Green 1983 PDF

Chapter 6
Research on Teaching as a Linguistic Process:

A State of the Art
JUDITH L. GREEN
University of Delaware
In the last decade, several new approaches to the study of teaching-

learning processes have emerged. These approaches have their roots in
disciplines such as linguistics, information processing, and cognitive
psychology. By adapting theoretical constructs and methodological
advances from these disciplines, researchers concerned with educational
processes have developed new ways to study a variety of educational
processes including teacher planning, teacher decisionmaking, the nature of
effective instruction, evaluation of student performance, as well as the
relationship among such factors and the relationship between these factors
and student learning.
This paper presents a synthesis of and state of the art for one emerging
field, teaching as a linguistic process. The synthesis focuses on a cluster of 10
projects (Borman, Piazza, Barrett, & Sheoran, 1981a, 1981b; Cole, Griffin,
& Newman, 1978-81; Cook-Gumperz, Gumperz, & Simons, 1981; Cooper,
Ayers-Lopez, & Marquis, 1981; DeStefano & Pepinsky, 1981; Erickson,
Cazden, Carrasco, & Guzman, 1978-81; Hrybyk & Farnham-Diggory, 1981 ;
Hymes, 1981a; Merritt & Humphrey, 1981; Morine-Dershimer &
Tenenberg, 1981 ~) sponsored by the National Institute of Education through
its 1978-79 grants program. These projects constitute a core of first-genera-
tion studies and projects (see Table I). Additional work related to these
The work on which this publication is based was performed pursuant to Order No. NIE
P-81-0084 of the National Institute of Education. It does not, however, necessarily reflect the
views of that agency.
The author would like to thank Dr. Judith Harker (Veterans Administration Hospital,
Northridege, California). Dr. Marjoris Arnold, and Dr. Gladys Knott (Kent State University)
for their editorial comments and Deborah Smith (University of Delaware) for her work on the
analysis of findings and insights into the uses of the c o n s t r u c t s presented.
The editorial consultant for this chapter was Courmey Cazden.
151
TABLE I
u1
NIE-sponsored Projects: General Structural Components of the Research Design ro
Project Population General Participant Topic-centered Natural Experiments,'

Observation Observations Experimental Tasks
Borman et al. 2 classes, grade 2, 3, Playground: layout Games students played Behavioral measures:
(1981a, 1981b) 5, 6 structure and organized a. Hopscotch/Freeze tag ~"
(NIE G-79-0123) Inner city school activities --formal (Videotape)
Low SES --informal Hopscotch/kickball O
Cincinnati, Ohio Patterns of player activity (Videotape)
of turn taking game rules
Patterns of time spent in b. Chandlers bystander {=
activities preceding cartoons
game, organizing game, perspective taking
and playing game c. Selman's "First ~.
Patterns of distractions Things"
associated with turns at moral judgments
play d. Bruininks-Oseretsky
Patterns of game mainte- Test of Motor Profi- "
nance strategies asso- ciency o
ciated with boys' and (Short Form)
girls' games motor skills
e. Raven's Coloured
Progressive Matrices
intellectual and con-
ceptual skills
f. Playground logs
g. Teacher ratings
academic perfor-
mance
social skills
h. Peer nominations
popularity/status
i. Friendship ratings
Cole, Griffin & Newman Third grade Classroom: Observe and record con- Cycle of planning and in-
(1978-81) Integrated school organization tent delivery using vid- struction
(NIE G-78-0159) San Diego, California grouping eotapes, audio record- Curriculum experiment:
teacher-student inter- ings, and field notes Project personal, with
actions 1. tutorial (adult-child) teacher involvement,
curriculum 2. teacher-led large group plan unit of instruction
3. teacher-led small group that meets following cri-
4. student cooperative teria
groups 1. content fits general
5. clubs curriculum
a) computer 2. content novel; that
b) backpack is, not within regular
curriculum or student
activities (e.g., chem-
ical intersection task) .~.
Content was planned for
delivery under five con- =
ditions ~.
1. tutorial (adult-child) ~"
2. clubs: Computer
Capers and Back- ~==
pack Bears to ex- =
plore occurrence in
nonschool settings
3. Lessons
a. teacher-led large E
groups ~"
b. teacher-led small
groups 0
c. student only worL" o
time (cooperative- =
ly designed by
teacher and re-
searcher)
Each unit is designated .~
(Continued on next page) as a cycle. Each cycle
informs the next; there-
fore, inquiry evolves.
TABLE I Continued
NIE-sponsored Projects: General Structural Components of the Research Design
Project Population General Participant Topic-centered Natural Experiments/
Cook-Gumperz et al. First and fourth grade Teacher planning and Select activities--using Natural experiments used
(1981) Integrated school pedagogy--participant contextualization cues to explore target phe- ~"
(NIE G-78-0082) Berkeley, California observer is participant and other observed pat- nomena (e.g., narrative
in class and assumes terns, identify signs of structure, metalinguistic
three roles: discrepancy in events awareness, storytelling) :~
1. participant in events, that reflect problem or in more constrained
then records events is an example of differ- way. These activities
in notes ential learning form a contrastive set
2. observer of events; Focus on discrepant with the naturally occur- =r
participates only if events--begin to predict ring instances of the ~"
approached for help cues of what will and phenomena
3 teacher's aide will not occur Pear stories (Chafe) to i=
Each role provides differ- 1. verbal explore oral and written
ent view (insiders' and 2 nonverbal narrative, to explore
outsiders' perspectives). Begin systematic observa- thematic cohesion de-
Observation of teacher tion and select video- vices. Contrast with
planning taping sharing time activity in
1. explicit information Collect data on phenome- kindergarten
from teacher interna in other contexts; Referential Communica-
views e.g., home tion Task (Kraus &
2 implicit information Naturally occurring activi- Glucksberg)
from observing ties studied include: and
teacher behaviors 1 reading group in- Phonemic Segmentation
a plans struction Task
b. organizational 2. sharing time To explore children's
structures 3. peer network forma- use and its relationship
c. delivery of instruction to reading achievement
tion To explore children's
Observation of classroom phonological awareness
processes to ascertain with reading achieve-
1. formal segments of ment
day To explore children's
2, ways in which teach- metalinguistic aware-
er frames activi- ness with reading
ties/events achievement
3. contrastive situations Storytelling Task
in which unexpected To explore the relation-
or atypical events ship between recogniz-
contrast with expect- ing and producing well-
ed or typical situa- formed stories and the ~,~
tions development of reading
Interview participants for- skills
mally and informally to Science Laboratory .~.
obtain their perceptions Experiment: Lawrence
and verification Hall of Science
To explore language :~
use and problem-soIv- ~"
ing in setting other than
school. Videotape visit (n
to Lawrence Hall of Sci- =0
ence and work in la- r'-
boratory. This is a peer ~"
language situation, i=
Audiotape collection of ~"
.q.
parent-child discourse in ~"
home
To explore narrative
structures used at
home and to compare
with those at school
(Continued on next page)

TABLE I-Continued
NIE-Sponsored Projects: General Structural Components of the Research Design o}

Cooper, Ayers-Lopez, & 1 kindergarten working Peer interactions during Peer communication tasks
Marquis class individual work time to explore children's
(1981) 1 second-grade parochial Field notes describe non- discourse skills under
(NIE G-78-0098) Austin, Texas verbal behaviors, seat- conditions of minimal o
ing arrangements, and distraction and to ex- :~
other pertinent contex- plore skills when role of
tual information cooperating partner,
Audio recording of chil- teacher, and learner did
dren working on class not have to be negotiat- =r
assignments ed. ~-
Tapes reviewed and in- Task 1: dyadic inter-
dexed for occurrence of action~Cooperative ~.
instructional episodes /earning activity
Typology constructed Task 2: dyadic inter- ==
for episodes action--Asymmetrical
Patterns of student work knowledge (one stu-
during individual work dent had knowledge "'
time and taught other)
Patterns of teaching and Both activities involved
learning bids blocks and a pan-
type balance scale
DeStefano & Pepinsky 1 first grade Discourse of teacher and Teacher evaluation of
(1981) Inner City three target students each student's success
(NIE G-79-0032) Appalachian, black, during teaching/learning to construct "Literacy
white of literacy in the class- Success Profile"
Columbus, Ohio room was collected via 1. evaluation informa-
audio- and videotape tion as determined
1. individuals wore by assignment of
wireless microphone subjects to class-
to permit collection room reading
of subvocalization groups, interviews
and interactional dis- with teachers, report
course cards
2. videotapes available 2. score on Ctay's Con-
to provide check cept of Print Survey,
against audiotape Sand Test measures
records and to ex- limited set of con-
amine elements of cepts about print and
nonverbal behavior engaging with print
3. field notes served as 3. scores on criterion-
third type of check. referenced tests )
4. teacher interviews of 4. classroom read-
concepts, classroom ing/writing behaviors
values, and expecta- 5. scores on Clay's *o
tions tied to literacy written language

evaluation proce-
(1
dure.
Erickson et al. (1978-81) 2 first grades Classroom: Periodic videotaping and

(NIE G-78-0099) Bilingual school teacher-led large participant observation
Low SES groups Select observation of acti- w
Chicago, Illinois teacher-led small vities
groups Observation on nine tar-
informal peer small get students (selected
groups on basis of language
peer-led instructional proficiency and aca-
groups demic performance)
Range of sociolinguistic
variation in terms of so- Q
cial strategies and lan-
guage/nonverbal com-
munication used

TABLE I Continued
NIE-sponsored Projects: General Structural Components of the Research Design

Erickson et al. (Con'd) Subset of individual chil- q

dren's repertoires used
across event contexts
"instructional chains" o
and "naturally occurring
peer tutoring" t~
"Cultural" aspects of or-
ganization of social
strategies and commu- ~r
nicative functions in
classroom events, i.e.,
participation structures
of the events o
t=
Patterns of personaliza-
tion and 'privatization" .=
of instruction in
t, whole group lesson
in math or language
arts
2. assignment of seat-
work to individual
children
3. monitoring by teach-
er of individual chil-
dren's seatwork as
they do it
4. patterns of praise
and/or feedback to
children in privatized
context, and the dif-
ferences between
private and public
personalizing
Ethnography of learning
1. how children hold
each other account-
able for "the social
order"
2. how adults hold chil-
dren accountable for
the "social order" Q
3. how teachers learn
about what children Q
can do academically .=.
and socially
Hrybyk & Farnham-Dig- Second grade, private, Observation of Sociometric question-

gory (1981) parent cooperative 1. characteristics of the naires to assess group
(NIE G-79-0124) Eighth grade, parochial school organization structure
Delaware-Maryland area 2. children's social or-
ganization
3. neighborhood and
family organization 5"
Interviews to obtain infor-
mation on:
1. children's conception
of school organiza-
tion
2. children's conception o
Q
of peer organization w
of neighborhood or-
ganization
(Continued on next page) of the workplace
TABLE I Continued
NIE-Sponaored Projects: General Structural Components of the Research Design o

Hymes (1981),
Smith
(University of Pennsylva-
nia) o
(NIE G-78-0038)
DavisD Two classes each, grades Focus: school-community

4,5,6 perspective on back-to- a
Three schools Inner city basics school. Explored
Philadelphia, Pennsylva- perceptions of
nia 1. school policies
2. homework
3. literacy materials in (1
home m
4. parent expectations
for students and .=
school
o
5. community perspec-
tive
Andersonb Focus: life-history of an
administrator explored
contrasting view of
school administrator
and expectations par-
ents hold for school
aThis study is composed of eight substudies, focusing on three different schools. Each study focused on one theme
bSame as information for General Participant Observation.
Watkins ~ Focus: reciprocal perspec-
tives of teachers/parents,
Explored
1. teacher expectations
of parents
2. parents' perceptions
of school expecta-
tions
3. the continuity or dis-
continuity of these
perceptions
Also explored teacher Q
and parent perceptions
and expectations of
1. homework .=
2. reading at home
3. parent involvement
(3
Gilmore b Focus: interdependence
of student expressive
behavior with school
=
success in terms of
1. admission to special =
programs
2. extinguishing unde-
sired behavior within
the program
Also explored communi-
cative competence in
1. the home
2. the school (3
CO
3. peer situations h
4. teacher-student situ-
ations
5. formal activities
6. informal activities
(Continued on next page) 7. the community (e.g.,
Girl Scouts)
TABLE I Continued
NIE-sponsored Projects: General Structural Components of the Research Design ~o

Q
Gilmore ~ (con'd) In terms of language, she ,,=
explored the functions of ~"
silence in an interactive
situation and sulking o
Lussier o Focus: tactics of individual 23
boys in the classroom
against the background
of school and communi-
ty environment. Ex- ~"
plored participation of ~"
individual boys and m
peer culture's role in
this participation
May~ Focus: monitoring atten- =
tion. Explored O
1. what counts as at- .=,&
tention/inattention o
2. verbal styles that
can mask attention
Also explored teacher-
student interactions and
responses during infor-
mal and formal time in
classrooms
Woods-Etlio~ Focus: a case study on
writing instruction in
one classroom. This
project explores how
writing was used and
taught as part of learn-
ing-centered routines in
content areas as well
as during formal writing
instruction. The study
explored
1. rules for writing
2. individual's use of
writing
3. official writing
4. unofficial writing
5. collaborative writing
6. teacher orchestration
and demands on .=.
teachers
FieringD Focus: written literacy in Informal experiments in {D
(1
children's lives. Ex- curriculum based on
plored systematic observations ~Q
1. the meaning of liter- of the ethnography.
acy Titled: "What ifs"--pro-
2. difference between vides basis for explor- f~
directed and sponta- ing uses of informal ac- r"
neous reading and tivities of student, cul-
writing in classrooms ture for formal
3. nature of official and schooling activities
unofficial reading (1
and writing
4. variation in student
participation style (1
Q
=
,.L

TABLE I--Continued
NIE-sponsored Projects: General Structural Components of the Research Design ~'

