This is the fifth in a series of notes to fifteen lectures for my class ITSF5001: Ethnography and Participant Observation.
  1. The fundamental issue remains:
    1. Note that I remain agnostic here as to whether this 'knowledge' is positive (platonic?) or constructed for a time and place. are we getting, as researchers, what there is to get? This assumes that there is indeed something to get, a new form of knowledge for oneself and one's audiences.
      1. This is a methodological issue given what we have learned about the difficulties of getting such knowledge.
      2. It is an epistemological postulate that such knowledge is available for those who seek it.
      3. It is an ethical choice that such knowledge is worth trying to obtain.
    2. As Kaplan noted, in this process, we, as researchers, are all limited, and enabled, by the same processes that limit all human activities, and these are above all symbolic. In the context of ethnography, these limits have to with language as the means through which all knowledge is mediated. More specifically, they have to do with the limits and possibilities of conversation: language in use with specific others.
      1. Technically, I am not only talking about "language" as common sensically known but about the ensemble of semiotics means (including gesture, dress, body decorations, architecture, furniture, etc.) that are available for the expansion and narrowing of propositional meaning and thereby makes uncertain what the propositional meaning could be, as well as produces new possibilities.
    3. There are two aspects to this critique of methodology:
      1. What is there to know?
        1. Work in sociolinguistics and ethnomethodology, building on work in sociology and anthropology, have greatly expanded our sense of what is significant in the shaping of human action
      2. How are we to know it?
        1. If
          1. much that there is to know is not directly accessible to semantic elaboration,
          2. and we cannot be sure that anybody's semantic elaboration covers the same areas of what there is to know,
          3. and if we suspect that the very act of asking limits what we can know (even as it makes up new realities to handle):
        2. then methodological choices are fundamentally theoretical
    4. To simplify, there many ways for getting at what we want to know about human beings at least:
      1. asking (and trusting in our questions, whether shaped as questionaires, structured interviews, focus and unfocused group, chance encounters);
      2. listening and watching (and trusting the answers, whether given as words or movement);
      3. eliciting or, more controversially, provoking
        1. this include any form of experimentation when we have to trust the dialogical encounter between the researcher (whether ther researcher is visible or not) and the participants;

      In each case the researcher acts practically with the people. He is necessarily a full participant. But, in each case the researcher takes a different stance vis-a-vis the other participants. Together they make different "polities (communities) of practice" that will highlight different aspects of the settings (allow for different kinds of learning).

      We will discuss asking in this session, and then listening and eliciting.

    2. on actively asking

  2. Let us first pay attention to Briggs
    1. The history of asking in the social sciences and the developing awareness of the issues.
    2. The referential illusion: What is the relationship between knowledge sought and the organization of the interactions through which it is sought.
      1. The continual demonstration (one of the major contribution of anthropological research) that
        1. it is dangerous to ask questions of the type: "I want to know X about people A"
        2. and even more dangerous to plan to answer such a question by proceeding by assuming that one can know X among the A's by asking one of the A's
        3. most dangerous of all is the assumption that what the A's tell about X is what there is to know about X (the validity issue again)
      2. There is never any reason to take systematically the naive questions one may start with and the leave behind:

        "How do you do know if the informants is telling the truth?" (Dean & White p. 105)

        (note the quote marks in the title: Dean & White know that this question cannot be taken literally)
        The authors from the book by McCall & Simmons stand at a time (the late 60s) after several generations of ethnography research in sociology. They are summarizing the ethnographers' experience.

        1. (note that they are writing about doing research in institutions in the United States and thus that their discussions are particularly relevant to people doing research in educational settings)
        Given their position in a research world getting more and more hostile to naturalistic research in the behavioral sciences, they present their summary in terms used by those outside.
        1. [note that this continually sends them towards epistemological asides on matters like "truth" or "real belief" (p.105)
        2. note further that they then fall back into the vocabulary: "perception ... filtered by ... the informant's cognitive ... reaction" (p. 105)]
        The issue here is not "truth" but
        1. what the observer is to make of what had been told.
        2. what the observer it to do next.
      3. Anthropological disinterest in maintaining reliability and establishing validity formally and a priori
        1. much of what anthropologists observed is done at such a scale that the question of reliability does not arise (e.g. major rituals, laws, myths, etc.)
        2. anthropological work is best typified by continual challenges on validity (e.g. how are we to see "teaching" or "learning"?). We are continually fighting with each other about the relationship of what we see to some aspect of the theory the previous observations was supposed to be relevant to.
      4. The continued push to find new forms of interaction ("participant-observation" recast) that will reveal features that would otherwise remain hidden.
        1. This is the major point of Briggs's book: the more common forms of interview as social interaction may prevent us from seeing what we would see if we stood with our informants in different positions: we must multiply the biases rather than try to eliminate them.
    3. An epistemological, philosophical issue: who knows best (or most usefully) among the questioner's community and among the questioned's community
      1. obviously the natives always "know best." But what is the form of their knowledge and how is it related to the form that our knowledge must take?

        Obviously I write "natives" to refer to ongoing participants in pre-existing nodes of interactions probably controlled by participants in other nodes of the network. Garfinkel talks about such participatory grousp as "cohorts," "congregations," etc. to escape the history of works like "group," "community," etc.

      2. Given that the form of the native's knowledge may not be discursive (that is available to referential questioning), then what other techniques must be sought that will translate their formed knowledge into our formed knowledge?
      3. Given that it is likely that the set of participants we study (e.g. in a family or classroom) are internally differentiated in terms of their rights, priviledges, and interests, as related to the setting, then who are we to trust?
    4. Confronting the limits of participation:

      First, recognize one's position as participant (-observer) (caveat: questions of scale):

      1. no way to prefigure where one will stand (how one will be interpreted by the observed)
      2. no replication
      3. no way to know whether what one has seen is all there is to see (question of boundary)
      4. no "sampling" possible.
      5. no way the observed can construct themselves as "observed" and fit themselves within the observer's categories. (the observed may reveal the categories to which they hold each other accountable, but these are not those which will be useful for the observer even (particularly if) the goal is "the discovery of the participants' own categories" (after all participants do not need to discover their own categories!)
    5. Criteria:

      These issues are still with us:

      1. Classically:
        1. reliability: (LeCompte and Preissle p. 332-341)
        2. validity: (LeCompte and Preissle p. 341-354)
      2. But rather
        1. The move towards precision in the account of the happening (given the likelihood that all that is knowable about humanity is the "how" of action: constraints and possibilities): Byers
    6. Still there may reason to 'merely' ask using any of many possible techniques from
      1. close questioning on a narrow set of issues ("questionaires," including censuses, demographics, genealogy, etc.)
      2. focused interviews
      3. unfocused interviews (for example, from instructions on what to organize a group discussion with mothers of children with autism)
      4. conversations (one on one, or one on many)
      5. focus groups, formal and informal
      6. quasi-experimental including
        1. reactions to texts or artifacts produced by the researcher (projective ests, Tobin on viewing videotapes
        2. breaches in the expectable and their handling (Garfinkel)