The Language of Law School This page intentionally left blank

Participant Observation
, 3–16 and passim; Geertz, 
Interpreta-
tion of Cultures
, 3–30; Greenhouse et al., 
Law and Community
, 7–21.
2. See, e.g., Mehan, 
Learning Lessons
; Cazden, 
Classroom Discourse
; Heath, 
Ways with
Words
.
3. Cazden, 
Classroom Discourse
, 3–4.
4. For a study of differential treatment of students in high- and low-ability reading
groups, see Collins, “Socialization to Text”; for a study of schools in two neighboring rural
communities, see Heath, 
Ways with Words
; for a comparison across classrooms with stu-
dents from differing cultural and social class backgrounds, see Cazden, 
Classroom Discourse
.
5. Conley and O’Barr, 
Rules versus Relationships
, 125, 172–177.
6. Sarat and Felstiner, 
Divorce Lawyers
.
7. Conley and O’Barr, 
Rules versus Relationships
, 165, 173.
8. Matoesian, 
Reproducing Rape
 (tracing gendered patterns in talk about rape in the
courtroom); Merry, 
Getting Justice and Getting Even
 (tracing differences across class in
discourse and orientation toward law); Frohmann, “Discrediting Victims’ Allegations.”
9. If a student’s identity was not clear from the tape and was missing from the in-
class coder’s sheet, we coded it as “unknown.”
Notes to Pages 29–33
239

---

10. Interestingly, students in several of the schools actually approached our coder
before the coder had invited students to these small group interviews; they wanted to be
sure their perspectives would be included in the study. As noted, I also performed on-site
intercoder reliability checks. In addition to passing review by the Human Subjects Com-
mittee of the American Bar Foundation, we underwent additional IRB review as requested
by individual schools.
11. Thus, for example, a “focused” dialogue was one in which the professor inter-
acted with one student for more than two exchanges (four turns); a “nonfocused” dialogue
involved the professor engaging several different speakers for no more than one consecu-
tive turn each.
12. In several cases where coders left the project before completing an entire semes-
ter of tapes, another coder checked each turn that had already been coded, in effect
recoding, and then finished the remaining classes. We were fortunate to have one stalwart
coder who single-handedly coded many of the schools, thus ensuring quite high consis-
tency even between schools. She also served as a standard when other coders encountered
difficult coding decisions. We also conducted formal intercoder checks.
13. Nancy Matthews and Susan Gooding, the two project managers, were each me-
ticulous in their concern for detail, often personally checking turn-by-turn through the
schools to ensure consistency in coding.
14. In addition, we developed overviews of the semester for each classroom, both
through overarching school ethnographies and through tracking sheets that charted how
doctrines and legal issues were presented through the semester, and kept track of how social
context, role-playing, and metalevel discussions were handled in each class.
15. A classic discussion can be found in Sacks et al., “A Simplest Systematics,” 7; see
also Atkinson and Drew, 
Order in Court
; Matoesian, 
Reproducing Rape
.
16. Transcription conventions are outlined at the front of the book. In adopting them,
I have been responsive to Conley and O’Barr’s admonition: “Those of us who do law and
language research must . . . rethink our transcription conventions. . . . [Some] transcripts
are extremely difficult for nonspecialists to work with.” Conley and O’Barr, 
Just Words
,
139.
17. For purposes of much of the quantitative analysis reported here, we removed
backchannel and background comments. However, we retained the coding for future analy-
sis, which can compare and contrast different kinds of classrooms and speech styles along
these dimensions as well.
18. As noted, conversation analysts have developed a useful analytic framework for
studying conversation. An adjacency pair is a unit of conversation consisting of two turns,
in which one speaker takes the first turn (the first “pair-part”) and another speaker takes
the second (the second pair-part). Generally, the first pair-part provides a frame for the
second; so, for example, if I greet you (“Hi”), my first turn places some pressure on you to
greet me back. Or, more to the point here, if I ask you a “question,” there is a strong con-
versational framing favoring a second pair-part “answer” to that question. See Levinson,
Pragmatics
, 303–304, for an introduction to these concepts.
19. The silent turn here functions in the same way that a spoken response might if
the student had said “I don’t know” or “Could you give me some help?” In this sense, it
could be interpreted as a “zero sign,” indicating, perhaps, that the speaker is having diffi-
culty answering.
20. The term “minority” has rightfully been critiqued as Eurocentric; worldwide,
it is clearly the case that the many groups lumped together under that aegis are the ma-
jority. Because I am synthesizing work from numerous studies (focusing on the United
240
Notes to Pages 34–38

---

States, where whites have traditionally formed the majority of the population), I use
various terminologies; when the word “minority” is used, it is to be read as having a
bracketed meaning.