Smith Discusses the nature of
findings in terms of ~"
1. official and unofficial
activities that make o
up everyday life in :~
classrooms t~
2. nature of ethno-
graphic monitoring
3. goals of ethnograph- =r
ic monitoring ~.
4. criteria for ethno- rn
graphic research O.
5. limitations of this re-
search
Merritt & Humphrey Two classes each: nur- Focus: review corpus and Focus: search data bank 0
(NIE G-78-0159) sery school, kindergar- familiarize researchers for instances of service-
ten first, second, and with primary project in like events. Explore: "~
third grades which the present sec- 1. over grade levels
Private school ondary analysis is em- 2. ways teachers man-
bedded age primary and
1. review videotape secondary vectors si-
records and field multaneously
notes using indexing 3. ways students gain
system help while in the
2. site visit to original secondary vector
site 4. shifts in these events
3. systematic search of in different contexts
corpus 5. factors contributing
to successful nego-
tiation of these
events by students
Morine-Dershimer & Ten- Two classes each, grades Student and teacher per- Stimulated recall inter-
nenberg (1981) 2,3,4 ceptions of classroom views
(NIE G-78-0161) Integrated school (Span- language were ex- Using six 1/2 hour video-
ish-speaking, black, plored. The project ex- tapes, each pupil
white) plored viewed three different
San Francisco Bay Area, 1. pupils' conceptions lessons. Student re-
California of units, salient fea- sponded to series of
tures, functions, and tas ks:
rules of discourse 1. sentence completion
2. match between stu- task on "rules" of
dents' and teacher's discourse, construct-
conception of dis- ed on the basis of
course features and pupil response to an
rules open-ended question
3. factors that support about "how people .~.
or constrain the cor- talk in classrooms"
respondence be- 2. Generating sen- {,
tween teacher and tences which might ~,
student conception be said by (or to) ~"
of discourse features the pupil to "get
and rules someone's attention"
4. difference in percep- or "get someone to ~,
tion of features and do something" p
rules at home and in 3. Reporting "what you ,~
play settings heard anybody say-
5. the relationship being" after playbacks ~"
tween teacher evalu- of short video seg-
ation of students and ments of lessons in -~
student perception which pupils had
and expectations of participated (re-
students' communi- sponses recorded
cative behavior verbatim)

TABLE I Continued "
o~
NIE-sponsored Projects: General Structural Components of the Research Design o~

Morine-Dershimer & 6. continuities and dis- 4. Organizing 3 x 5

Tennenberg (con'd) continuities per- cards of "what you =E
ceived by students heard" into groups of
between language of cards that "belonged o
home and school together because
people were saying Q
7. relationship between
communication be- the same kinds of
havior and academic things"
performance in read- 5. Studying a set of
ing teacher questions 5"
Teacher taught six lan- asked in the lesson
guage arts lessons at (also pupil re-
researcher's request. sponses) and ex- 3
Teacher selected the plaining who said
content. Criteria: these things, to 0
.=
1. Lesson should not whom for what rea-
focus on spell- son
ing/handwriting Similar procedures used
2. Lesson should infor stimulated recall in-
clude the whole terviews for teachers
class Similar procedures used
3. Lesson should infor language of home
clude some verbal and of play settngs
interaction (i.e., no Videotapes used to mea-
seatwork) sure frequency of talk
Presentation of students
with array of photo-
graphs of children in
class. Students asked
to select three children
most likely to fit scenar-
io given.
Used to assess status
Teacher perceptions of
pupils obtained by ask-
ing teacher to group
children on basis of
several different lan-
guage characteristics
which had been identi-
fied in earlier studies as
salient features to
teachers
Pupil entering reading
i
.
achievemenEfinal read-
ing achievement
=}
3
Q
==
Z.
168 Review of Research In Education, 10
projects will be included to complete the state of the art. The work presented
in this paper is a representative sampling of research in the emerging field; it
is not all inclusive.
The synthesis will be presented in four parts. Part 1 provides a historical
perspective on the emerging field and further delineation of the projects that
form the central corpus for the synthesis. Part 2 describes the conceptual
basis for the field. Part 3 discusses the methodologies and the principles
guiding them. Part 4 presents findings in four areas: (a) findings related to
the nature of teaching-learning processes from the teacher's perspective, (b)
findings about the nature of teaching-learning processes from the student's
perspective, (c) findings on the relationship between teaching and learning,
and (d) findings about the relationship between learning processes at home,
in the community, and at school.
THE EMERGING FIELD: AN HISTORICAL PERSPECTIVE

Antecedents of the Fleld
Work on teaching as a linguistic process draws on theoretical constructs
and methodological practices from sociolinguistics, psycholinguistics,
ethnography of communication, child development, anthropology, cogni-
tive psychology, sociology, and educational research on the nature of
teaching-learning processes. These disciplines are the antecedents of a new
discipline which is still in its formative stage. This discipline is concerned
with collecting and analyzing human behavior in natural settings and in
exploring what is learned from and how people learn through interacting
with others. In other words, teaching as a linguistic process research is
concerned with how people learn language, learn through language use, and
learn about language use in educational settings (cf. Halliday, 1982).
Specifically, research in this area is concerned with how language in the form
of interactions between teacher and students, among peers, and between
children and adults functions in classrooms, on playgrounds, at home, and in
the community in support of the acquisition and development of other types
of knowledge (e.g., academic content, social cognition, and knowledge of
procedures for participation in ongoing events).
Consolidation of the Field: NIE's Role

Consolidation of this research can be traced to two separate but related
actions on the part of the National Institute of Education (NIE). In 1974, the
Teaching and Instruction division held a benchmark conference on research
on the study of teaching. Researchers from across the country and from a
variety of disciplines and approaches met and participated in 10 panels to
discuss what was known about teaching and where research on teaching
might go. Each panel represented a different orientation to the study of
Green: Teaching as a Linguistic Process 169
teaching and had its own configuration of participants. Panel 5 focused on

the study of Teaching as a Linguistic Process in a Cultural Setting and
included researchers from linguistics, education, child development, and
anthropology. Each panel produced a written report that was and is
distributed by the National Institute of Education (1974).
In 1977, 3 years after the Panel 5 report was completed, NIE issued a
grants announcement (FY 1977) calling for research that focused on the
social aspects of classroom language in contrast to the language of the home
and playground, and that explored the relationship of teachers' and
students' language to achievement. This grants announcement suggested
that current advances in disciplines such as linguistics and anthropology
might be adapted and applied to educational problems to produce new
information and understanding of educational processes.
These actions on the part of NIE produced two outcomes, one direct and
one indirect. First, in 1978-79, 10 projects were funded by two NIE
programs. The majority of the grants were funded under the research
program for Teaching and Instruction. A smaller series of related projects
was funded by Reading and Language Studies through the unsolicited grants
cycle. These projects, therefore, form a core of work, and, as will be
discussed later, have been instrumental in developing the new constructs
and methodologies to be used to collect and analyze teaching-learning
behaviors in the natural educational settings of school, home, and
community.
The second aspect of NIE's role in the consolidation of the field is less
direct. The panel report and grants announcement, when combined with
initial dissemination efforts of these projects, have stimulated several
privately funded studies, which are direct spinoffs of the original studies
(e.g., Ford Foundation, Sloan Foundation, and Spencer Foundation), as
well as second- and third-generation studies funded by other agencies
(Bureau of Educationally Handicapped) and other divisions of NIE (Home,
Community, and Work). In addition, issues in research and general findings
have been discussed at more than 39 professional meetings across various
disciplines including anthropology, linguistics, psychology, child develop-
ment, sociology, and education. Within education, dissemination efforts
have also addressed a wide audience including researchers, practitioners,
teacher trainers concerned with language, reading, writing, bilingual
education, teacher preparation, technology, and accreditation. Therefore,
this body of research, while relatively new, has already developed a diverse
audience of individuals concerned with teaching and learning processes
which occur as part of everyday, face-to-face interactions in educational
settings of school, home, and community.
170 Review of Research in Education, 10
Between Education and Other Disciplines

The centrality of this work makes the findings important to both educators
and researchers in the fields that are the antecedents of this work. The
knowledge gained from these and related studies adds to our understanding
of educational processes and to our understanding of how the constructs and
methods adapted from disciplines such as linguistics, cognitive psychology,
and anthropology function in educational settings. Therefore, work in this
area informs both educational theory and theory in other academic
disciplines and helps further our knowledge about how people behave and
learn in natural settings. In other words, this work helps extend our
knowledge of how people learn to be more competent communicators and
learners, and about factors that serve to support and constrain such learning.
CONCEPTUAL BASIS OF THE FIELD

A set of constructs help define what is meant by the conceptualization of
"'teaching as a linguistic process." These constructs form a mental frame or
grid (Gumperz, 198 lb) for the discussion of findings and for discussion of the
c o m m o n elements of the core studies. The parts or pieces of information for
this section will be those constructs that represent an overlapping set of
related concepts, to the majority of studies. The whole is the emerging
nature of teaching-learning processes as linguistic processes and the nature
of communication in educational settings of school, home, and community.
Extracting Common Elements

While the projects vary in the literature and research basis they use to
frame the questions explored, they are all concerned with how people learn
from and use language to participate in face-to-face interactions. All the
studies use a variety of observation tools selected from a common set of tools
which are guided by principles of communication in face-to-face situations
(e.g., field notes, videotapes, audiotapes, case studies, and stimulated recall
interviews using records of observation). Each is concerned with capturing
human behavior that occurs during face-to-face interactions and exploring
the relationship of this learning to various aspects of cognitive and linguistic
development and school achievement. Therefore, identification of the
constructs underlying the various approaches helps establish a shared
framework with which to explore the work presented in this paper. Table II
presents a summary of the constructs extracted.
The constructs selected were stated explicitly in some of the studies and
implicitly in others. Implicit adherence to a construct was inferred from
examples of behavior sequences that were reported, from descriptions of
methodological tools used, from analysis procedures, and from the
discussion of findings and implications of the research. Constructs were
extracted from three sources: published papers, unpublished papers

(including unpublished manuscripts and midquarter and final reports), and
indepth interviews with principal investigators and research team members
of the larger, multiyear projects (Cole, Griffin, & Newman; Cook-Gum-
perz et al.; Erickson et al.; Morine-Dershimer & Tenenberg).
Three additional points must be considered. First, these constructs serve
multiple purposes. They are the underpinnings of the emerging field in that
they form a framework that guides observation, collection, and analysis;
they are also the products of the analysis. That is, the findings from these
studies further refine our knowledge of each construct, how it functions as a
guide in observing behavior in natural settings, and how it functions within
and across various classrooms and other educational settings. In other
words, these constructs form a nucleus of concepts that guide collection and
analysis of human behavior related to learning through communicating with
others.
The second point carries a note of caution. The definition of constructs
presented here is limited by space and the general goal of this paper. Each
construct could be the focus of a more extended discussion or a separate
treatment. However, for the purpose of this paper, a working definition for
each construct will be provided, adapted from sociolinguistics and
ethnography of communication. This procedure was necessitated by the
range of terms used in each project. When terminology was extracted and
compared, only 20 percent of the terms overlapped, given the range of
disciplines represented (e.g., anthropology, education, linguistics, psychol-
ogy, and sociology); therefore, to facilitate discussion, a common set of
descriptors for similar constructs was used. (More comprehensive treatment
of these and related constructs can be found in reports and papers of each
project, and in sources such as Cherry Wilkinson, 1982; Crystal, 1980;
Erickson, 1982a; Erickson & Shultz, 1981; Gilmore & Glatthorn, 1982;
Green & Wallat, 1981a; Gumperz & Hymes, 1972; Hymes, 1974, 1981b;
Kendon, Harris & Key, 1976; Mehan, 1979a.)
The third point is also a caution related to the summary of the constructs
presented in Table II. These constructs were extracted from a limited set of
each principal investigator's work. Therefore, they reflect only those
constructs that were readily identifiable; they do not represent the complete
theoretical history of each author.
Constructs that Define and Guide Teaching as a Linguistic

Process Research
The constructs have been grouped into six general categories to facilitate
discussion: (a) Face-to-Face Interaction: A Rule-governed Phenomenon,
(b) Contexts as Constructed, (c) Meaning as Context Specific, (d)
TABLE II .,~
Constructs Underlying Core NIE Studies: Teaching as a Linguistic Process ~o
~4
e J # ~_
Q
Constructs 13
~-._~ ~
FOCUS
- - l ~ ~ h e r - s t u d e n t interactions X X X X X X X 7
Student/peer interactions X X X X X X 6
a
~r
Contexts are constructed durin 9 interactions X X X X X X X X X X 10
Activities have participation structures X X X X X 5
Contextualization cues signal meaning X X X X X X X 7
Rules for participation are implicit X X X X X X X X 8 (1
Behavior expectations are constructed as part of interactions X X X X X X X X 8
Meanin1 is context specific O
All instances of a behavior are not equal X X X X X X X X X X 10
Meaning is signaled verbally and nonverbally X X X X X X X X X X 10
Contexts constrain meaning X X X X X X X X X X 10
Meaning is determined by and extracted from observed X X X X X X X X X X 10
sequences of behavior
Communicative competence is reflected in appropriate X X X X X X
behavior
Inferencin 9 is required for conversational comprehension X X X X X X X X X X 10
Frames of reference guide participation of individuals X X X X X X X X X X 10
Frame clashes result from differences in perception X X X X X X X 7
Communication is a rule-governed activity X X X X X X X X 8
Frames of reference are developed over time X X X X X X X 8
Form and function in speech used in conversations do not X X X X X X X X 8
always match
Classrooms are communicative environments
Differentiation of roles exist between teachers and X X X X X X X X X X 10
students; relationships are asymmetrical
Differential perceptions of events exist between teachers X X X X X X X 7
and students
Classrooms are differentiated communication environments X X X X X 5
Lessons are differentiated communicative environments X X X X X 5
Communicative participation affects student achievement X X X X X X X 7
Teachers orchestrate different levels of participation
Class X X X X X X 6
Group X X X X X X 6
Individual X X X X X X 6 )
Teachers evaluate student ability from observing X X X X X 5
performance during interactions
Demands for participation co-occur with academic demands X X X 3
.
Teachers signal their theory of pedagogy from their X X X X 4

behaviors (1
Teachers' goals can be inferred from behaviors X X X X 4
Note. The constructs here reflect those that were readily extracted from the studies. These constructs do not reflect the entire theoretical
orientation or history of the different authors.
p,
I=
)
Comprehension as an Inferencing Process, (e) The Classroom as a

Communicative Environment, and (4) The Teacher and the Communicative
Process.
Face-to-Face Interaction: A Rule-governed Phenomenon

Recent work in sociolinguistics, child language, ethnography of
communication, and ethnomethodology has demonstrated the rule-gov-
erned nature of face-to-face interaction. Rules of conversational participa-
tion and discourse construction have been shown to be culture specific and
learned from interacting with others. Some of the patterned or rule-go-
verned ways of interacting that have been identified to date include gaining
access to conversations (Garnica, 1981; Gumperz, 1977), taking turns
speaking (Sacks, Jefferson, & Schegloff, 1974), constructing narratives
(Tannen, 1980), interacting appropriately (Cherry Wilkinson & Calculator,
1982a; Hymes, 1974), demonstrating group membership and interactional
synchrony (Erickson & Shultz, 1982).
The conceptualization of face-to-face interaction as a rule-governed,
patterned p h e n o m e n o n is central to understanding the nature of teaching as
a linguistic process. Each of the projects reported in this synthesis adheres to
this conceptualization to some degree, which is demonstrated by the fact
that all are concerned with capturing how participants in educational
activities use language to achieve goals, to learn, and to participate in the
everyday activities of the classroom and other educational settings or in
natural experiments designed to focus on specific types of learning contexts
and learning phenomenon. The goal of this work is to understand the nature
of teaching-learning processes from the perspective of the participants and
to identify contextual factors that support or constrain participation,
cognitive and social knowledge gained, and/or communicative factors that
lead to evaluation of student ability. An observer, therefore, looks for
recurrent patterns of language use, observes how a message is received and
responded to, and attempts to identify typical cases of interaction within and
across various educational settings (Cook-Gumperz et al. ; Erickson, 1982b;
Erickson & Shultz, 1981; Florio & Shultz, 1979; Green & Harker, 1982).
The conceptualization of face-to-face interaction as rule governed does
not mean that participants follow fixed scripts, that conversations or
activities are invariant. Rule governed, in this instance, means that
expectations for performance exist that are culturally determined, and that
these expectations guide participation and act to constrain the options for
what will or can occur. For example, if someone asks a question, the listener
is expected to respond, or if someone greets you, you are expected to return
the greeting. While the expectation exists that a response will be given, in
neither instance can the form of the response or even the occurrence of a
response be predicted with certainty. For various reasons, an individual can

elect not to respond, can respond to the question with a question, or can
remain silent. These instances of unexpected behavior form contrast points
with what is expected and thus make the expected behavior more overt
(Cook-Gumperz et al., 1981; G r e e n & Wallat, 1979, 1981b; Mehan, 1979a;
Mehan et al., 1976). While contrast points make rules more evident,
appropriate behaviors also indicate rules; therefore, exploration of
appropriate behavior and chains of behavior also provides a basis for
identification of rules.
The potential variation in the flow of the conversation means that to
understand teaching as a process of face-to-face interaction, streams of
behaviors people use as they engage in face-to-face interaction must be
captured and analyzed. As the work presented later will show, awareness of
the discourse rules can help the observer make sense of interactions and
capture the nature of the process from the participants' perspective.
Contexts as Constructed
In addition to viewing conversations as constructed by the conversational
actions of participants, the researchers in the core group also view the
activity being observed as being constructed. In other words, the contexts of
the interaction are constructed by people as they engage in face-to-face
interaction. Contexts viewed from this perspective are not given in the
physical setting (e.g., rug area, reading circle) but are constructed by the
participants' actions as part of the interactions (Erickson & Shultz, 1981;
G r e e n & Wallat, 1979, 1981b). By observing how people hold each other
accountable to what is occurring and how they signal through verbal and
nonverbal actions what the activity is, the observer can begin to identify the
differentiated activities that make up the everyday life of a classroom.
A word of clarification is needed. While contexts are viewed as
constructed, some activities become ritualized and therefore more
constrained and more predictable. For example, a church service is an
extremely ritualistic context (Erickson, 1982b). In schools, some activities,
such as morning opening exercises, become rituals for some teachers. In a
ritualistic situation then, the demands for participation remain extremely
stable; variation in structure and types of language used is kept to a
minimum and the outcome is more predictable than spontaneous. Erickson
(1982b) argues that classroom activities and lessons are somewhere between
ritualistic and spontaneous on a continuum.
Three constructs are directly related to the view of contexts as constructed
entities. The first is the notion of contextualization cues (Cook-Gumperz &
Gumperz, 1976; Corsaro, 1981; Gumperz & Herasimchuk, 1973).
Contextualization cues are the verbal and nonverbal cues that people use to
transmit meaning. They include verbal aspects such as syntax and lexical
items and nonverbal aspects such as gesture, facial expression, proxemic

distance, and prosody (pitch stress, intonation, juncture). Observation of
these aspects of communication and behavior provides a range of cues about
intended meanings. Researchers in this area do not assume that they can
specify the exact cues used but that the range of cues from within a channel
and across channels of communication provide a degree of redundancy and
can be specified reliably. For example, as the teacher welcomes children to
the circle, she may also motion with her hand, turn her head from side to
side, and gaze at the students as she moves her head around the circle. All
these cues signal a welcome and provide redundancy of information. In the
core studies, seven of the projects use contextualization cues to aid in
collection and analysis.
The second construct is participation structure, which builds on the work
of Philips (1972, 1974) and Erickson and his students (Erickson & Mohatt,
1981; Erickson & Shultz, 1981; Florio, 1978; Florio & Shultz, 1979). As
indicated in Table II, five studies in the core group focus on this construct.
Participation structure refers to the demands for participation and the
varying rights and obligations that occur within and across activities. For
example, Florio and Shultz (1979) showed that a math lesson is not an
undifferentiated whole; rather, the lesson was divided into parts and each
part had a different set of expectations for performance. Similar findings
have been reported for reading (Cook-Gumperz et al., 1981; Green, 1977)
and for group discussion time (Green & Wallat, 1979, 1981b). Philips (1972,
1974) showed that the participation structures used by the Warm Springs
Indian children to learn at home differed from those required in school. The
clash betv, een the two types of structures affected students' achievement
and performances in school. The work of Erickson and Mohatt (1981)
supports this finding.
By extracting the participation structures of general events, researchers
using this construct can identify functionally equivalent activities in terms of
structure and demands for participation across a variety of settings;
ascertain if a single instance of an event is typical of the event or just a
random occurrence; and explore the degree of routine that exists or, rather,
the degree of stability of expectation in a recurring event. The existence of a
participation structure also suggests that expectations for behavior are
signaled as part of the delivery of the activity. In other words, as teachers
deliver instruction, the manner in which this information is delivered also
indicates what is required.
Erickson (1982b) suggests that two types of structures are signaled as part
of the evolving activity: the academic task structure (ATS) and social
participation structure (SPS). These two structures co-occur and can work
together or be problematic. Green and Harker (1982) have identified three
types of demands: academic, social, and contextual. They suggest that
failure to read the demands can lead to inappropriate performance during a
lesson. Michaels and Cook-Gumperz (1979) have shown that students'
linguistic performances within a participation structure are related to
achievement and to evaluation of student ability. Finally, Green (1977), in a
qualitative-quantitative study of effective teaching, found that the way in
which teachers used language across different participation structures was
correlated with student performance on comprehension of a story in the
lesson. These findings support early work by Bales and Strodtbeck (1967)
who found that teacher influence varied with the phase of the lesson. This
work, then, provides preliminary evidence of the relationship between
linguistic phenomena and student achievement.
The third construct, communicative competence (Hymes, 1974), refers to
an individual's ability to enter a situation and to engage in appropriate
behaviors. In other words, to know when to talk, to whom to talk, and how
to talk within a given situation. This ability, as Cook-Gumperz and Corsaro
(1976) have shown, is developmental. This construct underlies 8 of the 10
studies.
A caution is needed, howeverl The notion of competence is not used in
this work in a test sense. Competence in a sociolinguistic sense refers to an
idealized set of rules in the mind of the individual (Cherry Wilkinson &
Dollaghan, 1979; Hymes, 1974). Researchers using this construct assume
that only performance can be measured; therefore, competence can only be
inferred by observing an individual's performance within and across a
variety of contexts. It is not possible, therefore, to test for communicative
competence in a psychometric sense.
Meaning as Context Specific

The conceptualization of meaning as context-bound is closely related to
the concept of context as constructed. From this perspective, how a behavior
" m e a n s " is determined by considering how it is used, what precedes it, and
what follows. All instances of behavior are not considered functionally
equivalent. For example, the utterance " O k a y " has different meanings in
different contexts depending on how it is delivered. It can mean, " G o o d .
Well d o n e . " It can also mean " H o l d on. Stay with me." In the latter
instance, it serves as a place holder or device for maintaining engagement of
the students and the teacher's control of the lesson. It can also act to frame or
signal what comes next, as in "Okay, now we're going to go on to the next
item" (Green & Wallat, 1979, 1981b; Sinclair & Coulthard, 1975). The
meaning in each instance is signaled by the verbal and nonverbal aspects of
the delivery--whether there is a pause surrounding the item, the speed of
delivery, and the intonation pattern used. The position of the item alone
does not, in this instance, provide sufficient information about its meaning.
Further understanding of the nature of meaning as context-specific comes
from the consideration that no particular conversational behavior exists in a

vacuum. Each message in a conversation is surrounded by other messages.
The messages and actions of participants that precede a given message act to
constrain what can or will occur at any point in a conversation. The most
obvious example of this is the response to a question. The question places an
obligation for response on the person to whom it is directed. Therefore,
unless the listener or another participant does something unexpected, the
behavior that follows a question will be a response; however, as suggested
previously, the form and content of the response cannot be predicted.
Conversations, therefore, are constrained by the definitions people
construct for what is occurring, by the "local" history that unfolds during a
conversation, and by the rules of conversational participation and discourse
processes. However, conversations are not composed merely of strings of
questions and responses. Even in more ritualized settings such as lessons
where question-response sequences occur more frequently (Mehan, 1979a,
1979b; Sinclair & Coulthard, 1975), not all behaviors fit the question-re-
sponse pattern (Green & Harker, 1982). In addition, as Gumperz (1981a)
suggests, a message interpretation at one point in a conversation can change
the meaning a person has of what has preceded.
The small set of constructs presented above is only a beginning in the quest
for understanding teaching as a linguistic process. This work points to the
need for precision descriptions of language use and demonstrates the
potential of this approach for the study of teaching-learning processes as
linguistic processes. This approach provides precision information about
how behaviors are used within and across the different contexts of everyday
life in classrooms and the relationship of such use to student participation
and learning. In addition, the degree of precision that is obtained provides a
basis for identification of new variables, refinement of previously
determined variables, and provides for the exploration of the relationship
between language use and achievement (Au, 1980; Cole, Griffin, &
Newman, 1979, 1978-81; Erickson, 1982a; Green, 1977; Griffin, Cole, &
Newman, 1982; Michaels & Cook-Gumperz, 1979).
The constructs presented above provide general information about how
researchers in this emerging area view and approach research on the nature
of teaching as a linguistic process. In the next section, the question of how
participants in conversations "make sense" of what is occurring and the
nature of the resources used will be explored briefly.
Comprehension as an Inferencing Process

Two constructs have emerged from recent work exploring the intersection
between information processing and semantic analysis and comprehension
in conversational situations. All the studies in the core of work discussed in
this paper are concerned with what is learned from participating in
face-to-face interactions. However, the works of Cole, Newman, and

Griffin (Cole, Griffin, & Newman, 1979, 1978-81; Gearhart & Newman,
1980; Griffin, Cole, & Newman, 1982), Cooper (Cooper, Ayers-Lopez, &
Marquis, 1981; Cooper, Marquis, & Ayers-Lopez, 1980, 1982), Borman
(Borman et al; Borman & Barrett, 1981), and Hrybyk and Farnham-Dig-
gory focus on the cognitive aspects of the face-to-face process. In addition,
subcomponents of the Home/School Ethnography Project (Cook-Gumperz
et al.) focus on the cognitive knowledge of participants and its impact on
participation in and learning from face-to-face interaction (Collins &
Michaels, 1980; Michaels & Cook-Gumperz, 1979; Simons & Gumperz,
1981; Simons & Murphy, 1981).
When combined with work that merges information-processing/semantic
analysis procedures with procedures to capture evolving conversations
(Bloome, 1981; Bloome & Green, 1981; Frederiksen, 1981; Green &
Harker, 1982), a new direction for the study of cognitive processes in natural
educational settings of home, community, and school emerges. This
direction focuses on capturing cognitive processes that occur as part of the
everyday face-to-face interactions in naturally occurring activities.
In this section, two constructs that underlie the microanalytical
approaches used in the core studies will be discussed, which provide
systematic ways of obtaining precision descriptions at various levels of
unfolding cognitive, social, and linguistic processes that co-occur with
conversational processes. Work in this area, therefore, does not depend on
tests of cognitive ability to assess cognitive performance, but is concerned
with capturing naturally occurring cognitive processes (Cole, Hood, &
McDermott, 1978; Hrybyk & Farnham-Diggory).
Meaning is extracted from face-to-face behaviors. As discussed previously,
conversations are constructed by participants. They are not scripts to be
followed rotely. Participants in face-to-face situations, therefore, must
continually process information across various channels (verbal and
nonverbal) in order to participate appropriately in and acquire cognitive
knowledge from the evolving conversation. To accomplish this, participants
must use their communicative, social, and cognitive abilities to extract
meaning. By processing contextualization cues, a participant is able to infer
what is required. Meaning, therefore, is viewed as a process of extracting
verbal and nonverbal information so that a person can "make sense" of the
evolving events and gain access to the cognitive, social, procedural,
contextual, and communicative knowledge provided during these face-to-
face interactions.
The inferencing process is made complex by two characteristics of
conversations. First, different forms of messages can co-occur. For example,
a message can be aimed at an individual nonverbally, while the teacher is
presenting information to the group verbally.
Second, messages can serve multiple functions. That is, they can function
as controls on behavior simultaneously with providing content information.
For example, a teacher might say, "Now we're going to read from page 35 of
our texts." As the teacher delivers this message, she stresses the words
"Now w e ' r e . " At the same time she stresses these words in her message, she
speaks louder. She returns to normal level as soon as she gets the students'
attention. Control, therefore, was achieved indirectly, and the teacher
never had to break the flow of the content presentation.
Microanalysis procedures grounded in the constructs presented above
permit description of these co-occurring functions. These approaches
provide precision descriptions of the range of behaviors that occur in the
different communication channels. They also permit precise descriptions of
how behaviors build on other behaviors to accomplish the goals of each
aspect of a lesson or conversation. These descriptions have led to the
identification of variables that permit exploration of effective practices and
their relationship to cognitive performance (Au, 1980; Cook-Gumperz,
1981a; Cole, Griffin, & Newman, 1981; Green, 1977), the effects of
linguistic performance on the assessment of cognitive ability (Cook-Gum-
perz et al.; Griffin, Cole, & Newman, 1982; Hymes, 1981b), and student
perception of situation demands and language use to student performance
(Morine-Dershimer & Tenenberg; Hymes).
Frame of reference. One additional concept related to inferencing needs
to be considered: people in conversations bring resources and frames of
reference to the situation to help them interpret the conversation. A frame
of reference is developed for events and for general conversational
participation. Frames are established by extracting from face-to-face
situations the expectations for behavior. That is, from participating in a
variety of situations, a participant develops a frame of reference or set of
expectations for what should or might occur in similar situations. Therefore,
frames can be formed for local events and/or can come from past events.
Recent work by Frederiksen (1981) and Tannen (1979) suggests that as a
person interacts within a situation, the frame of reference he or she is using is
modified. Modification of the frame of reference comes from overt as well as
covert feedback. Covert feedback comes from observing the participation of
others. Work by Morine-Dershimer and Tenenberg shows that students
monitor not only teacher behaviors (e.g., questions) but student responses.
They report finding a pattern in students' report data which indicates that
students focus on the responses of other students so that they may learn what
the correct answer is or so that they can check their own knowledge. The
Morine-Dershimer and Tenenberg work further supports the view of the
participant as an active information processor during face-to-face inter-
actions; that is, the view that participants monitor a wide range of cues
within conversations to determine expectations, extract information, or
confirm these products of conversations. These cues are also open to

observers. Therefore, microanalysis of behaviors from the participants'
perspective and the focus on microanalysis of evolving behaviors permit
exploration of shifts in the frame of reference being used.
Observation of frames of reference have already been undertaken. This
past work suggests that frames for activities become stable over time
(Elkind, 1979; Wallat & Green, 1982) and that analysis of the language used
by teachers and students during interaction provides information about the
teacher's frame of reference and, therefore, helps make the teacher's theory
of pedagogy overt and available to observers and participants (Cook-Gum-
perz & Gumperz, 1982; Elliott, 1976; Green & Harker, 1982; Griffin, 1981;
Gumperz, Cook-Gumperz, & Michaels, 1981).
The concept of frame of reference also carries with it the notion of frame
clash, which can occur when two participants hold different frames of
reference for the same situation (Elkind, 1979; Heap, 1980). The frame
clash can be overt or covert. Overt clashes form observable contrast points
as previotlsly discussed (Cook-Gumperz et al.; Green & Wallat, 1979,
1981a; Hymes, 1981a, 1981b; Mehan et al., 1976; Mehan, 1979a).
Covert frame clashes can contribute to negative evaluation of student
ability (Griffin, Newman, & Cole, 1981; Michaels & Cook-Gumperz, 1979;
Scollon & Scollon, in press). Such frame clashes require a level of
microanalysis that is finer than do those instances of overt frame clash. In
addition, covert frame clashes require some form of retrospective analysis
that permits a systematic search of the permanent records. Often interviews
of participants are needed to provide both clarification and verification
(such interviews are used in 8 of the 10 projects). Exploration of covert
frame clashes has shown that these frame clashes can contribute to negative
evaluation of student ability (Griffin, Newman, & Cole, 1981; Michaels &
Cook-Gumperz, 1979; Scollon & Scollon, in press). In addition, because
clashes reflect a difference in perceptions of a phenomenon, the contrast can
be used as a starting point in exploring the unstated problem. For example,
Watkins (1981) reports a contrast in perceptions between parents and
teachers about homework. The frame clash occurs because parents are not
certain what is meant by the term "check" homework and are concerned
about signing for work that is not accurate. The teachers, on the other hand,
perceived the parents as not caring when the homework was not signed and
returned.
The Hymes project is important to our understanding of frame clashes on
a general level; that is, the frame clash that was uncovered by this project
occurred outside the immediate face-to-face interactions between parents
and teachers. The interaction occurred in the parents' homes and was the
result of written communication. This work suggests that researchers
interested in certain aspects of classroom processes must go beyond the

immediate context.
Work by researchers concerned with the linguistic match between home
and school has shown how knowledge about language use at home can
inform our understanding of what occurs in school (Au, 1980; Cazden, 1979;
Cazden, Carrasco, Guzman, & Erickson, 1980; Cook-Gumperz, 1981b;
Gilmore, 1981; Lussier, 1981; Michaels, 1981a, 1981b; Morine-Dershimer
& Tenenberg; Philips, 1972, 1974). In one notable instance, the
Kamehameha Early Education Program (KEEP), the information on the
types of interaction patterns in the community was used to modify the way
teacher's delivered reading instruction. This change came after careful
ethnographic study of what was occurring in school and home settings and
after careful curriculum planning and experimentation. This work showed
that curriculum changes based on ethnographic findings could lead to
increased achievement. What is important to note in this work is that time on
task was high before and after the change. This work shows that changing the
task, not increasing the amount of time, might be a crucial factor. This
project also demonstrates that findings from ethnographic work can be used
as the basis for training teachers and improving instruction (Au, 1980).
The constructs discussed to this point have demonstrated the nature of
teacher-learning processes as linguistic processes. The last two constructs to
be presented focus on the general nature of the classroom as a
communicative environment and the teacher's role in orchestrating this
environment.
T h e C l a s s r o o m as a C o m m u n i c a t i v e E n v i r o n m e n t
This section explores the structural aspects of classrooms that help

researcher and teacher understand the nature of communicative demands
placed on both teachers and students by the diversity of classroom lessons
and structures. The work on participation structures has demonstrated that
the classroom is a differentiated communication environment with shifting
requirements and obligations for participation (Cherry Wilkinson, 1981;
Eder, 1980; Erickson, 1982b; Erickson et al.; Erickson & Mohatt, 1978;
Florio, 1978; Florio & Shultz, 1979; Green & Harker, 1982; Gumperz, 1977,
1981; Philips, 1972, 1974; Wallat & Green, 1979, 1982). This work has also
shown that lessons within classrooms are differentiated communication
environments (Bales & Strodtbeck, 1967; Bellack, Kliebard, Hyman, &
Smith, 1966; Collins & Michaels, 1980; Erickson et al., 1978; Fiorio &
Shultz, 1979; Green, 1977; Gumperz & Cook-Gumperz, 1981a; Guzman,
1980; Hymes; Michaels & Cook-Gumperz, 1979).
The differentiation that occurs across lessons and within lessons means
that special demands are placed on students and teachers with regard to both
cognitive and communicative participation. In addition to shifts in demands
for general participation, recent research has shown that teachers

differentiate instruction within lessons for individuals (Guzman, 1980),
differentiate feedback about rules for participation (Stoffan-Roth, 1981a),
and differentiate instruction or expectation for performance by group
(Cherry Wilkinson, 1981; Collins, 1981; Collins & Michaels, 1980; Griffin,
1981; McDermott, 1976). Also, the ways in which the teacher distributes
language use across the phases of lessons has been shown to affect student
performance (Green, 1977).
This work has implications for future research and for understanding
teaching, and the differentiated nature of classroom environments has
implications for sampling. Researchers who want to compare demands and
performance of students in different activities need to be concerned about
the functional equivalence of the activities in terms of demands for
participation; they also need to consider how many events to sample and
how generalizable their findings are across various content areas. Teachers
must become more overtly aware of the differing demands for participation
that are required across the various lessons. The value of using constructs
from the study of teaching as a linguistic process have been identified by
various teachers in current and past work (Carrasco, Acosta, & De La
Torre-Spencer, 1981; Florio & Walsh, 1976; Hymes; Stoffan-Roth, 1981b;
Wallat, Green, Conlin, & Haramis, 1981).
The Teacher and the Communicative Process

The constructs in the previous sections converge to form the background
for this section on the teacher's role in orchestrating and managing the
differentiated contexts that make up daily life in classrooms. The constructs
will not be restated here; instead this section will focus on communication
from the teacher's perspective and role as instructor. Four aspects of this
process will be discussed: (a) teachers orchestrate levels of participation
(class, group, and individual); (b) teachers evaluate student ability from
interactions; (c) teachers signal their theory of pedagogy by their behaviors
and expectations; and (d) both researchers and participants can infer
teachers' goals from their behaviors.
Teacher as orchestrator. The teacher is responsible for orchestrating the
actions and events of the classroom. Brophy and Evertson (1976) suggest
that orchestration is not merely using a set of strategies, but it is using
strategies to achieve a variety of goals in a variety of ways. Teaching,
therefore, is a creative process; it is a process of creating environments, of
creating activities, of creating situations with children so that children can
master the academic and social content of schooling. The vehicle for this
creative process is communication--communication between teacher and
students, among students, and between students and other adults.
Teaching, therefore, is a linguistic process. As such, classroom communica-
tion is subject to the rules and expectations of conversation. Classroom

events, like other communicative events, are constructed by participants as
they engage in face-to-face interactions. The teacher in this process plays an
asymmetrical role, since he or she is ultimately responsible for what occurs.
While most work on classroom language use has focused on what occurs
during lessons, several studies have begun to explore how teachers move
students from one place to another and manage interruptions and appeals
for help (Merritt, 1982; Merritt & Humphrey, 1979, 1981), how teachers
establish norms for behavior (Stoffan-Roth, 1981a; Wallat & Green, 1979,
1982), and how peer groups work together. In this latter area, three aspects
have been explored: playground games and peer groups (Borman, 1979;
Borman et al.), peer groups in instructional situations (Cherry Wilkinson &
Calculator, 1982a; Cherry Wilkinson & Dollaghan, 1979; Cole, Griffin, &
Newman, 1979; Cook-Gumperz et al.; Cooper, Ayers-Lopez, & Marquis,
1981; Cooper, Marquis, & Ayers-Lopez, 1981, 1982; Hymes; Steinberg &
Cazden, 1979), and peer group formation and structure and school status
and performance (Garnica, 1981; Hyrbyk & Farnham-Diggory, 1981).
This work suggests that teachers not only orchestrate what occurs with the
lesson of the moment, but also are responsible for orchestrating students in
simultaneously occurring groups and in peer situations; that is, the teacher
must monitor and orchestrate the lesson that is the primary vector of activity
(Merritt & Humphrey) as well as those that compose secondary vectors of
activity. As Merritt (Merritt, 1982; Merritt & Humphrey, 1979, 1981) has
shown, the ways teachers manage the flow between vectors of activities is
rule governed or patterned. Cahir (1978) also reports similar patterns of
orchestration of activity during transitions between events.
Within a lesson, the teacher not only presents academic content but must
also orchestrate the structure of the activities, distribute turns to speak, and
maintain order and flow of the activity. This work suggests that teachers
must simultaneously orchestrate academic content, management, and
discipline aspects of lessons. The task of instruction, therefore, involves
more than planning the sequence of presentation of academic content.
Griffin (1981) suggests that teachers must also orchestrate delivery
strategies of varying types. For example, if students ask for help, teachers
often need to ask for additional information to provide the appropriate help;
that is, teachers need to develop ways to get some information in order to
give some information. This notion demonstrates the mutual interaction
that must be orchestrated.
Teacher as evaluator. One aspect of orchestration frequently overlooked
is the idea of teacher as evaluator of performance during interactions.
Evaluation in this instance does not refer to standardized or teacher-made
tests, but to the ongoing monitoring of student performance that occurs
during lessons. This evaluation can be deliberate and overt, as in the
initiate-respond-evaluate sequence (Mehan, 1979a; Mehan et al., 1976), or

as frequently occurs, covert (Cahir & Kovacs, 1981). Work on linguistic
frames and frame clashes (e.g., Cook-Gumperz et al., Scollon & Scollon, in
press) demonstrates that teachers often evaluate student performance in
face-to-face situations. For example, Michaels and Cook-Gumperz (1979)
found that during sharing time, the teacher made judgments about the
students' story-telling ability. Those students who "stayed on topic" by the
teacher's definition (i.e., were topic-centered) were evaluated more
positively than those who used a topic-chaining style of narrative.
Microanalysis of the children's storytelling ability in more experimental
contexts showed that the children who used the topic-chaining style were
adhering to a style that is culture specific. The frame clash between the
expected style and the style used by some students led to negative
assessment of their ability to use "appropriate" language in sharing time.
This evaluation by the teacher did not lead to instruction in appropriate
language. Scollon and Scollon (in press) report similar findings related to
reading performance.
Cole, Griffin, and Newman, 1981, suggest that teachers must continually
make judgments about student performance as they are teaching, but the
information-processing demands often make this task difficult. They
describe one case during an experimental tutorial in which the tutor
misinterpreted a child's action. In attempting to "correct" the student's
"error," the tutor apparently confused the student. The result was that the
student, when later working on the task alone, appeared to know less than
he did before the tutorial and when working cooperatively with the tutor.
This work suggests that a teacher's actions, based on online assessment, may
inadvertently have a negative influence on a student's performance.
This work also suggests that teachers must become more aware of the
basis of their evaluations. It indicates that teachers may be using linguistic
performance in making their judgments. A warning in terms of evaluation is
necessary. Work by Carrasco (1981) showed the potential problems that
come from teachers' evaluations made from observations during teacher-
child interactions. He reports a case where a student was about to be
retained in kindergarten because of lack of appropriate performance during
group activities directed by the teacher. The child was accidentally captured
on videotape working in a peer situation. The child was extremely
competent in this situation and actually took a leadership role. The peer
situation tape made the teacher reassess her judgment and the child was not
retained. This data suggests that single-context sampling for assessment may
not provide a complete picture of student ability.
Observing teacher pedgogy and teacher goals. The last two constructs in
this section will be considered together. By observing teachers' actions as
they work with and build on their own behaviors and the behaviors and
actions of students, researchers and students can extract the expectations for
behavior. These expectations relate closely to teacher perceptions of the
situation and the expectations teachers hold for performance. Such
expectations are indicative of both the teacher's goals for the situation and of
the teacher's theory of pedagogy (Cook-Gumperz, Gumperz, & Michaels,
1981; Griffin, 1981; Hymes, 1981b). Griffin (1981) suggests that the
differences in approaches to high, middle, and low students may be due to
differing theories about what each group of students needs. Work on math
instruction within the Cole, Griffin, and Newman (Petitto, 1982) project
showed that even when the teacher used the same plan for instruction in long
division, she treated the groups differentially and when queried about the
shifts could provide a rationale. This suggests that rather than a single theory
of pedagogy, teachers may have different theories for different students
and, therefore, shift instruction according to these theories.
Summary. In the section above, a series of constructs underlying the
central core of studies and the field of teaching as a linguistic process were
presented. While these constructs are not a complete set of those involved in
the conceptualization of teaching as a linguistic process, they help define the
difference between this field and previous approaches to the study of
teaching. The discussion of the methodologies emerging from this work and
the section on the findings will presuppose knowledge of this framework. It
is not the intent of the following sections to reiterate the discussion above;
however, where appropriate, the discussion will extend and refine
knowledge of these constructs.
METHODOLOGY: A PRODUCT OF THE RESEARCH

The general approach to the study of teaching-learning processes as
linguistic processes used in these studies centers on the exploration of
naturally occurring behaviors that people use as they engage in everyday
teaching-learning activities at school, in the home, and in the community.
The methodology used by the majority of these studies is an adaptation of
ethnography of communication generally used in anthropology. The
methods were not transported from anthropology to education blindly; the
researchers were sensitive to special demands of classrooms, schools,
homes, and communities, which has led to modifications of these
approaches to meet the special conditions and demands of educational
settings and issues. For discussion of the general approach see Spradley
(1980a, 1980b), Spindler (1981), and Agar (1980). For discussion of the
issues involved in using ethnographic methods and doing ethnography in
educational settings, see Cherry Wilkinson (1982), Erickson (1977, 1979,
1982b), Gumperz and Cook-Gumperz (1981a), Hymes (1981b), Sevigny
(1981), Lutz (1981), and Spindler (1981).
This section illustrates the central issues in doing naturalistic research
from the perspective of the core studies and provides representative models
of various aspects involved in using these approaches.
Capturing Complex Processes: The General Approach

As indicated previously, each study has a unique aspect in its literature
base and also a degree of overlap with the other studies. This section
explores the ways in which the differences in theoretical base influenced the
focus, collection procedures, analysis, and interpretation procedures used in
the core studies.
In exploring human behavior in natural settings, most of the projects used
ethnographic methods or techniques adapted from ethnography. These
methods are guided or driven by the mental gird or framework that is
derived from the theoretical basis of the study. Methods in these studies,
therefore, are theoretically grounded. This conceptualization of methodolo-
gy will become clearer as the different aspects of the methods are considered
for the 10 projects. Three aspects of the methods will be considered: the
general approach and analysis focus, general structural components of the
research designs, and ethnographic tools for data collection.
The general approach and analysis focus. The exploration of methodology
begins with a summary of the general approach and focus used in the core
studies. Five general approaches were used: single case studies (i.e., focusing
on a single school or classroom [Cole, Griffin, & Newman; DeStefano &
Pepinsky]); double case studies (i.e., using contrastive settings or. grades
[Cook-Gumperz et al.; Erickson et al.]); multiple case studies (Hymes);
developmental studies (Borman et al.; Cooper, Ayers-Lopez, & Marquis;
H r y b y k & F a r n h a m - D i g g o r y ) ; and cross-age sampling (Merritt &
Humphrey; Morine-Dershimer & Tenenberg). Although these are general
characterizations of this body of work, each of these general approaches is a
complex entity. That is, within the single case studies, a wide variety of issues
were explored and methodologies used. This characteristic, therefore, is of
limited help in characterizing the 10 studies.
Task focus. Task focus refers to the primary basis of the literature framing
the studies or the specific task explored and the types of analysis procedures
used. The complexity of these studies becomes clearer when the task focus is
considered. The 10 studies can be characterized by the nature of the tasks on
which they focused. Various types of general task focus were considered:
linguistic processes, cognitive processes, the relationship between language
and cognition, the construction of cognitive tasks through language use, or a
combination of these. While all the studies considered language in some
form and most considered a cognitive dimension, the studies differed in the
emphasis placed on these processes. Therefore a project might be
characterized by a primary focus and also have components requiring a shift
in focus. In the following section, these shifts are indicated by the placement
of a project under more than one area of task focus. In addition, these
projects contained various substudies or a range of questions.
For example, five projects adopted a general linguistic focus (Cook-
Gumperz et al.; DeStefano & Pepinsky; Erickson et al.; Merritt &
Humphrey; Morine-Dershimer & Tenenberg). Within this general task
focus, however, further differentiation occurred. Three studies had
substudies that focused on language and reading (Cook-Gumperz et al.;
DeStefano & Pepinsky; Erickson et al.). Four had components that focused
on language of home and school (Cook-Gumperz et al.; Erickson et al.;
Hymes; Morine-Dershimer & Tenenberg). Eight had substudies that
focused on language and instructional participation (Cole, Griffin, &
Newman; Cook-Gumperz et al.; Cooper, Ayers-Lopez, & Marquis;
DeStefano & Pepinsky; Erickson et al.; Hymes; Merritt & Humphrey;
Morine-Dershimer & Tenenberg). Three had components that focused on
language and written literacy (Cook-Gumperz et al. ; DeStefano & Pepinsky;
Hymes). One focused on language and math (Cole, Griffin, & Newman).
And three focused on language and the language arts (DeStefano &
Pepinsky; Erickson et al.; Morine-Dershimer & Tenenberg).
In addition to a linguistic focus, three studies adopted a general focus on
cognitive processes and social processes (Borman et al. ; Cole, Griffin, &
Newman; Hrybyk & Farnham-Diggory). Within this focus, further
differentiation occurred. For example, two studies focused on issues in
social cognition (Borman et al. ; Hrybyk & Farnham-Diggory); three onpeer
network development (Borman et al.; Cook-Gumperz et al.; Hrybyk &
Farnham-Diggory); and two on playgrounds and children's life on
playgrounds (Borman et al.; Hrybyk & Farnham-Diggory).
Four projects adopted a focus on the relationship between language and
cognition (Cook-Gumperz et al.; Cooper, Ayers-Lopez, & Marquis;
DeStefano & Pepinsky; Erickson et al.). Within this area, three studies had
components that focused on participation structures and academic task
structures (Cook-Gumperz et al. ; DeStefano & Pepinsky; Erickson et al.).
Two had substudies that focused on task construction through language
(Borman et al.; Cole, Griffin, & Newman). Six had components that
focused onpeer teaching (Cole, Griffin, & Newman; Cook-Gumperz et al.;
Cooper, Ayers-Lopez, & Marquis; DeStefano & Pepinsky; Erickson et al. ;
Hymes). Four focused on task occurrence in and out of school (Cole, Griffin,
& Newman; Cook-Gumperz et al.; Hymes; Morine-Dershimer &
Tenenberg), and four focused on task occurrence in different participation
structures (Cole, Griffin, & Newman; Cook-Gumperz et al.; Merritt &
Humphrey; Morine-Dershimer & Tenenberg).
As indicated above, no single characterization of these projects is
possible. Projects may have a primary focus on linguistic processes and still
have substudies or components that focus on cognitive processes or the
relationships between language and cognitive processes. This work,

therefore, is a rich source of information about a broad range of phenomena
and processes found in everyday events of educational settings.
These projects can also be characterized in terms of general methodologi-
cal approach; for example, the use of natural observation and natural
experiments within a project (Cole, Griffin, & Newman; Cook-Gumperz et
al.; Cooper, Ayers-Lopez, & Marquis; Hymes), the use of a triangulation
approach (Morine-Dershimer & Tenenberg), and the use of qualitative and
quantitative measures (Cook-Gumperz et al.; Cooper, Ayers-Lopez, &
Marquis; DeStefano & Pepinsky; Erickson et al.; Morine-Dershimer &
Tenenberg). These last two characteristics show the comprehensiveness of
these 10 projects. The wealth of data suggests that further analyses are
possible and that secondary analysis of these data banks, which is grounded
in the original goals and frame of reference, will be profitable (cf. Merritt &
Humphrey). (Information about potential areas for secondary analysis is
presented in the second section on Future Directions: Research.) While all
the studies considered language in some form and most also considered a
cognitive dimension, the studies differed in the emphasis placed on these
processes.
The general analysis frame. The last type of information presented is
related to the use of both natural and experimental methods and qualitative
and quantitative methodology. The majority of studies (7 of the 10) used one
or more of these procedures in the analysis and design stages. These projects
can be described as broad based. These are not eclectic studies; rather, the
multiple loci are guided by specific questions and specific theoretical
frameworks. The cohesion comes from the fact that each focuses on multiple
levels of interaction in and across a variety of contexts; the variety of types of
information collected permits the researcher to explore a construct in
different contexts and under different conditions. In addition, the various
types of data often provide convergent sets of information on and validation
of findings.
To demonstrate the cohesion, the concept of tracer unit was borrowed
from the work of Cole, Griffin, and Newman (1981; Griffin, Cole, &
Newman, 1982). A tracer unit is a type of behavior, type of information, or
type of construct that is traced or followed across and within various settings
and/or contexts. This unit becomes a primary "locus of observation"
(Merritt & Humphrey, 1981) within a given study. Any given study may
have multiple tracer units. Each unit involves shifting the locus of
observation to foreground and background processes, variables, contexts,
and so forth.
For example, Borman and her colleagues (Borman & Barrett, 1981;
Borman et al.) observed instances of negotiation of rules for participation in
games, formally organized and spontaneously organized games on
playgrounds, children's interactions on playgrounds, and children's social

cognitive skills. By tracing each of these phenomena, Borman and her
colleagues were able to explore how boys and girls negotiate playground
games, the ecology of children's play, children's interpersonal relationships,
social-cognitive skills, and the relationships between and among these
phenomena.
Cooper and her colleagues (Cooper, Ayers-Lopez, & Marquis) also
focused on peer situations. Their locus of observation, however, differed
from that of the Borman team. The students were observed in a didactic
teaching situation and a cooperative learning situation under quasi-experi-
mental conditions, either teaching or learning in a peer learning situation.
These researchers focused on interactions that occurred in different contexts
(e.g., cooperative learning situations, classroom spontaneous interactions,
didactic teaching situations). By focusing on patterns of interaction in these
contexts and identifying those that lead to effective learning, these
researchers were able to describe developmental patterns of interaction in
effective learning, characteristics or developmental patterns of children's
spontaneous interactions, and the consequences of such interactions. In
other words, patterns of interaction were traced within and across contexts
and instructional situations to provide a broad picture of students'
competence as tutors, resources for others, and collaborators in learning.
Other tracer units included discourse behaviors, academic content,
perceptions of behavior, and specific teacher strategies (e.g., praise,
personalization of instruction). In addition to a focus on a type of behavior,
tracer units were considered at different levels; for example, on
microbehaviors such as a teacher's use of praise and on more macrolevels
such as peer networks.
Table I provides further information about the types of units,
information, and contexts traced, and the loci of observation used within
and across studies. In the next section, further clarification of the projects,
their approaches, complexity, and methodologies will be presented.
General Structural Components of the Research Designs

Each of the 10 projects are multifaceted. Each explores a variety of
related and interrelated phenomena to obtain a broad picture of the nature
of the phenomena studied (e.g., discourse rules at home and in informal and
formal school settings). The complex designs also permit exploration of the
factors that constrain or support phenomena in different contexts and the
relationship of the phenomena to other aspects of teachingqearning
processes, achievement, or learning measures. The knowledge gained from
these studies provides a broad picture of the phenomena across various
activities and settings. This information also helps refine our understanding
of not only the phenomena in isolation (as a bit of information) but our
understanding of the phenomena in the larger pattern (Erickson, 1981).
Because multiple levels of context were studied simultaneously (e.g.,
individual teacher-child interaction within lessons, across lessons, within the
general classroom, at home), it is possible to explore or interpret the bits in
terms of how they fit into ever-increasing contexts. This approach,
therefore, can be considered an exploration of part-whole relationships.
One outcome of this approach is that some researchers have found it
possible to predict the types of phenomena that will occur (Gumperz &
Cook-Gumperz, 1981b); what cannot be predicted is the specific event
structure or the specific types of language used.
While each study is complex and unique in design, the group as a whole
can be viewed in terms of three types of design components: general
participant observation, topic-centered observation, and natural experi-
ments~experimental tasks. These three components are considered general
approach structures and can be used to contrast the 10 studies. Table I
provides an overview of each study, including the specific focus of various
aspects of the study.
General participant observation refers to participant observations of
phenomena on a general level in broad terms, which permits identification
of where, when, and how the phenomena occur, as well as general structural
aspects. Topic-centered observation refers to the focused observation of
phenomena that is usually undertaken after general participant observation
or along with a natural experiment or experimental task (cf. Cole, Griffin, &
Newman, 1981). However, topic-centered observation can become the
primary focus when the researchers use past research about the phenomena
to frame the observations. This latter aspect does not mean that those who
begin with general participant observation ignore past work, but that some
researchers elect to begin with the topic focus. The last category, natural
experiments/experimental tasks, refers to more constrained tasks designed
by the researchers to explore systematically a given phenomenon or set of
phenomena or to manipulate the occurrence of the phenomena in different
contexts.
As indicated in Table I, all but two of the studies used two of the three
categories (Cole, Griffin, & Newman; Cook-Gumperz et al.). However, as
indicated, not all used the same two categories in creating their design for the
project. Some studies followed the same phenomena from general
participant observation to more refined topic-centered observations
(Erickson et al.; Hymes; Merritt & Humphrey). Others used topic-centered
observation and natural experiments (Borman et al.; Cooper, Ayers-Lopez,
& Marquis; DeStefano & Pepinsky; Hrybyk & Farnham-Diggory;
Morine-Dershimer & Tenenberg). Two studies (Cole, Griffin, & Newman;
Cook-Gumperz et al.) utilized all three categories in planning and

implementing their study.
Detailed presentation of each study is not possible given the general
purpose of this synthesis and the restricted space; therefore, the
Cook-Gumperz et al. study has been selected as indicative of the general
approach because it involves all three aspects in a design which moves from
broad, general observation to more and more focused observations. Figure
1 provides a detailed diagram of this study. The model was extracted from
published papers, the final report to NIE (Cook-Gumperz et al.), and
interview data.
Like the Cook-Gumperz et al. project, each project begins with a mental
grid; each grid, however, is different. The difference stems from the
research and theoretical base of the study. In addition, each engages in some
form of observation. Like the Cook-Gumperz et al. study, the observations
are not random; rather, they are grounded in and driven by the conceptual
framework of the study. This grounding is reflected in the way in which field
notes are generally taken. Field notes generally include: (a) descriptive
comments, (b) methodological comments, and (c) theoretical comments
(Corsaro, 1981; Erickson et al. ; Sevigny, 1981; Spradley, 1980a, 1980b). In
addition, some participant observers include personal comments separate
from theoretical and methodological comments (Corsaro, 1981).
Observation, therefore, is a principled phenomenon. It occurs in a
systematic way. Inquiry into naturally occurring phenomena evolves within
this group of projects. Initial questions become refined as the observers
explore them in specific activities and contexts. The phenomena become
refined as does the researcher's understanding of when, where, and how
they occur within and across a variety of activities and contexts.
As indicated in Table I and Figure 1, the refinement can occur by moving
from general participant observation to topic-centered observation (e.g.,
Erickson et al. ; Hymes) or by planning contrasts (e.g., the cycles used by
Cole, Griffin, & Newman) and moving from general observation to natural
experiments (e.g., curriculum experiments) to topic-centered observations
and back to natural experiments in a cyclical manner.
An analogy for these approaches might be an objective microscope. 2 An
objective microscope has three lenses on it, each with a different power. At
the lowest power, the field on the slide is clear. As the observer changes
lenses, the power is increased; the field begins to blur and the details come
into focus. At middle power, however, a portion of the field does remain. At
highest power, though, the details can be seen and the field disappears. By
shifting lenses, the researcher is able to move backward and forward across
settings, bits of information, and in some instances, phenomena, exploring
the part-whole/whole-part relationship between bits and larger patterns.
Each of these studies uses this approach to a greater or lesser degree.
Ethnographic tools. Each of the studies in the core group used some form
of systematic formal recording. In this section, the tools used to collect the
data are specified and select issues related to the use of the tools are
discussed. Table III presents an overview of 16 data collection procedures
which were used in the 10 studies. As indicated, all studies used
multiple-collection techniques; however, which ones were used depended
on the questions being asked. Of concern in all the studies were questions
about what counts as a task, what level should be observed, and what counts
as knowledge. None of the studies answers these questions; rather, new
definitions or understandings of these issues can be seen as a product of the
different studies (e.g., the definition of formal and informal writing and
literacy was explored in depth within the Hymes project: Gilmore (1981),
Woods-Elliott (1981), and Fiering (1981). The multiple collection
procedures also provide ways of moving between specific foci and broader
streams of behaviors; the use of multiple collection procedures also provides
a source of convergence of information.
One of the common criticisms leveled against this type of work is that an
inordinate amount of material is collected. The material collected, as
discussed above, is used for a variety of substudies and as grounding for the
analysis and interpretation of data; therefore, this criticism in not valid. The
data form a wide-angle lens for future work (cf. Merritt & Humphrey, 1981).
One product of this work is systematic indexing. Indexing systems by
Erickson and his colleagues and Griffin (Merritt & Humphrey) are
presented in Table IV. These systems demonstrate systematic ways of
cataloging complex data to facilitate retrieval and analysis.
The Griffin system is of special note. The Merritt project involved a
secondary analysis of data collected in a Carnegie Foundation study granted
to the Center for Applied Linguistics (see Merritt & Humphrey for details).
Merritt, with the aid of the indexing system designed by Griffin (Merritt &
Humphrey), was able to review the corpus, set criteria for selecting events
that serve as a "locus of observation," and identify occurrences of
service-like events across the corpus.
The work of Erickson and his colleagues and Griffin and Merritt
demonstrates the viability of secondary analysis of data from these studies.
A word of caution is needed. The secondary analysis that Merritt undertook
was carefully grounded in the theoretical orientation which guided the
original collection as well as in the larger patterns in the corpus. In addition,
two researchers from the original project participated in the secondary
analysis. Secondary analysis that does not provide grounding, both
theoretically and historically, in the original data might be troublesome.
Morine-Dershimer and Tenenberg's work shows the potential problems.
This research team deliberately subjected the same data to researchers with
different analysis frames. While all were concerned with language, each
FIGURE 1. Design of Study: School~Home
STAGE I: Framing and Planning the Project Components
Mental Frame-Mental Grid Participant Observation

Assumptions guiding/driv-
ing the collection, analysis, Single observer assigned to
and the interpretation of each classroom. Observer
data assumes three roles:
1. participant observer: par-
Assumptions are derived ticipates in events and ob-
from theoretical and research serves during participa-
literature which includes work tion-records information
on: after event
discourse processes 2. observer participant: pri-
conversational analysis marily observes--partici-
ethnography of communi- pates only if approached
cation by students for help
classroom organization 3, aide: acts as aid for
teaching-learning pro- teacher, helps students
cesses Each role provides a different
adult-child interactions view of events. This ap-
child language proach allows observers to
cross-cultural communica- assume an insider's view at
tion times and an outsider's view
evaluation of performance at other times. Ongoing in-
socialization volvement provides time for
informal interviewing, captur-
ing developmental aspect of
events, and establishing a
shared perspective of events
with teacher and students
Ethnography Project (Cook-Gumperz et aL)
STAGE I1: Data Collection-General Participant Observation
Participant Observation I: Explicit Definition of Behav-

Teacher Planning-organiza- iors
tion
Participant observer (PO) Teacher definition of actions
works with teacher before available through question
school year begins, before and answer sequences be-
class, at breaks, and after tween PO and teacher in
school as aide. PO is not PO's role of aide
trained teacher, is naive, and
can ask questions on a "real
need to know" basis; this talk
is "talk for doing" job as Implicit Definition of Behav-
aide. Approach is used to iors
observe:
t. teacher plans and plan- Inferred from practice and
ning behaviors from directions given to PO
2. teacher organizational be- as aide. Can be made expli-
haviors and practices citly during informal interview-
3. teacher theory of peda- ing as part of aide's role
gogy
Participant Observation I1:

Classroom processes~prac-
tices Implicit Definition of Behav-
iors
Observer participant of total
day to ascertain: Inferred from sequences of
1. formal segments of day behavior, from actions of
2. ways in which teacher teacher and students working
frames activities/events
3. orchestration of events ~ with each others' behaviors
and observation of contex-
Participant observation of tualization cues.
events to help students and
to ascertain:
1. segments of events
2. conflict/contrast points
3. expectations for behavior
Basis of segmentation:
contextualization cues and
participation structures, ob-
servable behaviors.

F I G U R E 1 (Con'd)
STAGE II1: Topic-centered Participant Observation
Participant Observation II1: Participant Observation IV:

Observe for contrastive Observe and videotape se-
situations~signs of discrep- lect events
ancy Observe participation struc-
1. Look for instances of dif- tures and obtain rights and
ferential learning obligations for events
2. Look for events of day
that reflect problem. Best Observe verbal signals and
site is event with high inci- conventions or f o r m u l a i c / r i t - ~
dence of problems (e.g., ualistic uses of language
miscommunication).
3. Look for atypical happen- Observe target individuals
ings within typical events selected during earlier
4. Begin to predict type of stages and who permit ob-
event that will occur, not servation of contrasfive be-
specific event haviors
Green: Teaching as a Linguistic Process lg7
STAGE IV: Inferencing/Hypothesis STAGE V: Natural Experiments

Generating
Observe discourse strategies Plan and execute natural ex-

used within the context periments that permit con-
trast of observed phenomena
Observe indications of evalu- with similar phenomena in
ation of student perfor- controlled or contrastive set-
mance (verbal and nonver- tings
bal)
Science laboratory:
Observe students and iden- Lawrence Hall of Science
tify differential performance Explore whether the dif-
and treatment within and ference in participation
across settings setting and structure
produce differences in
Observe recurrent events performance
and begin to predict occur-
rence of types of events Pear stories (Chafe)
Explore narrative product-
ion (oral & written) in
control situation with nat-
urally occurring narratives
in classroom (e.g., shar-
ing time)
Storytelling task
Further explore students'
narrative abilities and dif-
ferences in narrative style
among groups of stu-
dents of different lan-
guage traditions
Referential communica-
tion taskJphonemic per-
ception task
Explore students' ability
to use decontextualized
language and contrast to
reading-skill performance
V. Home data collection

Collect data on narrative
events in home. Work
with parents to select
events, have parents
tape events (no PO),
suggest event.
TABLE III ~)
Ethnographic Tools Used for Data Collection m
~ .-~=~
=E
o
=
~r
Videotape records X X X X X X 7
Audiotape records X X X X 4 c
Field notes X X X X X X X X 9
Review notes of videotapes X 2
o
Formal indexing of tapes X X X X X 6 .=
Collaborative planning with teacher of activities 2
to be taped
Participant observation X X X X X X X X X X 10
Participant by researcher in classroom as aide X X X 3
Elicitation tasks X X X X 4
Interviews of teachers X X X X X X X X 8
Interviews of students X X X X X X X 7
Diaries X 1
Stimulated recall interview using ethnographic record X X X X X 5
teachers
Stimulated recall interview using ethnographic record X X X 3
students
Tests X X X X 4
Cognitive tasks administered X X X X X 5
TABLE IV
Model Indexing of Data Procedures
Type of System Developer Procedures

Computer Erickson et al. (1978-81) 1. Review and catalogue content
2. Write Review Notes: interpretive comments stimulated by viewing comments are
guided by mental grid. Raise question about observed phenomena. Analogous to
field notes during participant observation
Content: descriptive, methodological, and theoretical comments
Provides synoptic view
3. Catalogue includes:
(a) major events or activities (e.g., morning business, reading period)
(b) constituent phases or subactivities (e.g., for reading: "get reading, lesson," e
"wind-up of lesson," "transition to seatwork") ""
(also called Participation Structure)
t=
(c) particular sets of interactions of individuals (teacher-student, student-student)
are identified
(d) overall topic of talk within the interaction (e.g., academic subject matter-related Q
or nonacademic subject matter-related)
(e) time-date generator used to superimpose numbers over copy of original picture m
(digital clock with time elapsing in hours, minutes, seconds, and tenths of t-
seconds). Provides time reference points for analysis in index system.
4. To retrieve data
(a) scan field notes and review notes
(b) view video
5. Transcription of segments. Attempt to display
(a) original text of speech, indicating breath-group phrase lengths, speech rhythms, n
Q
pitch changes, other aspects of speech prosody
(b) some indication of sequential flow of action accompanying speech
(c) description reporting location of speakers in space relative to one another
(d) description of most salient nonverbal behaviors (gesture, touching, facial
expression) that co-occur with transcribed speech
TABLE IV--Continued
o
Model Indexing of Data Procedures o
Type of System Developer Procedures
Computer (con'd.) Erickson et al. (con'd.) (e) translation of Spanish when Spanish was language
(f) display in three columns:
Text Translation Comments/Sequence o

Include:
overall sequencing
nonverbal behav- =~
iors
Filing System Griffin (used by Merritt to obtain 1. Note size of group videotaped (whole class, dyads, triads, etc.)
5"
data for secondary analysis of 2. Note participants (only children, children and teachers)
data collected for another 3. Record a rough characterization of the topic and nature of verbal and nonverbal o
project. See Merritt & behaviors (synoptic view)
Humphrey, 1981, p. 46) 4. Record names of participants entering and leaving the video field 0
5. Cross-reference to other videotapes recorded at the same time or to the audio
back-up.
6. Record grade, teacher, date of taping, time of day, indexer, equipment used, etc.
7. Record technical quality of the video and audio recording
used his or her own approach, for example, speech act analysis,
sociolinguistic perspective, teacher-child questioning sequences. Each
analysis produced a different picture of classroom discourse (Morine-Der-
shimer, Galluzzo, & Fagal, 1981; Morine-Dershimer, Galluzzo, & Tully,
1981; Morine-Dershimer, Lay-Dopyera, & Graham, 1981; Morine-Der-
shimer, Ramirez, Shuy, & Galluzzo, 1981; Morine-Dershimer, Tenenberg,
& Shuy, 1981; Tenenberg, Morine-Dershimer, & Shuy, 1981).
One way to think of this work is as a series of pyramids. The triangle that
forms the base of the pyramid is composed of a different approach or
perspective at each point. The point at the top of the pyramid is always
occupied by the research team of Morine-Dershimer and Tenenberg. The
research team, therefore, serves as the controller. All pyramids rotate
around or cycle through the team. While multiple perspectives are involved,
the research team's perspective remains the unifying theme and insures that
the focus remains true to the negotiated contract.
The question of secondary analysis is illuminated by the work in these
studies. Possibilities exist for additional secondary analysis given the
indexing systems developed and used for these projects. However, the
caution mentioned above must be considered further. That is, secondary
analysis consistent with the theoretical frame that guided and drove the data
collection will further increase our knowledge of aspects of teaching as a
linguistic process not explored in the studies to date. Secondary analysis
studies that use other groundings and other approaches to analysis of data
will provide other realities, which may or may not complement this work.
These studies will not disconfirm the findings presented in the next section;
rather, as Mead (1975) points out, they will merely provide a picture of
another reality, one obtained by using a different "locus of observation" or
different "tracer units."
Summary
The discussion of methodology, while brief, provided a picture of the
studies and their designs. The studies were shown to be multifaceted
projects concerned with identifying phenomena as they occur within and
across a variety of activities and settings. The work included naturalistic
observations, focused or topic-centered observations, and various beha-
vioral measures and natural experiments. Collection and analysis proce-
dures were shown to be systematic and often convergent. The next section
will present a summary of the major findings that come from this work.
TEACHING AS A LINGUISTIC PROCESS: SUMMARY OF FINDINGS

The framework presented in the previous sections identified a series of
constructs which underlie the emerging approach to the study of
teaching-learning processes. This framework provides a mental grid that

guides the methodologies used to collect data and the procedures used to
analyze and interpret findings. In this area, therefore, all aspects from
conceptualization to interpretation, including design, collection, and
analysis, are theoretically driven. This framework also was used to shape the
discussion of findings that follows; therefore, before proceeding to the
discussion of the findings, a brief summary of the central aspects of this
framework will be provided.
At the center of the mental grid is the conceptualization of the classroom
as a communicative environment in which communication between teacher
and students and among students is rule governed. Rules, in this instance,
are not posted and static rules like discipline standards often found on
classroom bulletin boards. Rules, from a linguistic perspective, are
culturally determined expectations for how to speak, when, to whom, and
for what purpose (Hymes, 1974). These rules form the frame of reference or
expectations that a person has for participating in face-to-face interactions
on both a general level and for a specific activity or event. In addition, rules
from this perspective can be brought to a situation (retrospective or past
rules) and can be signaled or cued in the present setting (prospective rules).
In classrooms, as in other communication settings, participants bring a
frame of reference to the event; however, the specific rules or expectations
for performance are signaled by the participants as they work together and
build on each others' messages and behaviors to construct the activity and to
reach the instructional goal. Because the teacher is the instructional leader,
the teacher's expectations or goals dominate even if the teacher chooses to
delegate decisionmaking to students. That is, as the teacher and students
work together, the teacher guides the construction of the activity and signals
expectations for when to talk, how to talk, and how to interpret the meaning
and goals of this talk. In other words, as the teacher presents content, the
academic task (cf. Erickson, 1982b), he or she also presents information
about the participation requirements (Erickson, 1982b; Green & Harker,
1982; Wallat & Green, 1982).
These structures co-occur and must be inferred from the verbal and
nonverbal behaviors of teachers and from the sequences of interactions
between teachers and students (e.g., question-response-evaluation, what is
positively and negatively sanctioned, and what information gets stressed).
Individual participants and researchers alike must observe not only what
language is used but how it is used and how various types of language
processes and devices are related and function. To do this, the participants
and the researcher must consider the local meanings and expectations, the
relationship of these meanings to such behaviors and processes in other
similar contexts, and whether or not these behaviors match what is expected.
In other words, the observer uses a form of contrastive analysis to determine
what is occurring and to predict what type of behavior or activity is expected

to occur. When what is expected does not occur, a frame clash is produced
(Cook-Gumperz et al. ; Green & Wallat, 1979, 1981b; Hymes; Mehan et al.,
1976). Identification and analysis of contrastive situations and frame clashes
serve to highlight what the expected behavior or process is and the rules for
academic, social, procedural, and contextual participation. Participation in
teaching-learning processes, from this perspective, requires active moni-
toring, adapting, checking, processing, and so forth, of information across
many channels and for a variety of co-occurring purposes.
In the following sections, findings from the 10 multifaceted projects that
form the core work considered in this synthesis and state of the art will be
presented. As in the case of constructs underlying this body of work, not all
findings will be presented. Together, these projects have produced more
than 200 individual findings to date. Still others will be forthcoming as
further analyses are undertaken, as the longer range studies are completed,
and as additional secondary analyses are undertaken. The findings,
therefore, have been clustered and the data reduced to provide clarification
of the features and processes involved in understanding the nature of
teaching and learning processes as linguistic processes, as well as the
relationship of these processes to student participation, knowledge, and
achievement.
The findings will be presented in three clusters. Each cluster has been
constructed to reflect a different aspect of the evolving picture of the nature
of teaching-learning processes as linguistic processes. Data will be presented
about ( a ) t h e nature of the classroom as a linguistic environment,
(b) linguistic skills required to participate in and meet the demands of the
everyday activities, lessons, and events of classroom life for both teachers
and students, and (c) the relationship of different patterns of linguistic
participation to student participation and developmental changes in
language abilities and demands for performance.
A word of caution is needed before proceeding to the discussion. These
studies were funded to explore the nature of classrooms as linguistic
environments and teaching-learning processes as linguistic processes. With
the exception of such works as Bellack, Kliebard, Hyman, and Smith (1966),
Smith and Meux, (1970) and work reported in Cazden, John, and Hymes
(1972), systematic exploration of these classroom processes had not been
undertaken; this past work was exploratory and limited in scope. The core
studies funded by NIE, therefore, are a series of first-generation studies that
focused on systematic exploration of classroom life over time. These
projects, then, explored a small group of classrooms in depth to explore the
value of and develop procedures for looking linguistically at teaching-
learning processes. This intent has been met. As suggested in the
methodology section, the tools for engaging in this type of research have
been adapted to classroom needs and the needs of educational researchers,

and systematic designs and procedures have been identified that can serve as
guides for future researchers and for funding agencies seeking to fund
additional work in this area.
In the sections that follow, the results of this set of projects will be
discussed, further demonstrating the value of this work. The limited number
of classrooms studied potentially poses problems in terms of generaliz-
ability; however, the range of classrooms was representative of grade levels
and different populations. In addition, the classrooms explored represent
both the public and private sector. Therefore, while limited in number, these
in-depth studies are representative of a wide range of educational settings.
The question of generalizability in these studies was one of within-classroom
generalizability; that is, the researchers were concerned with whether the
phenomena identified were representative of life in this classroom, the
developmental nature of these phenomena within classrooms, and the
relationship of these phenomena to group and individual student
performance, learning, and achievement. Therefore, rather than study a
large number of classrooms, this work begins to address the concern voiced
in Dunkin and Biddle (1974), that we do not know what teaching is for one
teacher throughout the day or over time. These studies are concerned with
the qualitative as well as the quantitative differences in teaching-learning
processes and their relationship to school performance, learning, and
achievement.
One advantage of this work was discussed briefly in the previous section;
the scope of information collected and the systematicity of the indexing
procedures developed make secondary analysis not only possible but
potentially valuable as sources of initial information about processes
previously unnoticed or undocumented. As the Merritt (Merritt &
Humphrey) study shows, secondary analysis grounded in the primary
analysis and primary data sources permits exploration of specific topics
within the body of data. The larger studies have already used this
topic-centered approach within the projects (e.g., Cole, Griffin & Newman;
Cook-Gumperz, Gumperz & Simons; Erickson et al.; Morine-Dershimer &
Tenenberg). Therefore, while limited in number, these studies can continue
to generate topics to be explored in larger, multiple-class studies. One such
project is currently under way at Harvard. Michaels and Cazden have
received funding from the Spencer Foundation to extend the work on
sharing time and narrative production begun by Michaels (Michaels, 1981a;
1981b; Michaels & Cook-Gumperz, 1979) as part of the Gumperz and
Simons project.
With these cautions as a frame, the discussion will now turn to an
exploration of the findings produced by these in-depth explorations of
classrooms as linguistic environments and teaching-learning processes as
linguistic processes. Data will be presented about (a) the nature of the
classroom as a linguistic environment, (b) linguistic skills required to
participate in teaching-learning events, and (c) the relationship of different
discourse patterns to student learning.
The Classroom as a Linguistic Environment

Analysis of the patterns of discourse used by teachers and students
revealed the differentiated nature of classroom activities and lessons.
Differentiation occurred in terms of requirements for participation as well as
in terms of content, of perceptions about language use and participation,
and of the types of interactions teachers had with different groups of
participants and individuals within these groups. In addition, various aspects
of this differentiated behavior on the part of teachers and students were
found to relate to student participation, performance within activities, and
student achievement.
Differentiation by classrooms. The findings on between-class differences
are limited. The primary focus of this group of projects was to obtain
precision descriptions and understandings of linguistic patterns and
demands within and across the differentiated tasks of the classroom. This
description and in-depth study of a small group of classrooms was a
methodological change in direction and was called for by Dunkin and Biddle
(1974) who suggested that rather than studying large groups of classrooms,
researchers needed to identify what teaching was like for individual teachers
and students within and across days. Much of the past work had focused on
time sampling and cross-class comparisons. This work, then, provides for
in-depth descriptions of the nature of teaching-learning processes within
individual classrooms.
Within the core group of projects, however, three projects engaged in
contrastive analysis: Morine-Dershimer and Tenenberg explored the
differences in language perception and language use and the relationship of
these factors to achievement and participation; Erickson et al. undertook a
"double" case study of two bilingual firsbgrade classrooms with bilingual
teachers and different classroom organizational patterns (e.g., one was
more individualized and one was more traditional); and Cook-Gumperz et
al. undertook a double case study with the contrast in the grade levels (first
and fourth grades).
Morine-Dershimer and Tenenberg, in their explorations of language use
and the relationship of this use to student performance over six language arts
lessons, found that three distinctive patterns of questioning behavior could
be identified: text-guided questions, pseudoquestions, and real questions.
The lessons analyzed were taught at the request of the researchers, but the
content of the lesson and the form was left to the teacher. The teachers were
asked to teach a language arts lesson to the class that included some form of
discussion and was not composed of worksheets (individual work). Within

these lessons, a series of linguistic analyses were undertaken. The
questioning chain analysis produced the three types of questioning styles.
Further exploration showed that these questioning styles were related to
student attention during the lesson and student final achievement in
reading. Each style will be explored briefly.
Teacher A used questions that students had difficulty in recalling what
they remembered as hearing. These questions were categorized as "not
quotable"; in fact, one student remarked that to ask questions the student
needed the textbook. Similar findings were also reported by DeStefano and
Pepinsky (1981; DeStefano, Pepinsky, & Sanders, 1982) who found that the
teacher talk during reading instruction resembled text language rather than
natural discourse. This pattern was related to low attention and low
achievement.
Teacher B used pseudoquestions (cf. Mehan, 1979b). Pseudoquestions
are found primarily in classrooms or formal learning situations. These
questions are questions to which the teacher has the answer and the student
does not. The student's task is to give the teacher the answer desired.
Pseudoquestions, therefore, contrast with real questions in that in the latter,
the person asking has or is perceived as having a genuine need for the
information. This style was related to low attention and low performance on
final reading achievement. Morine-Dershimer and Tenenberg found that
students perceived these questions differently from the other two types; they
categorized these questions as serving as "instructing" function.
In contrast to these two styles, the pattern of questioning used by Teacher
C was related to high attention and high final reading achievement. This
style was labeled "real" questions. Students perceived these questions as
serving an informing function. The teacher wanted to know the information
and also used these questions to inform students about what was important.
This project also demonstrated that students could clearly distinguish
between the functions of questions in school, at home, and at play. A caution
is needed here, however; the types of lessons sampled were limited and the
findings reveal a questioning style with a dominant approach. This does not
mean that the teachels only asked these types of questions and did not use
other forms, but that a definite pattern of use could be identified. This work
suggests the need to explore the functions of questions within and across
contexts as well as teacher and student perceptions of what questions do.
This work suggests that questions serve multiple purposes and that both
teachers and students are aware of these different functions.
Erickson et al. focused on a different aspect of the classroom as a linguistic
environment. In this project, naturally occurring events in two bilingual
classrooms were explored. The two teachers studied had different
organizational structures; by peeling back the layers of difference, that is, by
moving from macrofeatures (e.g., physical organization) to microfeatures,

the researchers were able to identify pervasive similarities in patterns of
linguistic interaction between teachers and students. On a class level, a
lesson level, and even an individual-within-a-single-lesson level, these
researchers found pervasive differences in the approach to teaching.
However, when they looked at the ways in which these teachers interacted
with individual students over time, they found that the two teachers were
similar. They used what is defined by these researchers as a "personalized
approach," an approach more like the patterns of interaction in the student's
community. They included students in the lessons; that is, they personalized
the instruction by including the student's name within the lesson (e.g.,
spelling sentences using the students' names); they engaged in private
conversations and de-emphasized competition; they used "culturally"
appropriate modes of address even though the participation structures
varied.
Early work by Piestrup (1973) suggests that such matches may support
differences in achievement. Work by Philips (1972) and Erickson and
Mohatt (1981) also suggests that the school-community match can support
student participation and learning. These studies, however, are only a
beginning. They indicate that surface level differences may not contain the
entire answer and that pervasive linguistic patterns that cross lessons need
further exploration if we are to understand the effect on student
performance of different types of linguistic environments. Work by Green
and Harker (1982) also suggests that the differences in interactional
approach used by teachers place different demands on students for
appropriate performance. These authors suggest that the difference may not
be so much in terms of academic performance as in terms of the acquisition
of strategies for participating in schooling activities; that is, different
approaches, as reflected in patterns of interaction, produce different types
of student knowledge about "learning how to learn."
The last contrast that will be reported in this section comes from the work
of Cook-Gumperz et al, This project contrasted a first- and a fourth-grade
class along various features. One feature was the types of knowledge
teachers expected of students in terms of performance in instructional
activities. Schafer (1981) found that by fourth grade, the teacher they
studied expected students to have mastered the general rules for
participating in instructional activities.
These studies of classroom similarities and differences suggest that further
work is needed on crossing institutional borders. This work indicates that
each classroom is a differentiated linguistic environment and that students
often perceive the differences between aspects of these environments. More
work is needed, however, to understand how the shifting demands of these
different environments influence students' knowledge of academic content
and what types of knowledge are required for effective and appropriate
participation.
Differentiation by group. In considering the findings on the differentiation
of linguistic environments within the classroom, the discussion will move
from a macrolevel (group) to microlevels, from lesson to individual, to
individual over time.
Discussion begins with consideration of group-level differentiation. Two
projects, Cole, Griffin, and Newman and Cook-Gumperz et al., focused
specifically on differentiation that occurs with task at the group level.
Building on the work of McDermott (1976, 1978), Collins (1981) explored
content of group reading instruction as well as the discourse patterns in
teacher-student interaction. Collins found that teachers created differen-
tiated environments for high-, middle- and low-group students. He found
that in low-group lessons for reading, the teacher consistently placed greater
emphasis on pronunciation, grammar errors, and single word decoding.
Less emphasis was given to content and meaning. This behavior pattern
contrasts sharply with the behaviors used with the top reading group. The
high group was encouraged to "go for the meaning." When members of this
group made errors in pronunciation, grammar, or decoding, these errors
were often ignored.
Similar findings also were reported by Eder (1982). In addition, Eder
found that teachers in her study did not change the composition of their
groups during the year. If, as suggested in this work, students extract
definitions for activities from participation requirements and form frames of
references for activities from chains of behaviors, the effect of this
differential treatment becomes clearer. Students in low groups have
different input in terms of content, strategies for reading, and definitions of
reading. Further work needs to be undertaken to explore the relationship of
this input to student reading performance. Microanalysis allows precision
description of the unfolding processes both within and across lessons;
therefore, by adopting a topic-centered approach and by using the
microanalysis procedures developed, future work should provide a
systematic picture of the effects of different types of reading practices on
student performance and acquisition of reading skills and processes.
In a related study, Cook-Gumperz and Worsley (1981) explored the
ability to tell a well-formed story which has connected sequences with group
placement in reading and with language arts skills. They found that in first
grade, there is not a direct relationship between good storytellers and
reading group placement. Nor did they find a relationship between this
ability and school tests of language arts or component reading skills. They
suggest that
as classroom curricula through the grades from first grade on place more reliance on written
literacy tasks, the relationship between low reading group, good story tellers and the high
reading group, good story tellers will diverge sharply. The former group will lack or be slower
with other supporting literacy skills with which to capitalize on their . . . flair for narrative
expression. (p. 15)
This work suggests that the resources students bring to the task of reading
are not being utilized. H e a p (1980) raises the same question when he
explores what counts as reading. H e suggests that students' performances
may be due to factors such as prior knowledge (resources), reading the task
requirements and not the text (e.g., reading the teacher behaviors), and
frame of reference (having a different view of the task). W o r k on linguistic
abilities of students, student-teacher interactions, and evaluation of student
p e r f o r m a n c e within and across groups suggests that the question of what
counts as reading, as performance, and what contributes to evaluation needs
further exploration. This work also shows that reading groups are not
treated equally even when time constraints and time on task may be equal.
The work of Cole, Griffin, N e w m a n , and others at the L a b o r a t o r y of
C o m p a r a t i v e H u m a n Cognition provides further clarification of the
relationship between task structure and student learning. In a tutorial
situation, a tutor misinterpreted a child's action. In attempting to " c o r r e c t "
the student " e r r o r , " the tutor apparently confused the student. The result
was that the student, when later working on the task alone, appeared to
know less than he did either before the tutorial or while working
cooperatively with the tutor. This work suggests that a teacher's on-line
assessment may inadvertently have a negative influence on a student's
performance.
Similar findings are reported by Harste ( G r e e n & H a r k e r , in press). In
this instance, a reading teacher signaled a rule to the student to help with
decoding. The rule signaled is not correct; the teacher corrects her behaviors
but does not overtly signal the change. The student continues to use the
frame set by the teacher and is unable to decode the word. No other problem
occurs of this sort during the lesson. When asked about the word at the end,
the student gave a nonsense word that included parts of all words missed.
The teacher's instructional strategies affected student performance. In the
end, the student was made to look as if he had less skills after the lesson than
at the beginning of the lesson.
Cole, Griffin, and N e w m a n (1978-81; 1981) also report the differentiation
a teacher used while using the same instructional plan. The teacher
a t t e m p t e d to teach the same instructional plan for long division to her entire
class. The class was divided into a series of ability groups and the teacher
proceeded to teach long division. Time of instruction was held constant. As
the teacher m o v e d from group to group, her instructional patterns changed
with perceived needs of students. This shift in instruction was accompanied

by differences in the amount of content covered. The top groups were able to
complete the task and engage in extra activities. The slower the group, the
less content was covered, even though the teacher intended equal content
coverage.
Hrybyk and Farnham-Diggory report a similar finding in terms of learning
centers. In observing classroom activities, they found that only those
students who completed assignments got to go to learning centers designed
to reinforce skills being taught. This practice meant that better or more
competent students got reinforcement activities, and those who were slower
and did not finish were unable to use the centers. They suggest that this
practice leads to differential treatment of groups and does not help those
who may need the most help.
To summarize, the work on group instructional practices by teachers
suggests that classroom instruction differs by group. Griffin (1981) suggests
that this differential treatment is due not to a single theory of pedagogy a
teacher holds but to differential theories of pedagogy based on perceived
needs of students. Work on how teachers perceive different groups of
students in terms of abilities and how these perceptions influence the
strategies used and the activities selected needs to be undertaken. As
indicated above, this type of issue can be explored on a microanalytic basis as
well as on a more global basis (e.g., interviews, planning). This work also
suggests that a focus on the preplanning stage will get at teacher intent;
however, if we are to understand the relationship between planning and
instruction, we need a precise description of the evolving instructional
event.
Differentiation within lessons. One of the areas of greatest knowledge gain
from these core studies and other related work is the area of language use
within lessons. This area has a high degree of convergence across studies;
therefore, rather than listing each study under every finding, representative
studies will be listed for each finding. To show the scope of convergence with
other work beyond the core set of studies, where appropriate, additional
work will be cited.
Within lessons, teachers were found to establish routines to guide
participation ( C o o k - G u m p e r z et al.; Cook-Gumperz & Gumperz, 1976;
Erickson, 1982a; Mehan, 1979a, 1979b; Merritt, 1982; Merritt &
H u m p h r e y , 1979, 1981; Wailat & Green, 1979, 1982). These routines were
found to be stable over time (Guzman, 1981; Michaels & Cook-Gumperz,
1979; Shultz & Florio, 1978; Wallat & Green, 1979, 1982). For example,
teachers were found to establish routines for conversational access (Merritt,
1982; Merritt & H u m p h r e y , 1981; Stoffan-Roth, 1981a) which signaled
accessibility or nonaccessibility to students. This set of signals is important
for managing the flow of activity. If the teacher is working with one group in
what is called the primary vector of activity (Merritt & Humphrey, 1981), he
or she must also monitor students in other groups and activities (the
secondary vector of activity). The teacher does this by using verbal and
nonverbal signals to indicate when a child can enter, when he or she is
leaving the primary vector (called "slotting out") to address someone in the
secondary vector. This signaling is important for maintaining engagement of
students in the primary vector of activity as well as for meeting the needs of
individual children in other vectors of activity.
Merritt and Humphrey (1981) found that the basis for such differentiation
depended on several contextual parameters: (a) the teacher's philosophy
about ideal student behavior during individual worktime, (b) the nature of
the school setting, (c) the mutual biographical context of teacher and
student (e.g., academic standing, extent of child's willingness to work alone,
frequency of previous attempts and successes at getting help from the
teacher), (d) the nature of the interaction in the primary vector (e.g., is
there "down time," a break in the activity), (e) the degree to which the
solicitor's address overlaps utterances of those in the primary vector, and
(f) the degree to which the teacher judges that he or she can successfully
respond to and dispatch the query without impairing the events in the
primary vector (p. 186). In addition, this work showed that once a teacher
has actually become engaged in a particular vector of activity, he or she will
aim to preserve the integrity of that vector.
However, there are times when demands from outside the primary vector
are made. At these times the teacher may (a) include primary vector
participants in the change of attention, (b) use split modality (verbal and
nonverbal together) involvement to partially slot-out, or (c) totally slot-out
(temporarily leave the vector). They found that the last strategy may take
several forms. The teacher may (a) act as if he or she had never left (e.g.,
inserting "Hi Carter" [the name of the child seeking entrance or help] in the
middle of talking to another child), (b) use a ritual form of slotting-out
(e.g., "excuse me for a moment"), or (c) use a ritual form of "slotting-in" or
returning to the primary vector (e.g., " O K , I'm sorry I had to interrupt").
If we shift the focus of the lens for a moment and look at the differentiation
process from the child's view, Merritt and Humphrey identify four strategies
that students use to seek individual attention. Students can (a) attempt to
overcome the problem or make a decision on one's own, (b) turn to another
student for help, (c) switch to an alternative activity, or (d) approach the
teacher anyway. Each decision carries different outcomes for the student.
This detailed discussion of signals for accessibility demonstrates the
complexity of the decisionmaking process within each classroom, highlights
the multiple processing strategies required of both teachers and students,
and identifies the constraints on teachers in giving differentiated help and on
students in obtaining differentiated help. This work also highlights the need
for an approach to the study of teaching-learning processes that can capture

more than one vector of activity at a time as well as the relationship between
vectors and participants.
In addition to having routines for accessibility, teachers were found to
have routine places for giving important messages (Shultz & Florio, 1978),
and for establishing requirements for participation in a lesson (Stoffan-
Roth, 1981a, 1981b; Wallat & Green, 1979). This latter area refers to the
establishment of systematic and recurring expectations or norms for
behavior that teachers hold for student behavior within and across activities.
Of special note is the fact that these norms and expectations are not signaled
overtly but are signaled by the way the teacher distributes turns, permits
talk, acknowledges contributions, and so forth. In other words, the teacher
signals the rules for conversational participation both verbally and
nonverbally (Erickson, 1982a, 1982b) and students must infer what the rules
are from the sequences of behavior.
Teachers were also found to vary the structure of lessons. For example,
teachers varied the types of opening moves they used (DeStefano &
Pepinsky: G r e e n , 1977; Guzman, 1980; Morine-Dershimer & Tenenberg).
Opening moves tended to serve dual purposes: to enforce orderliness and to
illustrate and make the lesson explicit (Guzman, 1981). Within lessons,
teachers were found to vary the use of directives. DeStefano and Pepinsky
found that the use of directives varied by reading group. In the lower group,
directives were used to direct activity, while in the middle group, they served
to control behavior. Teachers also varied the distribution of opportunities to
talk (Erickson, 1982a; Merritt & Humphrey) and the types of talk
permitted, for example, chatting (Erickson et al.) or informal talk (Hymes).
A n o t h e r area of variation or differentiation was expressive style.
Teachers varied in terms of positive and negative praise used, reprimands
given and positive and negative sanctions used (Cahir & Kovacs, 1981;
Erickson et al.; Guzman, 1981). Merritt and H u m p h r e y also found that
teachers differentiated or shifted roles and language use with different
organizational structures. For example, when the teacher used a structure
that required individual student participation, he or she shifted from the
director of the activity to the monitor. As the monitor, the teacher used
strategies to solicit information, evaluate students, and provide procedural
guidance. This work illustrates how the use of individual activities increases
the complexity of monitoring and managing required of teachers and
students. In this type of organization, the teacher must monitor more than
one activity vector in order to provide help, to know what is occurring, and
to be able to maintain the flow of activity in the classroom. The students also
have a complex task; they must monitor what the teacher and other students
are doing as they complete work on their own activities, and they must
monitor what the teacher is doing in order to get help. Individual work,
therefore, requires multiple processing of cues (Merritt & Humphrey).
The question of preferential treatment was explored also. Guzman (1980)
found that preferential treatment was not a unidimensional phenomenon
but varied by context and activity. Cooper, Marquis, and Ayers-Lopez
(1982) report related findings in their study of peers as teachers. They found
that students used different peer "consultants" in different contexts. In
other words, students differentiated requests for help and distributed these
requests among different consultants. This work suggests that students are
aware of the resources other students can provide and therefore distribute
the role of teacher to a peer differentially. Their work also indicates that
students use language for teaching and learning differentially; that is, they
use different strategies in situations that are cooperative problem-solving
situations than they do in asymmetrical situations in which one student is in
the role of teacher (Cooper, Ayers-Lopez, & Marquis; Cooper, Marquis, &
Ayers-Lopez, 1980). Similar findings were reported by Olmeida-Williams
(1980) for peer teaching situations in bilingual classrooms.
In summary, while these findings are only a representative set, they
demonstrate the variation and differentiation of language use, language
purposes, and language functions that occur within lessons. Recently, this
differentiation process was shown to reflect differences in teacher
instructional or pedagogical style (Erickson, 1982b; Green & Harker, 1982).
Style, from this perspective, is signaled tacitly and can be inferred from or
extracted from the ways in which the teacher interacts with students within
and across contexts, from the types of behaviors that are sanctioned, from
the types of feedback teachers give about appropriate behavior, and from
the types of behaviors used by teachers to manage the flow of activity across
various activity vectors. As indicated above, microanalytic approaches
permit precise descriptions of processes and relationship in "real time"
among processes. This type of approach also permits exploration of
variables defined by sequences of behavior, for example, the type and range
of opening moves, the effect of these moves on student participation, and
the intent of teachers in using the moves. In addition, these approaches can
help to identify the conditions for learning for individuals that occur within a
lesson; therefore, this type of research can explore the influence of specific
types of behaviors and activities on individual students' participation and
performance. (For further discussion see Carrasco, Vera, & Cazden, 1981;
Cole, Griffin, & Newman, 1979; Griffin, Cole, & Newman, 1982; Griffin,
Newman, & Cole, 1981; Newman, Griffin, & Cole, in press).
This work also suggests that the decisionmaking process for teachers
occurs within lessons and is based on student responses, that decision-mak-
ing is not a preplanning stage alone but rather an ongoing process in which
teacher and students both influence what occurs. This work provides a
description of the ways in which teachers shift their behaviors based on

perceived needs of students, the activity, or their own pedagogical intent. It
also provides in-depth, precise descriptions of how these teacher behaviors
support, constrain, or interfere with student performance and knowledge
gained from the activity (Newman, Griffin, & Cole, in press).
In the next section, additional discourse features will be considered.
Discourse Processes and the Relationship to

Student Knowledge and Participation
This section discusses the findings on discourse processes that are
characteristic of face-to-face interaction in classroom settings and the
relationship of these processes to student academic achievement and to the
acquisition of learning strategies, norms and participation in activities. This
section, therefore, combines information about features of discourse and
the purpose and outcomes of such discourse.
Before proceeding to the discussion of the findings, a word of clarification
is needed to frame the discussion. The studies in the core group focused on
identification of discourse processes and face-to-face behavior using current
advances in sociolinguistic and psycholinguistic analysis. One of the goals of
work in this area was to describe the nature of teaching-learning processes
that are part of the ongoing demands of classrooms and teaching-learning
processes. This goal has been met in part. What emerges from this work is a
set of systematic descriptions of lessons and e v e n t s - - a picture of how
participants in teaching-learning processes use language to learn and at the
same time acquire new knowledge about language use. Outcomes in this
area are of two types: outcomes related to academic performance and
outcomes related to participation in processes of interaction.
One caution must be considered. The work on instructional processes has
been intense and in depth for a limited set of classrooms and instructional
activities within these classrooms. However, when taken as a whole, this
information provides precise descriptions of evolving processes and initial
explorations of these processes to academic achievement. These results are
representative of life in classrooms studied and are suggestive of other
classroom settings. The findings come from studies that used contrastive
settings, events, or students. They represent, with one exception
(DeStefano & Pepinsky, 1981), a planned contrast both within and across
classrooms. