles seemed most similar—especially in the gross
outlines, such as turn-taking conventions, use of expressive para-
linguistic features, and so on—were of similar ethnic and geographic
background. Steve, Peter, and I all grew up in middle-class Jewish-
identified families and social networks in New York City. (The fact
that Steve and I met in summer camp when we were teenagers is evi-
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
Summary of Style Features | 185
dence that our families had similar orientations.) Chad and David,
on the other hand, grew up in sections of southern California which
were not ethnic identified. Chad, however, was less disconcerted by
the fast pace of the evening’s talk, and he was better able to partici-
pate. One cannot help but wonder whether the fact that his mother is
an Italian from New York City plays a part in this difference.
The one who was least able to participate in terms of rhythm and
the establishment of thematic cohesion was Sally, who was born and
raised in England. Moreover, anyone who has experience with people
from the backgrounds represented in this group immediately identi-
fies the devices used by the New Yorkers in this group as somehow remi-
niscent of the styles they have observed in people from that background,
and of Sally’s style as somehow typical of upper-class British speech.
It is certainly not the case that everyone from these backgrounds talks
just like this. People differ in individual ways. Nonetheless, the use of
such conversational devices and the expectation that others will use
them are certainly learned the way language is learned, that is, in inter-
action with family and friends. Although there is no inherent disposi-
tion toward particular stylistic devices associated with ethnicity or class,
nonetheless ethnic and other subcultural identities often involve one
in social networks in which particular linguistic strategies are exercised
and thereby learned (Gumperz [1982b]). It would be surprising indeed
if people who habitually interacted with each other did not develop ways
of talking that became generalized among them.
Conversational style, then, may be seen as a mix of linguistic de-
vices made up of features and used according to strategies for serving
the human needs for interpersonal involvement and independence.1
In one sense, each device is represented by a kind of continuum.
Speakers may be distributed on one continuum with respect to how
fast they pace their comments relative to previous comments, another
1. Kochman (1981) identifies similar opposing values as the rights of feelings versus the rights of sensibilities, corresponding to my involvement-considerateness (or indepen- dence) principles. He finds that differing values placed on these rights account for black- white style differences. The present study suggests that white style is not monolithic.
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
186 | Conversational Style
A continuum representing pacing practices might look something like
this:
Subtle ________________________________________ Gross
humor humor
C h
ad
Sa ll
y
P et
er
D eb
o ra
h
D av
id
St ev
e
Slow ___________________________________ Fast
pace pace
Sa ll
y
D av
id
C h
ad
P et
er
D eb
o ra
h
St ev
e Percentage of talk devoted to stories yields this:
But numbers of narratives told yields yet another:
Low High
percentage ________________________________ percentage
of narrative of narrative
turns turns
C h
ad
D eb
o ra
h
P et
er
Sa ll
y
D av
id
St ev
e
If these continua are superimposed one upon the other, an overall
continuum representing relative signaling of involvement and con-
sideratenes might yield something like this:
Few Many
narratives narratives _____________________________________
C h
ad D
av id
S al
ly
P et
er
D eb
o ra
h
St ev
e
for how gross their paralinguistic features are, another for how many
stories they tell, and so on. In this sense, speakers in the Thanksgiv-
ing group occupy different places on different continua. A continuum
representing grossness of styles of humor might look like this:
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
Summary of Style Features | 187
An interesting insight is suggested by the different ways in which the
people in this group of speakers perceived each other. In recalling the
occasion several months after it had occurred, Sally referred to the group
as a “rambunctious crowd,” and she identified the participants of that
crowd as Steve, Peter, David, and me. David, in contrast, thought the
evening was dominated by Steve, Peter, and me. Peter thought the
evening was dominated by Steve, and Steve thought it was dominated
by me. I had the impression that Steve, Peter, and I had participated
equally, to the exclusion of the other three.
Reference to the considerateness-involvement continuum may ac-
count for this discrepancy.2 Distinctions among speakers become more
refined, the closer one is to those speakers in style. Thus, to Sally, David’s
style was more like the styles of Peter, Steve, and me than like her own,
so she perceived him in association with us. To David, it was clear that
Peter, Steve, and I differed from himself, so he did not make distinc-
tions among our styles. Peter, however, is very familiar with fast-talking
style, and he thus perceived differences between Steve and me. Steve,
on the other hand, might be disinclined to see himself as dominating,
so he naturally perceived the one with the next grossest style in that role.
