What is always confusing about focusing on talk as action is that so often discourse is positioned in lay conversation or in substantive fields as necessarily not focusing on social action. For example, we might think about learner’s discursive patterns as not focusing on their actual practices (in the behaviorist sense, perhaps). We’ve talked back-and-forth for a while now about whether or not DA (and specifically DP) is a new linguistic behaviorism, which it absolutely is not, but the point stands: for many readers, the idea that talk is action/-oriented is a bit of a hard sell.
I think that’s partially because when we as conversation analysts talk about what talk is doing, we’re talking about things like making offers, requests, soliciting information, accounting for absences, assessing self-perceptions, etc. This is maybe different from other kinds of actions, like building a fire or cooking a steak or driving a car or lifting a box, all physical kinds of (observable behaviors?) actions. And we’re slightly deviating from classical speech act theory, although not entirely; we are focusing on speech as performing action IN the talk rather than action OUTSIDE the talk (like marrying two people).
I belabor this because it’s important to think through this action orientation as “talk does action,” but it isn’t quite the action someone focused on the observable and physical would be thinking about, and also not world action in the speech-act-theory sense of the word.
I think there is something I remain unsure about which is how we can talk about agency and intentionality in CA. On the one hand, we don’t want to assume people have no goals, intents, or desires when they are speaking, and it feels silly to pretend that people are not agentic, etc. On the other hand, we don’t want to assume people’s intents, and want to focus on what talk actually does, how it is taken up, etc., without necessarily assuming that this was the person’s intent. How can we analyze the talk without disregarding the speaker? I guess the traditional answer is maybe we interview them after the conversation and see what they meant? Or perhaps, from within the CA community, the idea that their goals are automatically apparent in the talk, because in the event the goal is not met, further instances of attempting to accomplish it will be present, or there will be some trouble in the talk, or some such.
Assorted other thoughts:
What talk is accountable? Ok, but then surely not all talk is accountable talk. Did Sacks or others talk about this? To some degree all talk needs to account for itself and/or have its relevance be self-evident (auto-accounting?). It’s only explicitly accountability talk when the account is vocalized, perhaps (like when you don’t know the answer to a question and are in this K- position).
A DP Conundrum? This is so simple and I’m amazed it’s taken me three years to get here, but this is a key question. Aren’t we in a bit of a bind if we say we should not take talk to be revealing of person’s internal states & we should understand how people ORIENT to talk. But then, often people orient to talk AS revealing of neutral mental states. I think this is a central tension in DP that I haven’t really thought through yet. I think it’s fair to say that people don’t ALWAYS orient to talk as “true” or revealing, but they do pretty often. How can the analyst reconcile all these things? Actually there doesn’t seem to be an inherent paradox (no actual contradictions here), but it’s a bizarre self-referential puzzles.
Timing Pauses. How does anybody time pauses ever? I finally took Wiggins’s advice and set a stopwatch to see if I was saying “one missisippi” at the right speed – turns out I have not been. I have been much slower. In some of my Jeffersonian transcription, I used a stopwatch as best I could to calculate pauses. What I used to call like (.7) is, it turns out, closer to (1.2). It seems as though my pauses will never be accurate, absent a sophisticated software or resorting to audacity. So far these pauses haven’t been too analytically relevant, but I have no doubt that there are more pauses in the talk than I am hearing or timing. This has actually rocked my world a bit – what I would call a micropause might be .2 or .3 according to a stopwatch, which means that what I would call no pause could very easily be a micropause! Everything about this has horrified me.
Validity & quality in data sessions. Here’s something. A key factor in the validity of a DP study is the use of data sessions to validate findings, suggest that the interpreter’s view is broadly agreed to and well-conceptualized. But no one ever talks about (in publications) what actually happened in the session (“I held a data session and Francesca said everything was great so I know this analysis is valid”). How should we think about quality within data sessions, and if we are gonna talk about the quality this gives to our analysis, how should we report that?
I bring this up because I have a data session schedule for Tuesday. I could use some feedback on this. I’ve had data sessions around this data multiple times (once in DP, a different segment in CADA group, different segments at LANSI if you call that a data session, etc.).
- Is it better to use those same data segments with this different group to see if the interpretation holds up across audiences, or
- Is it better to use new data segments to see if the interpretation holds up across segments?
Moreover, how do I negotiate the fact that I’m now in the findings portion of the analysis. I am past “unmotivated looking,” and in fact have already drawn some conclusions based on the data. Now I am trying to concretize and verify those conclusions across the data set. It seems like this makes for a very different data session. How much of this information about my own interpretation should I reveal before the session? I usually reveal as little as possible, but won’t I get more out of it if I say my own interpretation, or might this compromise the validity of the data session?