ED 737-013: Discourse and Multimedia Analysis
Notes on Readings and Media
Please also see Terminology & People
Wortham (2003). Linguistic Anthropology of Education.
This introductory reading is meant to help provide an overview of the use and relevance of discourse analysis in educational research and its relations to ethnographic and similar studies.
Wortham notes the history of a focus on language within the educational ethnography tradition, particularly the influence of Dell Hymes (who became in the 1980s Dean of the Graduate School of Education at Penn, late in his long career as an outstanding scholar in linguistics and cultural anthropology) and John Gumperz (at Berkeley, also in linguistics and anthropology) on such researchers as:
Shirley Brice Heath (Penn, Stanford)
Susan Urmston Phillips (Arizona)
Courtney Cazden (Harvard)
Sarah Michaels (Harvard, Clark)
Kit Woolard (Penn, UC San Diego)
Bambi Schiefflin (Penn, NYU)
Judith Irvine (Penn, Michigan)
He also draws on some related traditions in linguistic analysis, particularly the work of Michael Silverstein (Chicago), who was Wortham's teacher, and Roman Jakobson (Harvard), who was Silverstein's teacher. Silverstein is best known for developing the notion of "indexicality" in discourse: the way in which something you say has a meaning based on (a) the fact that you said it, and (b) what it says about you as the speaker. (It's more than this, and we'll discuss it later in the course). Jakobson is one of the founders of modern linguistics, coming from an Eastern European tradition, and worked mainly on the regularity of sound patterns in language, but also on "shifters" or signs that change meaning depending on who is using them and in what context.
The general theory of signs (semiotics or semiology) was developed by Charles Saunders Peirce in the US and Ferdinand de Saussure in Europe at the end of 19th and beginning of the 20th century. Peirce was the first to emphasize the importance of indexicality as one way that signs function and carry meaning (iconically, indexically, symbolically).
Wortham raises a number of issues that we will deal with many times in the course, particularly the ways in which expectations about meanings, often deriving from our participation in larger groups or institutions in which various forms of discourse have their broader functions, play a role in how we make sense of something heard or read and in how we try to make sense to others. An important idea in his approach is that our use of words or other signs helps to create a situation or context, and does not simply adapt to situations that already exist. Another is that as the situation shifts, so do the expectations about how we should act, what we could say, what it might mean to others, what their words and actions mean to us, etc. This is a dynamic view of discourse-in-context: each thing that happens changes the potential meanings of what might happen next. The trick in doing discourse analysis this way is to figure out what is producing the most important changes of interest to us, and how, so that we can track the changes and identify evidence for how they happened.
Wortham's example turns on the role of pronouns and changes in which pronouns are used to refer to whom. These sorts of changes are changes in the interpersonal aspect of the discourse (which I will refer to as part of the Orientational meaning function of language or other media): how we create, sustain and modify our relationships with others (and even influence how they relate to each other). You can see that these changing interpersonal relationships in the classroom group also depend on, and influence how we interpret, ideational meanings (representing a state of affairs in the world, part of what I will call the Presentational meaning function): "mangled bodies at the bottom of the staircase" is a statement about the world, but the choice of the word "mangled" and the image that the whole phrase evokes has an attitudinal meaning (i.e. this state of affairs is also very undesirable) which also contributes indirectly to the interpersonal relations. Even if the teacher means this in a teasing or partly humorous way, that also contributes to the interpersonal meaning. Of course the use of the pronouns also contributes to the ideational meaning by helping us to sort out who is who in this state of affairs. it is important to see that the same words, the same phrase, can function both ideationally and interpersonally (and also "textually" as part of the Organizational meaning function of language). Not all of these functions will be equally important to us, but we should be able to identify all of them when we need to.
Lemke (1990). From Talking Science: Language, Learning, and Values.
This book is the outgrowth of a major research project on the role of language in science (and some other) classrooms in several middle and secondary schools. It looks at both the interpersonal and the ideational contributions of discourse: how language mediates teacher-student relationships and classroom organization and how it mediates the communication of scientific knowledge. One of the major points early on in the book is that these two functions sometimes work at cross-purposes: teachers and students have to decide which is more important at a certain moment, relationships or learning.
