Stance and emotion
Tyler Schnoebelen (tylers at stanford dot edu)
This week I have been wondering about the relationship between stance and emotion. Most scholars
seem to divide stance into at least ―affective‖ and ―epistemic‖ stance.1 Consider Ochs (1996), for whom
any situation has time, place, social identities, social acts and activities, and the participants‘ affective
stances and epistemic stances.2
Affective stance refers to a mood, attitude, feeling, and disposition, as well as degrees of
emotional intensity vis-à-vis some focus of concern...Epistemic stance refers to knowledge or
belief vis-à-vis some focus of concern, including degrees of certainty or knowledge, degrees of
commitment to truth of propositions, and sources of knowledge, among other epistemic qualities.
(Ochs, 1996, p. 410)
Ochs goes back and forth with how she further defines affective stance. For example, at one point she
seems to be willing to reduce it two axes: positive/negative and intensity. At other points, she labels
affective stances more specifically: ―sadness‖ or ―sympathy‖, for example.
Appraisal theory seems to have a division that is similar to Ochs‘:
Stance is similar to appraisal and can be defined as ‗the expression of personal feelings and
assessments‘ (Conrad and Biber 2000: 57). The notion of stance includes three broad categories:
epistemic stance (certainty/doubt), style stance (discourse comments), and attitudinal stance
(positive/negative attitudes/feelings)...Attitudinal stance is a broader notion than affect, making
no distinction between the systems of affect, judgement and appreciation, and is more or less
equivalent to attitude rather than affect. (Bednarek, 2008, p. 16)3
Note that in these definitions, there is no possibility for someone to convey emotion without also
conveying stance—stance subsumes emotion/affect, by definition.
Kiesling makes a bit of a different divide, though it bears some similarity to appraisal theory. His division
is between how a speaker relates to the content of an utterance and how they are relating to the listener
they are speaking to.
Other scholars add an element of ―evaluation‖ to affective and epistemic portions (e.g., Bucholtz & Hall, 2005, p.
595). I haven‘t quite figured out what this really means. Typically, ―evaluative‖ is used for things like great!,
terrible!, terrific!, etc. Eckert (p.c.) asks whether this is just basically ―opinion‖—I‘m for or against a topic, or have
some more complicated evaluation of it.
The affective and epistemic stances are in some ways central—the basis of the social acts/activities.
These are all relatively straight-forward except ―style stance (discourse comments)‖. This is more about comments
on the speaking/writing itself, so things like briefly, literally, honestly (Conrad & Biber, 2000, p. 60).
Perhaps judgment and appreciation aren‘t quite clear. The claim in appraisal theory seems to be that attitude is made
up of affect/emotion, as well as ethics-type evaluations of behavior (a brutal tyrant, a skilful performer, don’t be
cruel), and aesthetic-like evaluations of text/process/phenomenon (it’s a fantastic book). Judgments can be about
social esteem (normality, capacity, tenacity) and social sanctions (veracity, propriety). Appreciation can be about
reaction (impact), composition (balance and complexity), valuation (overall social significance). I‘m not really clear
on how useful these are—that‘d require more reading on appraisal theory than I have done.
A person's expression of their relationship to their talk (their epistemic stance—e.g., how certain they
are about their assertions), and a person's expression of their relationship to their interlocutors (their
interpersonal stance—e.g., friendly or dominating). (Kiesling, 2009, p. 172)
As he points out, these are related—someone being patronizing is also likely to be expressing certainty.
Indeed, this reminds us again of the Goffmanesque idea that certain stances require particular social roles
and particular social roles require certain stances. Ochs calls these ―clusters‖. In addition to being apt, her
description helps to block simplistic analyses that would describe linguistic forms and social meanings as
having a 1:1 mapping.