It seems likely, considering my findings, that some aspects of style
are particularly salient: pacing, grossness of humor, storytelling. It
seems likely that Steve and I perceived David as a more active partici-
pant than he was because he told a number of long stories and because
he joked a lot, with a paralinguisically gross style. Similarly, we perceived
Peter as a more active participant than he was because he kept up with
the fast pace and told many stories.
2. Robin Lakoff alerted me to this perspective, by suggesting the continuum and not- ing the operating principle, “Anyone to the right of me [on the continuum] is rambunc- tious.”
Considerateness ——————————————— Involvement
Sa ll
y
D av
id
C h
ad
P et
er
D eb
o ra
h
St ev
e
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
188 | Conversational Style
Conversational style, then, is made up of the use of specific linguis-
tic devices, chosen by reference to broad operating principles or con-
versational strategies. The use of these devices is habitual and may be
more or less automatic. The goal of all conversation is to make clear to
others the speaker’s intentions. The degree to which one’s meaning is
understood as intended depends upon the degree to which conversa-
tional strategies, and hence use of linguistic devices, are similar. Fur-
thermore, the similarity of such devices makes for rhythmically smooth
interaction. Both the rhythmic synchrony and the construction of
shared meaning create the satisfying sense of harmony that often ac-
companies conversation among people who share social, ethnic, geo-
graphic, or class background. By the same token, the use of strategies
and consequent devices that are not understood or expected creates a
sense of dissonance, which often leads to negative or mistaken judg-
ments of intent. This, in turn, often leads one to walk away from an
encounter feeling dissatisfied or disgruntled. Thus an understanding
of conversational style explains in part what often appears as clannish-
ness among members of certain groups, or discrimination or prejudice
on the part of others.
The present study has focused on a single extended interaction,
during which a particular style in some way “dominated.” Although all
conversational devices can be successful when used by speakers who
share expectations of signaling systems, it is in the nature of interac-
tion that when devices are not the same, one style takes over. For ex-
ample, those who expect shorter pauses between utterances will
necessarily speak first; having spoken, they effectively block the con-
tributions of others, but just those others who cannot tolerate much
overlap. The voices of those who talk more loudly than their interlocu-
tors in a given interaction will necessarily ring out, and the voices of
those who talk more softly will, thereby, be drowned out or overshad-
owed. These effects are independent of the intentions of the speakers
involved. Thus, the fastest-paced speakers in any group are the ones who
will usually end up talking the most.
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
Summary of Style Features | 189
This study has been a first attempt to describe the features and de-
vices that constitute conversational style. Much work remains to be
done to continue and further develop this description, to understand
better the strategies and universal principles underlying styles, to docu-
ment the devices that make up different styles, to investigate styles in
different contexts and in interaction with each other, and to correlate
linguistic with nonverbal channels of communication and nonlinguistic
aspects of behavior.
The study of conversational style, moreover, is no more nor less than
the study of communication. The features that make up conversational
style are the conventions by which meaning is communicated in social
interaction. Differences in conversational style contribute to the phe-
nomenon that Becker (1995) discusses, paraphrasing Ortega y Gasset
to the effect that all utterances are both “deficient” and “exuberant.”
They are deficient in that they fail to communicate precisely and entirely
what a speaker intends, and they are exuberant in that they communi-
cate more than a speaker intends, including, as has been demonstrated
in the Thanksgiving interaction, images of oneself.
Understanding conversational style, then, is a matter of identifying
the system that links aspects of discourse realization and organization
to each other and also links this linguistic system to other aspects of
human behavior and consciousness. In this sense, the study of conver-
sational style is the study of discourse coherence.
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .
This page intentionally left blank
Tannen, D., & Tannen, D. (2005). Conversational style : Analyzing talk among friends. Retrieved from http://ebookcentral.proquest.com Created from apus on 2020-04-04 00:15:41.
C o p yr
ig h t ©
2 0 0 5 . O
xf o rd
U n iv
e rs
ity P
re ss
, In
co rp
o ra
te d . A
ll ri g
h ts
r e se
rv e d .