The book takes a few representative examples of classroom discourse and analyzes them in multiple ways, highlighting one or another aspect of what is going on in and through the talk. In some places it also discusses briefly the role of other media: written language in textbooks, diagrams on the chalkboard, gestures. The major contribution of the book was to show educators the importance and usefulness of studying talk in classrooms, especially for understanding how conceptual relationships are presented in talk between teachers and students (see notes below on Chapters 2 and 4).
The first chapter gives a simple analysis of the basic organization of talk in classrooms that is repeated again and again with small variations. This structure, which I called "triadic dialogue" and is today known most often in the research literature as I-R-F or I-R-E dialogue, has been noticed again and again by many researchers and was the first well-documented speech genre in educational discourse analysis (see work in the Bibliography by Sinclair & Coulthard, Mehan, and Cazden). There is a lot of debate about how it is used and whether its functions are mainly good or bad (see the classic article by Gordon Wells). It is not the only pattern of teaching-class dialogue, but it is very, very common. As an institutional discourse genre it provides a set of expectations about meanings of individual utterances, and a set of resources for creatively improvising around these expectations. An analysis can focus on the typical patterns or on the improvisations and exceptions, but each makes more sense if we also look at the other. We can also consider the functions of triadic dialogue: how does it contribute to personal and role relationships? to classroom organization and order? to authority and power? How does it contribute to the communication of scientific or other concepts and information? How does it function to build interpretive expectations for particular utterances? How does it function within the classroom as an institution on a longer timescale? What does its prevalence say about the functions of classrooms in schools and in society?
A large part of Talking Science is about the communication of scientific concepts and relationships. The analysis here is done through a second sort of institutional pattern that occurs not just in schools but also in other places where scientific discourse is used or taught. This is not a pattern of actions and moves (questions, answers, evaluations, etc.) in an exchange structure like triadic dialogue, but a different kind of pattern, a pattern of semantic relationships. I called this a thematic pattern (or later a "thematic formation"), and it is similar to the notion in science education of a concept network, except that it is about language meanings, not about "mental concepts" (if there are such things). Across the different statements of an episode of triadic dialogue, or even across lessons, and in principle between lessons and textbook chapters and between schools and science laboratories and conferences, there are certain typical and repeated patterns of meanings that occur again and again. They are in one sense the basic building blocks of scientific discourse about a particular topic. The example in chapter 1 is about electrons in atoms; the basic patterns is built out of sets of terms closely related in meaning (e.g. different kinds of atoms, different kinds of orbitals), and the semantic relationships typically constructed among these items (e.g. a certain number of electrons is always said to be located in a particular kind of orbital in a particular kind of atom). The semantic relationships are basic ones in our language: "number of", "type of", "location in" (quantification, classification, location). When used in a particular thematic pattern, these generic meaning relations take on more specific meanings: what "located in" means when applied to electrons "in" an orbital, or orbitals "in" an atom is different for this thematic pattern in the language of physics or chemistry than it might be in language about everyday objects. Learning to" talk science" is in part about learning to express these patterns correctly in statements and questions in various situations. It is about learning to use the thematic pattern in your language, and learning how the meaning relationship in the pattern are both similar to and different from the generic semantic relationships that we find everywhere in our language.
In terms of discourse analysis, this chapter illustrates the need for two kinds of analysis: exchange structure analysis (e.g. identifying triadic dialogue) and thematic-semantic pattern analysis (identifying the electron/orbital/atom pattern). The first of these applies mainly to the interpersonal meanings, the second mainly to the ideational meanings. The first depends on the general Orientational function of language, the second on its Presentational function. Both of these turn out to be especially simple and easy to do in the case of science classroom discourse. This is good practice for more complicated analyses, where the patterns are harder to identify.
Ochs, Elinor. (1979). Transcription as theory.
This is a classic paper which discusses the sense in which transcription conventions, for both verbal and nonverval data, represent choices which have important theoretical implications. Because transcription is always selective and can never display all aspects of the information from a sound or video recording, we need to consider what should be represented and how. As with all aspects of research methodology, the answers differ in each research project, depending on our research questions and the theoretical grounding of our approach to discourse and communication.
Some questions to think about as you read:
How is Och's approach influenced by her work transcribing children's talk and behavior?
What are the implications of keeping utterances by different speakers in different vertical columns in the transcript?
What should we take as the primary data of discourse analysis: recordings or transcripts? why?
In what sense is a video- or audio-tape already a kind of transcription of an event?
Are laughter, intonation, and pauses a part of language? of speech? of communication? of interaction?
What are some of the major difficulties in transcribing the kind of data that you are interested in?
Going further:
Ochs cites another classic paper by Sacks, Schegloff, and Jefferson (1974, see course readings and bibliography). This is the founding paper of the approach to discourse analysis called Conversation Analysis (CA). Gail Jefferson developed the transcription system used in this field. You might want to compare it with Och's recommendations and also to consider the ways in which it is specifically adapted to transcribing informal conversations (e.g. dinner table talk, telephone conversations) as opposed to more formally organized kinds of talk (lectures, news broadcasts, classroom lessons, courtroom discourse, etc.).
Later in this course we will look at an example of multimedia transcription (Thibault 2000) for a film/video commercial advertisement, and a visual-verbal analysis of videotapes of children playing and archeologists analyzing a site (Goodwin 2000). You might want to look ahead at these to get a sense of various ways in which nonverbal information can be included in a transcript.
Goodwin – MxM analysis – notes
Chuck Goodwin’s (UCLA) paper “Action and embodiment within situated human interaction” (1999) is an argument for multi-modal or multi-semiotic analysis.
He is mainly trying to show the ways in which language, gesture, posture, and other ways of making meaning (e.g. use of artifacts or tools) COMBINE or are strategically integrated by people who are talking and acting together. The important claim here is that to understand the meaning of what is going on, you need to pay attention to all the different semiotic fields (language, gesture, posture, action with artifacts, etc.) together, and especially to how they contextualize each other in regard to meaning.
At the same time, the multimedia version of the paper shows one way to present a combination of analysis, transcript, diagrams of gestures and movements, static images, and audio and video records of an interaction. We do not simply get each of these separately; we get linkages between them that enable us to integrate them into a coherent argument, presentation, or analysis. In this sense the multimedia paper itself is an instance of its own major theme: integrating different semiotic resources or fields.
Key Terms:
Semiotic field [cf. semiotic resource system]
Strip of talk
Talk in interaction
Deontic force [cf. Normativity, Deontic modality vs. Epistemic]
Participation framework [longer timescale, niche]
Speech act
Semantic gloss
Pragmatic action
Contextual configuration
Orientation in interaction
Proposals
Deictic expressions, indexical gestures
Reflexive awareness and adaptation
Heterotopia [Foucault ]
Kinesthetic vs. visual components in gesture
Key People:
Fred Erickson – leading educational ethnographer (USA)
John Austin – British philosopher of language
Michel Foucault – French historian and sociocultural theorist
Erving Goffman – pioneer researcher on human behavior in public places
Adam Kendon – leading researcher on nonverbal communication and gesture
Bruno Latour – French sociologist of science and sociocultural theorist
Harvey Sacks – (deceased) pioneer of research on conversation and meaning
Manny Schegloff – collaborator of Sacks, extended his work on conversation analysis
Ed Hutchins – ethnographic researcher who studies situated cognition
Questions:
What is Goodwin’s view of the relation of ethnography to micro-analysis of behavior?
What similarities are there (in terms of their function in the action-and-communication process) between the role of the hopscotch grid and that of the Munsell chart’s grid?
Why does Goodwin reject the idea of “image acts” similar to “speech acts”? Does he reject the idea in all cases, or just for his examples? What alternative does he propose?
How does Goodwin see the human body as a special kind of semiotic field? What is special about it?
What relationship does Goodwin see between “institutional” and “vernacular” action-talk?