Title: Computational Models of Discourse Analysis
1Computational Models of Discourse Analysis
- Carolyn Penstein Rosé
- Language Technologies Institute/
- Human-Computer Interaction Institute
2Warm-Up Discussion
- How does the notion of positioning in todays
chapter connect with Engagement? - How does it connect with perspective?
- Is there any connection with Personality?
Identity
Personality
Perspective
3What Jim Gee concept does this remind you of?
4Gee vs. SFL
- Heteroglossia is like a tapestry
- Gee was referring to the individual colored
threads being woven together - The subtance of the perspective
- Martin and White are referring to what holds the
threads together - More focus on alignment versus disalignment
5Connection between Heteroglossia and Attitude
But is this really different from a disclaim?
And is this really different from a proclaim?
6Can you explain why this concept is important in
the context of this class?
7Student Comment
- I'm not quite sure how one would go about doing
this but one way to computationalize emergent
identity is by creating an us class and a them
class and then populating each class by finding
what is attributed to first person pronouns
versus third person pronouns. To take this a step
further, one could first identify the us and them
in question to couple this with pronouns(I
believe researchers have been working on this, if
not, some human assistance should be ok too).
Then we would have to filter for embedded
clauses/reported speech he said that "we are
winners" identifies them as winners and
constructs us as losers. All this would be
greatly assisted by synonym-antonym lists (if we
had these, then we could predict implied
identities too).
8Processes of Language Adaptation
- Communities are defined by norms that evolve
through consensus - Linguistic norms can be modeled with
distributions - Those who conform are core members
- Centripetal participation you can view the
process of language change as a person moves
towards core participation over time by measuring
the distance between their language model and
that of the community (Nguyen Rosé, under
review) - Core members have power within the community
- Those who do not conform are part of the out
group - Within conversations, norms can be indexed
through language behavior to associate people
with a community, and therefore a more or less
powerful role within the interaction
9Processes of Language Adaptation
- Norms do not need to be taken as given
- Language innovation tends to happen from below,
i.e., from less powerful groups within
communities - Those in power dont have a reason to question
norms because theyre working for them - We learn more about language processes from
observing those defined as deviant -- thats
why theres so much focus on deviant language
in sociolinguistics
10Note Different authors within the Jim Gee style
of research will have their own theories of how
language changes and how that relates to societal
change so although they are working within a
common methodological framework with similar
terminology, their theories are not all the
sameThats why theory is a big focus in their
work.
11Important!Since language is constantly
changing, models that represent linguistic
norms also need to be relative to the time when
the language was uttered!!
Time variables Age vs. Time Period vs.
Position within a Community
12What about time?
- Anyways, I think in class when I said I believed
that gender is constructed, I had not intended
that gender is constructed (and constantly
reinforced and contradicted) through language. I
had thought of gender as a slightly more
permanent thing, probably an 'identity' that
someone has, although it can certainly change.
Gender is constructed by society. Society tells
us what makes a male and what defines a female.
Nature is not the one that tells us that men
cannot wear dresses and women can't smoke cigars.
13Student Comment
- Reading this article didn't change my opinions. I
still hold the belief that our identity (whether
it be gender or culture or anything else) is both
a reflection of the society in which we live and
the situations in which we express that identity.
According to pages 292-293 of the reading, the
author seems to be saying this same thing, except
adding that these identities are re-constructed
through discourse. I guess I'm not sure that I
understand the difference between that and the
earlier premise. - Where does the idea of Hegemony come in? How
does this relate to the idea of time?
14 Where does language change and language
differentiation come from?
15What positioning do you see in our two example
blog posts?
- Find examples of positioning from the Engagement
perspective - Find examples of positioning from the Moita-Lopes
perspective - Do the interpretations line up? Or do they
communicate something different? - Which one is more tricky to computationalize?
- Can we afford to just focus on the less difficult
one?
16Class Follow-Up
- Any interest in
- 6 unit Lab class
- An ongoing reading group?
- E.g., Digging in to more basics in discourse
analysis - Follow-up on experimentations leading to papers?
- Collaboration on SIDE extensions?
17Engagement
- Already established Positioning a proposition
- But can it also be primarily positioning between
people? - Patterns of positioning propositions as having
the same or different alignment between speaker
and hearer could do this - Is positioning in communication always
positioning by means of propositional content?
18Hedging and Occupation?
- And as such, I believe hedging is a much more
effective tool in showing generational or
occupational differences rather than gender
differences. - For example, teenagers often use verbs such as
'like' and 'all' to report speech he was all
'that's stupid' and then he was like ''but I'm
stupid too'. The occupational differences I would
attribute to the differences between people who
need exact values as opposed to people who can
accept generalizations or approximations.
19Questions?