Linguistic ethnography and pedagogy in conversation
In the preceding discussion we examined the main differences between the pedagogic and linguistic ethnographic perspectives on recordings of classroom practice, and explored their methodological, ideological and institutional roots. Before moving on to discussion of the actual and possible interactions between the two perspectives, it is worth noting that each perspective is better suited to some purposes than others. Indeed, it would seem that each perspective is relatively poorly suited for addressing the concerns central to the other. The pedagogic perspective's chief problem is its tendency to focus on what is missing rather than what is present in the recorded data. This tendency arises from participants basing their analyses on what they would have expected to find in a 'good' lesson, rather than what actually occurs. This is problematic on four counts: first, focusing on what is absent distracts us from making sense of what is actually going on. Second, it makes weighing opposing interpretations difficult, since they are only loosely related to the common object under discussion; in other words, the recorded lesson serves as a poor basis for assessing the relative merits of participants' different pedagogical preferences that they seek to project upon it. Third, discussion of good teaching that is based on idealised or imagined practice, rather than authentic representations of actual practice, can lead us to adopting or developing unrealistic pedagogical models. Finally, it is usually unfair to the recorded teachers and students: the practices that participants seek to find may have occurred prior to or after the very brief snapshot captured in the recorded episode. This final issue points to a further mismatch between the pedagogical perspective and the sort of materials we brought to the workshops: the 5-7 minute episode is not a good unit of analysis for studying pedagogic processes, which typically last for weeks or even months.
Conversely, linguistic ethnography is less well-suited for investigating learning. This limitation arises from a number of analytic prejudices: the tendency to privilege the 'here and now' of the recorded evidence over longer time scales, for which evidence is not as firm (see Lefstein & Snell, 2011a); the tendency to focus more on social interaction than on cognitive processes; and a preference for explanations based on visible social dynamics over inferences about what's going on inside actors' minds. Learning, which takes place over relatively long time-scales and to a certain extent within people's heads, is marginalised by these preferences. Nevertheless, since learning is not the only phenomenon - indeed, is often not the central phenomenon - to be found in classrooms, we argue that the sorts of issues that linguistic ethnography uncovers are crucial for a well-rounded and balanced understanding of pedagogy. Our challenge is to bring the two perspectives together.
How did the two perspectives interact in the workshops? Most of the time we talked past one another. Following acquaintance with the data
(listening and/or watching the recording twice, raising and responding to questions about the context) we conducted an initial review of issues that participants were interested in pursuing. This review was typically dominated by comments originating in the pedagogic perspective: e.g., what participants would have expected to see, what the teacher should be doing, how good the teaching is, or how the events observed resonate with their own experience. Some of these comments were contested, and discussion of the pedagogic merits of the case typically ensued. The course leader interjected here and there a request for evidence from the text, or a reminder that we should try to understand the event before judging it, but for the most part this pedagogic discussion continued for about half the workshop session. Eventually, the course leader shut down the pedagogic conversation, and focused the participants' attention on a specific moment or issue that he claimed was worthy of deeper (i.e. linguistic ethnographic) examination. In the subsequent micro-analytic brainstorm the course leader typically deflected attempts to re-inject pedagogic issues by requesting that participants attend to what is happening in the episode (usually, in the specific 2-3 turns that were the object of analysis) and by demanding specific evidence for interpretations offered. In such a way, the workshops were divided into two more or less independent segments - the first pedagogic, the second, linguistic ethnographic - which rarely interacted with one another in any meaningful way. The main exceptions to this generalisation were events in which the linguistic ethnographic analysis directly challenged pedagogic assumptions and interpretations, or occasions in which the class engaged in meta-methodological reflection, e.g., about our tendency to rush to judgment about teaching quality even before fully understanding what is going on.
While these interventions are useful for teaching linguistic ethnography and its merits, we wonder if it might be possible to accomplish richer and more productive conversations between the perspectives. In moving forward, we want to experiment with making space for quality pedagogical reflection. One of the problems in the way we structured the workshops was that we rarely had time, conditions or tools to explore seriously participants' pedagogic reflections. Paradoxically, though we argued that the pedagogic discussion should follow the linguistic ethnographic one, in actuality the order was reversed, and we never managed to come back to the pedagogic issues that were raised at the outset. Perhaps if participants knew that we would devote the latter 40-60 minutes of each workshop to pedagogic discussion they would more readily delay their pedagogic reflections until then.
No less importantly, quality pedagogic exploration demands appropriate materials and tools, including:
- (a) Data about learning processes, including e.g., examples of student work, sampling of discussions over time, etc. Comparative data may also be productive, e.g., data comparing two strategies for teaching the same materials or students.
- (b) Some agreement about pedagogical goals and principles, i.e. about what is good teaching and what it should strive to achieve - at the very least provisional agreement for the sake of the workshop.
- (c) Tools for analysing learning processes. Here we can build on microanalytic and transcontextual methods that have been developed outside of the linguistic ethnographic community, for example Parnafes and diSessa's (2013) method for 'microgenetic learning analysis', or the way in which Wortham (2006) investigates the evolution of cognitive models articulated in classroom discourse over the course of multiple events of cognition. One limitation of these and related methods is that they require reduction and organisation of relatively large data sets prior to micro-analysis: one cannot simply apply them to whatever 5-minute strip of interaction students have brought to a workshop.
- (d) Methods for relating analyses of learning processes and classroom interactional dynamics to conclusions about teaching. Just as we cannot derive ought from is, we cannot straightforwardly derive teaching strategies from theories or evidence of learning. We believe that a promising way forward is to focus on dilemmas, and to attempt to sharpen teachers' professional judgment by examining (a) What is happening (including, what are students learning)? (b) Why is it happening? (c) What are the possible strategies or moves the teacher can employ? And (d) What are the advantages and disadvantages of each? (See Lefstein & Snell, 2014, for an elaboration of the rationale behind developing teacher professional judgment through discussion of dilemmas, and for examples of analyses of classroom episodes.)
We view such tools and the pedagogic exploration they're intended to support as complementary to and potentially building upon linguistic ethnographic analysis. Our challenge is how to combine both perspectives in the limited time frameworks in which we work. We are convinced that linguistic ethnographic analyses have pedagogic value; indeed, it stands to reason that their relevance will become clearer when combined with quality pedagogic discussion (instead of competing with it).
In closing, we would like to reflect on the broader educational context of the issues discussed here. What's at stake is much larger than the success of a seminar course or its participants' satisfaction. It is now commonly accepted that what happens in the classroom is the key determinant of the quality of school education, and that teacher participation in critical, collaborative conversations about their practice is a crucial component of any attempt to improve teaching. However, not all professional conversations advance their participants' insights and judgment to the same extent. The analytic dispositions, forms of reasoning and breadth of considerations that linguistic ethnography offers can contribute significantly to teacher professional discourse and learning. Our challenge is to adapt, complement and mediate linguistic ethnography in ways that are constructive, have integrity, and are recognised as helpful by practitioners.