In all societies, members have tacit understanding of norms, preferences, and expectations
concerning how situational dimensions such as time, space, affective stance, epistemic stance,
social identity, social acts, and social activities cluster together. (Ochs, 1996, p. 417)
We may say that stance is itself a form of social action, where speakers are evaluating something and
thereby positioning themselves to align (or not) with the listener (e.g., Du Bois). Indeed, Kiesling makes
stance a primitive. For him, stance is the main interactional meaning being created and it’s a
precursor to any sociolinguistic variation. ―That is, sociolinguistic variants are initially associated with
interactional stances and these stances become in turn associated with a social group meaning in a
community over time and repeated use‖ (Kiesling, 2009, p. 172).4 Kiesling wants to know how far he can
go with the idea that the choice of a linguistic form is ultimately based on the interpersonal/epistemic
stance they want at a particular time.
This is a very conscious-choice kind of model. Kiesling is basically saying that all linguistic forms that
communicate stance come about from decisions based on the speaker thinking about ―who they are‖ in
relation to the person they are speaking with. Kiesling is explicit in saying that stance is always a
speaker‘s primary concern in conversation.5 Let‘s take this for a test drive with an example.
(1) I thought hhh that my lif:e was...probably not one=that=was=going=to=be=worth living
Please listen to it at: http://www.stanford.edu/~tylers/misc/GClip1.wav
―The social meaning of linguistic forms is most fundamentally a matter not of social categories such as gender,
ethnicity, age, or region but rather of subtler and more fleeting interactional moves through which speakers take
stances, create alignments, and construct personas" (M Bucholtz, 2009, p. 146)
He has a footnote saying that really conscious choice may not mean conscious calculations—his analogy is to
catching a baseball. This is a relatively minor point and his explorations and arguments militate against thinking that
he is serious about pulling out some of the consciousness. Kiesling explicitly wants to see how far he can go—in this
particular area I think he goes too far.
I am happy to go along with the notion that in each utterance, a speaker is trying to convey some meaning. But I
think scholars go wrong by focusing on that too much. In fact, I might even go to the other extreme and claim that
the majority of what is communicated is actually unintended. In terms of the intended message, the actual encoding
is far from a mapping (‗I will breathe here to indicate X‘, ‗I will drag out this word to indicate Y‘).
This clip comes from MJ Seide, who is being interviewed for Story Corps by her young granddaughter,
Genna Alperin, about falling in love with her partner, Genna‘s biological grandmother.6
Here are some of the main things going on in this utterance:
Evaluation of past thinking
Implicit citation of a past emotion
Conveyance of current emotion
Explanation to a granddaughter
In terms of epistemic stance, the use of an emphasized, past-tense thought and probably indicate the
reporting of a past feeling that turned out not to be the case (―certainty was high in the past, but now we
can see it was incorrect, the prediction didn‘t come to pass‖). It‘s a little hard to know what to do with
issues of knowledgability in this sort of utterance. In the narration of past thoughts, it‘s hard to have any
sources of knowledge other than the subject and it‘s a little hard to drop this sort of thing on to a
continuum of commitment. This is because part of MJ‘s point is that things can look bleak but then take a
direction you didn‘t expect: that is, that even the most knowledgable source of feelings (the subject) don‘t
actually know what could change. It‘s not clear to me that thinking through epistemic stance for this sort
of utterance is also that useful.7
In terms of affective stance, there‘s quite a bit of emotion in this utterance—the breath after thought, the
drawing out of life, the pause after was, the whisper-like voice quality, the actual content (―not...worth
living‖). These suggest a difficult recollection—one that actually has to be rushed through in its
articulation, given the sudden speed up at one=that=was=going=to=be.8 Two components of emotion I
have mentioned earlier are valence (positive/negative) and intensity. In this case, the valence seems to be
―negative emotion‖.9 The intensity is a little harder to pin-point. MJ speaks somewhat softly and I have
the sense that she is suppressing the intensity. That is, I would claim she is amplifying intensity by
conveying low intensity. In truth, the way I get understatement out of this is probably because the
intensity signals are mixed. There is the relative hush (low intensity) but the stresses, breathing, pausing,
and content indicate something we expect to have relatively high intensity.
The points I have made so far relate MJ to her utterance. But of course, MJ is also talking to her
granddaughter and she also knows that others (presumably friends and family, especially, but also
strangers like us) will likely be listening to her recording. In general, I am critical of researchers wanting
to make everything about intention and not really allowing things to bubble up, unbidden. Yet this is an
Since 2003, Story Corps has recorded over 50,000 interviews between people who know each other, talking about
parts of their lives. The entire archive is available through the Library of Congress, I believe. Selected excerpts are
available on NPR and on the Story Corps website (http://www.storycorps.org/listen). I‘m not aware of anyone else
using this as a resource—it‘s somewhere between natural conversation and a sociolinguistic interview.
Am I wrong? Is what I‘ve said already of use? Is there something I‘m missing?
There is also an aesthetic element to this sudden rush—the varying of the rhythm feels important to me in its, um,
Eckert (p.c.) also points out that you can tell a lot about what‘s going to happen in the utterance just from the
dropping down in ―I thought‖. Jurafsky (p.c.) adds that there is parallelism happening between the tone fall of ―I
thought‖ and that of ―my life‖, which comes right after. He also points out the shimmer/jitter in the F0 of ―one‖,
which seems to do a lot of emotion-work (such shimmer/jitter is often thought of as an indication of disfluency).
utterance that feels—while not insincere—fairly performative. I am not really sure how I would prove this
intuition—perhaps by sharing with you a larger excerpt. MJ tends to speak in a teacher-like way to her
granddaughter, her questions back are fairly leading:
(2) Do you know what I mean by that?
(3) You know, you and I talk about most things but this is the first time we‘ve ever talked about
the fact that I‘m gay. And I guess what I want to ask you is does it embarrass you to have a
gay grandma? [grand-daughter: No. [No.]]
I wonder also if this has been rehearsed before—that is, this utterance is part of a larger story of coming
to be loved and gaining a family (MJ‘s partner is the biological grandmother, she already had two
children when she and MJ met). Again, this isn‘t to question the sincerity but to point out that some
stories are important to people, some memories of past emotions are told repeatedly to loved ones.
Sharing this memory is likely to build engagement, solidarity, love.
These are types of meaning but they aren‘t the types of meaning that linguists are used to. Where are the
T‘s and F‘s? Veracity of a proposition seems to be a rather impoverished notion to me. Yet it comes alive
to me in Latour (1989)‘s definition of truth:
A statement is true that resists all attempts to bend it or break it. On this everyone agrees. The
disagreement starts when we want to account for the resistance of a statement to dispute and
dissent. (Latour, 1989, p. 102)
What if we take resistance seriously? Latour is interested in the history and philosophy of science, so his
example is the disagreement between Pasteur and Pouchet. He compares what each of them bring to the
controversy about spontaneous generation: heat and swan-necked flasks as well as the Academy, the
scholars‘ locations (Paris vs. the provinces), access to journals, and other factors. What shows up is that
each adversary matches the other‘s points up to the n+1st factor that finally tips the balance. Yet all of the
factors are part of the story, not just that ultimate one.
Something is true, then, because it resists attempts to break it—it holds together. This is a holistic view of
truth—the meaning that we pull out of an interaction, a speech turn, a word, a vowel is tied up with
anything and everything more solid than itself. It's because something holds together that we say it is true.
This may be our way out of all the ambiguity, vagueness, and indeterminacy we have run into in thinking
about emotion in language.
What can we say that is true about linguistic forms and indexical fields? How do the interpretations I‘ve
given hold together? I‘ve been curious about the limits of indexical fields and the range of meanings that
an utterance can actually have. Last week, I piloted a method to map these. I took five utterances—MJ‘s
was one of them—and ran two different Mechanical Turk surveys.
In one survey, I asked people to imagine that a speaker was their best friend and to convince me that they
were a good person, using only the one sentence as evidence and focusing on HOW things were said
rather than WHAT was said. In a second, separate survey, I asked people to imagine that they were the
speaker‘s worst enemy—it was their task to convince me how awful the speaker was, only having the one
sentence as evidence, and again focusing on the HOW (speed, accent, etc). A third survey will be run on a
separate group of people to see how plausible each of the meanings people came up with is (tbd). What
aspects are picked up on? Which conclusions are the most believable? What associations are there?
Among the ―best friends‖, sincerity and thoughtfulness were among the top evaluations. These were
somewhat questioned in the ―worst enemies‖ responses.10
Slowness =thoughtful=she thinks things through =overly dramatic=drama queen, =lazy=fat
before she says them
Pausing, =depression, =invitation to prove her =quitter=victim, =making it up
Dragging out (mentioned by not really interpreted) =attempting to be dramatic
“Pain” in her =sincere person=honest person=warm =border-line suicidal=don‘t trust with your
voice person kids=total failure, =smoked way too much,
=had conversation many times, =sad,
=seeking attention through pity,
Softness, =good listener, =likely to be listened to, =conspiring=they‘ve got something to hide
quietness in =authority
Careful speech =thoughtful mind
Enunciation (mentioned by not really interpreted)
Stutters, breath =change of mind, =a little ashamed of =mentally slow, =drama queen,
her past thoughts =complainer
“Thought” =rough, uneducated sounding accent
“Probably” =lip flapping moron
“gunna be” =stupid
Tone =remorseful=not fun to be around, =whiner
Low voice =lazy=fat
It‘s a small pilot—I gathered 15 responses to each of the five clips (limiting respondents to Americans). In
reviewing the submissions, I found that around 9-12 of the responses were ―useful‖.
I haven‘t put (4) in the table because I‘m not quite sure what to do with it. It came from someone giving a
positive evaluation of the speaker. It may be a useful comment because it shows that people are aware of
the way speech encodes multiple, sometimes-conflicting messages, or it may be nonsense—it‘s not easy
to figure out what to do with ―intonation‖ and ―breathing‖ here to attach them to frustration and hope:
(4) She sounded like a good person because although her voiced intonation sounded frustrated,
her breathing sounded like she had hope and the ability to triumph.
Together these responses show what is picked up on and the range of meanings they are given when
rooted to an utterance. This is still too free-floating and of course it misses the fact that MJ and her
granddaughter ought to have a voice here, too. But the method is meant to give us views from the outside
that might enrich ethnographic studies. Interviewing MJ and Genna about what‘s going on in their
utterances is important, but they will have blind spots, just as the outsiders do. But the outsiders give us a
chance to challenge the insiders‘ views (―When you did X, could it have meant Y?‖) We find out what‘s
happening by investigating the range of possible meanings, presenting them to participants and outsiders
alike and seeing what is resisted and how.
Some remaining questions
At one point, I had imagined tackling the following questions, but the essay took a different direction.
Nevertheless, I think they are worth later pursuit. If you have any thoughts, they are most appreciated.
How do linguistic forms convey (a) stances, (b) emotions? Is there any difference in the processes
of (a) and (b)? (If emotions are subsumed by stances, they should be the same or at least in a
Why do people convey such (a) stances, (b) emotions? What‘s the difference between (a) and (b)?
What‘s the role of choice in conveying (a) stances, (b) emotions? Is there a difference between (a)
Bednarek, M. (2008). Emotion talk across corpora. Palgrave Macmillan.
Bucholtz, M. (2009). From Stance to Style: Gender, Interaction, and Indexicality in Mexican Immigrant
Youth Slang. In A. Jaffe (Ed.), Stance: Sociolinguistic Perspectives (pp. 146-170). Oxford:
Oxford University Press.
Bucholtz, M., & Hall, K. (2005). Identity and interaction: A sociocultural linguistic approach. Discourse
Studies, 7(4-5), 585.
Conrad, S., & Biber, D. (2000). Adverbial marking of stance in speech and writing. Evaluation in text:
Authorial stance and the construction of discourse, 56–73.
Kiesling, S. F. (2009). Style as Stance: Stance as the Explanation for Patterns of Sociolinguistic
Variation. In A. Jaffe (Ed.), Stance: Sociolinguistic Perspectives (pp. 171-194). Oxford: Oxford
Latour, B. (1989). Clothing the naked truth. In H. Lawson & L. Appignanesi (Eds.), Dismantling Truth:
Reality in the Post-Modern World (pp. 101–126). New York: St. Martin's Press.
Ochs, E. (1996). Linguistic resources for socializing humanity. In J. Gumperz & S. Levinson (Eds.),
Rethinking linguistic relativity (pp. 407–437). Cambridge: Cambridge University Press.