Accounting for Conflicting Mental Models of Communication in Student-Teacher Interaction: An Activity Theory Analysis

Using activity theory as a framework, this article discusses a naturalistic study of two college classrooms in which the instructors often relied on transmission models of communication—models assuming that stable, fixed meanings can be neatly transmitted from person to person. Particularly noteworthy was that these instructors seemed to rely on transmission models despite training in recent theories of communication and that, contrary to previous assumptions that people's communicative models are stable, both teachers shifted in and out of these models. Based on an analysis of the contexts surrounding shifts into transmission models, the article argues that these shifts happened in patterned ways. It then accounts for the resilience of transmission models within a broader sociocultural framework. We have long seen as problematic reliance on transmission models of communication—models which do not recognize the role of context in shaping interpretation and which assume that stable, fixed meanings can be neatly transmitted from person to person. Reddy (1979) has noted English speakers' extensive reliance on this transmission model, or as he calls it, the conduit metaphor. Thanks to Bakhtin and others, however, a variety of academic fields now understand that people can and do interpret utterances differently because of the uniqueness of their personal histories, material conditions, and belief systems. Composition studies, in particular, has achieved a consensus in critiquing transmission models (Eubanks, 2001). The data discussed in this article, however, suggest that two scholar-teachers trained in recent theories of meaning-making nonetheless relied on transmission models. They did so, moreover, both when communicating with students and when reflecting on that communication in interviews. Drawing on a qualitative research project examining student writing and teacher response in two college classrooms, I note not only that these teachers seemed to rely on transmission models but also that, contrary to previous assumptions that people's communicative models are stable, they shifted in and out of these models. How is it possible that teachers trained in recent theories of communication can find themselves intermittently, but consistently, acting as if meaning were neatly and unproblematically transmitted? A useful account can be provided by activity theory, particularly as elaborated by Engeström (1996). Like other advocates of activity theory, Engeström points out that we need to broaden our unit of analysis when we examine complex human interactions; thus, in this study of Writing Selves/Writing Societies, Bazerman & Russell Published February 1, 2003 http://wac.colostate.edu/books/selves_societies/ Copyright © 2003 by the Authors & Editors Conflicting Mental Models of Communication, Evans Page 394 two college classrooms, the basic unit of analysis won't be discrete entities such as "the teacher," "the students," or "the student texts" but rather an entire system of student-teacher communication—a system that involves complex interconnections among teachers, students, student texts, and means of responding to those texts, in addition to the myriad beliefs, histories, material conditions, etc. that influence these people, tools, and practices. Engeström elaborates: An activity system is not a homogeneous entity. To the contrary, it is composed of a multitude of often-disparate elements, voices, and viewpoints. This multiplicity can be understood in terms of historical layers. An activity system always contains sediments of earlier historical modes, as well as buds or shoots of its possible future. These sediments and buds—historically meaningful differences—are found in the different components of the activity system, including the physical tools and mental models of the subjects. (p. 68; emphasis added) In the case of the research discussed here, activity theory helps explain "disparate elements" such as the conflicting mental models of communication that the teachers seemed to rely on; it helps us to better account for seemingly incompatible beliefs by seeing those beliefs as rooted in a complex system in which one shift in any part of the system can reverberate throughout the entire system. These reverberations in the system will be apparent throughout the article as I account for why the two teachers shifted in and out of transmission models of communication. I suggest that, at least for these two teachers, transmission models were triggered by the arena of discourse in question (related to the teachers' training) as well as a complex interaction among a variety of other implicit theories, including perceptions of student competence and effort and perceptions of task difficulty. Then, suggesting that transmission models are both pervasive and resilient, I account for this resilience on a macro-social level. I conclude by recognizing that transmission models are sometimes appropriate and by discussing implications for research, theory-building, and teacher training. Theoretical Background Limitations of Transmission Models of Communication Although scholars in the humanities and social sciences are keenly aware of the limitations of transmission models, a brief overview of recent thinking on this issue will help to lay out the specific sets of assumptions that are at play in the teachers' behavior and that they themselves critique in their stated beliefs. The limitations of transmission models are especially wellillustrated by recent theories of communication which hold that divergent understandings are an inevitability rather than a conversational aberration that can be avoided simply by being "open" and "honest." Among the primary scholars advancing this "miscommunication as the norm" claim are Coupland, Wiemann, and Giles (1991), who write: "language use and communication are in fact pervasively and even intrinsically flawed, partial, and problematic. To this extent, Writing Selves/Writing Societies, Bazerman & Russell Published February 1, 2003 http://wac.colostate.edu/books/selves_societies/ Copyright © 2003 by the Authors & Editors Conflicting Mental Models of Communication, Evans Page 395 communication is itself miscommunicative" (p. 3). These assumptions grow in large part out of ethnomethodology and discourse theory: Some semantic "slippage" is common in the management of meaning transfer, and in fact there are many reasons to suppose that this is inevitably the case. If we acknowledge that speaking occurs (a) under real-time processing constraints and (b) within the lexical and syntactic confines of particular linguistic codes, we must doubt that there are such entities as pure, unsullied and perfect semantic representations. In the ethnomethodological tradition, language use, the making of meaning and its reconstruction, has