What we see is shaped by what we expect to see, i.e. what we assume to be happening in classrooms. Such expectations and assumptions are not typically explicated by teachers or researchers, but we can infer them from what participants notice and how they talk about it. The key point of difference between the two perspectives is that teachers generally assume that most classroom activity is about learning, while linguistic ethnographers tend to be suspicious that displays of formal learning often reflect other, non-academic issues and concerns. Consider, for example, the following contribution to the event that Tomer brought to the data analysis workshop. Following a discussion of the stereotypes used to describe Jews in medieval Europe, one of the pupils, Noa, recounts how she learned to stop using the term 'blonde' as a stereotypical description of unintelligent girls (as in 'ditzy blonde'):
All my best friends are blondes. So on many occasions I've said to them, 'What a "blonde" thing to say', meaning like ditzy. Actually sometimes - once my friend told me that it is really not nice that I say this to her and that I'm actually making a generalisation and she's actually pretty smart and gets good grades. So I've stopped doing it because I also realised it was a generalisation which is not very nice of me. So basically I think they didn't under—. They generalised about all Jews and said that all of them are dishonest, all of them are this. When actually they're people who may be amazing.
A few of the teachers participating in the discussion pointed to this contribution as an ideal example of identity exploration. Noa, after all, appears to be relating the historical material being studied to her own life, and using the curricular contents as an opportunity to explore who she is: a person without prejudice, who doesn't make judgments on the basis of friends' hair colour.
As linguistic ethnographers, we interpreted Noa's contribution differently. The seminar lecturer noted, first, that part of what Noa was doing was to identify herself and friends publicly, in addition to commenting on the historical material. Second, he wondered to what extent she has actually learned something from the history lesson, and to what extent she is performing the role of the ideal 'identity explorer'. He pointed out what a well-crafted conversion narrative she has presented, and how it conforms to teacherly expectations for denouncing prejudice, being sensitive to others and learning from mistakes. The differences between the teachers' and our interpretations were shaped in part by divergences in our assumptions and expectations. Rather than assuming that students are by definition engaged in learning, we assume that much of what happens in classrooms is related to managing one's identity and social relations, and performing school-appropriate roles (without necessarily engaging in learning; see, e.g., Bloome, Puro & Theodorou, 1989).
A second key difference between the assumptions underlying the two perspectives relates to the presumed distribution of agency and responsibility. While the pedagogical perspective tends to attribute almost all responsibility for classroom activity to the teacher, who is assumed to be generally in control of the classroom, in the linguistic ethnographic perspective power relations are co-constructed by teachers and students, with the latter often constraining the former no less than the former influence the latter. So, for example, in many of our video clips the teachers talk most of the time. Teachers viewing these clips interpret the situation as one in which the teacher has 'dominated' classroom discourse, has occupied all available space and thereby crowded out pupil voices. When we look at these clips, we often see the teacher offering pupils opportunities to participate, for example, by posing questions to them, but the latter have not cooperated - they 'coerce' the teacher to talk through their own silence.
A third difference relates to assumptions about pupils' ability and how central it is for understanding what happens in classrooms.
Teachers often use 'high' and 'low ability' as key factors for predicting or explaining student classroom participation and academic success. Ability is often assumed to be innate, and inherited (either genetically or socially) and therefore for the most part beyond the control of the school. Thus, for example, in Tomer's lesson, the participating teachers explained the pupils' good behaviour and relative articulateness as a function of the school's location in affluent north Tel-Aviv. For linguistic ethnographers, low and high ability are viewed as social constructions, the context-dependent product of pupils' interaction with their environment rather than some innate quality that moves with them from place to place. Schooling is organised such that some pupils appear 'bright' and others 'slow', and part of the analyst's task involves uncovering these processes (see Varenne & McDermott, 1998, on the way that schooling constructs social categories of relative ability, and Snell & Lefstein, 2012, on how teachers' interpretations of classroom events are shaped by their assessments of pupil ability).