been viewed as inherently problematic, strategic, and effortful. Garfinkel's (1967) perspective on talk as "accomplishment" acknowledges the probability of communicative inadequacy and incompleteness. Discourse theory (van Dijk, 1987) similarly asserts that utterances are intrinsically indeterminate. Because meaning resides in the interaction of linguistic form and social context, exchanges of meanings operate under inherent constraints and communicative acts are creative in compensating for the inexplicitness and indirectness of speech acts and texts. (1991, p. 5; italics original) Particularly important here is the notion that "exchanges of meanings operate under inherent constraints [given that] meaning resides in the interaction of linguistic form and social context." Completely shared meaning, then, such as is assumed in transmission models, would require at least two (interrelated) phenomena; first, it would require a shared linguistic form, which is precluded by the fact that, even when people happen to speak the same dialect, each person speaks a slightly different idiolect. Second, it would require a shared context, which is precluded by the uniqueness of our personal histories. Even what may appear to be a shared context in the moment that two people are interacting is far more complex than it appears. While a given moment in time may be shared by two people, the lived histories of each person prior to that moment are unique. These unique histories (and concomitant assumptions and ideologies) shape each person's interpretation of that particular moment, and thus people's interpretations of a seemingly shared moment may be radically different. Thus, as Chin (1994) points out, it is important to examine not just particular material and social worlds but also people's individual readings of these worlds. Because each person's reading is necessarily unique in some way, we can see the limitations of assuming that our intended meanings are transmitted neatly from our minds to our interlocutors'. The limitations of transmission models are further implied by Goodwin and Duranti's (1992) point that context—in addition to being multiple and contested within a moment (i.e., one speaker's "context" may differ from another's)—is also dynamic rather than stable. That is, even if two participants did share a similar context at a given moment of an interaction, that context is constantly shifting—and often doing so in different ways for different participants. Interaction not only reflects but also shapes and creates context. In a given interaction, for example, one person may invoke a new context by referring to a new topic that would cue different schemata. New contexts might also be invoked by contextualization cues (Gumperz, 1982), features of linguistic form such as prosody and lexical and syntactic choice that help signal contextual presuppositions. Gumperz notes that new contextualization cues may be introduced in the course Writing Selves/Writing Societies, Bazerman & Russell Published February 1, 2003 http://wac.colostate.edu/books/selves_societies/ Copyright © 2003 by the Authors & Editors Conflicting Mental Models of Communication, Evans Page 396 of an interaction and that some participants may fail to notice this. In sum, then, not only are transmission models inadequate given that meaning resides in the interaction of language and context—neither of which is totally shared by any two given people—but it becomes even more inadequate given that these contexts are dynamic. Every "context" or aspect of the world is necessarily seen through the lens of our senses, thoughts, feelings, and assumptions. Thus we have interpretations of contexts becoming part of contexts which then shape interpretations which may in turn shape other contexts and interpretations, etc. Seeing context in this way—as dynamic, complex, multi-layered, and rooted in interpretation at every turn—confirms the limitations of transmission models. Transmission Models of Communication: A Continuum Because the term "transmission model" is used in different ways, and because scholars have different means of determining when such a model is being relied on, these issues are worth examining in more depth. To accomplish the work of monitoring the implicit models that shape how we communicate, it will be useful to conceptualize transmission models of communication not as a single entity but rather as a family of closely related models. Although many scholars who discuss transmission models do view them as a single entity (cf. Reddy, 1979; Rommetveit, 1988), this article will, instead, conceptualize them as a continuum. On one end of this continuum is the assumption that meaning is automatically transferred; at a mid-point is the assumption that meaning should be transferred but is occasionally blocked by a faulty sender or receiver (what could be called a "messy" transmission model); and at the other end is the assumption that a variety of factors such as "noise in the channel," faulty senders, and faulty receivers often block transmission of meaning. The research on which this article is based suggests that the latter point on the continuum—what might be called a "sophisticated transmission model"—is relatively rare in comparison to the previous two points, and it is these first two points to which I will generally refer by the term "transmission models." The alternative to transmission models is referred to as a "post-transmission" model; this term will be used to describe assumptions that interpretive difference is normal and that contexts shape interpretation. In both cases, the term "model" refers to people's implicit or explicit theories about how communication works. Reliance on Transmission Models in Two College Classrooms The remainder of the article will explore the influences that seem to trigger people's shifts into transmission models. These influences will be illustrated on a micro-social level by a semesterlong naturalistic research project that examined both the written and oral response rounds occurring in two college classes. One of the goals of the study was to explore the models of communication relied on by the two teachers, who will be referred to by the pseudonyms of Rick and Lynn. In discussing these cases, the goal—like that of most case study research—will be to illustrate the existence rather than the typicality of the phenomenon in question. While I do Writing Selves/Writing Societies, Bazerman & Russell Published February 1, 2003 http://wac.colostate.edu/books/selves_societies/ Copyright © 2003 by the Authors & Editors Conflicting Mental Models of Communication, Evans Page 397 believe that transmission models are typical and pervasive, the case for this will be made later, in the macro-social analysis. In this section, the micro-social analysis should help readers decide how many of the patterns illustrated by Rick and Lynn apply to their own situations.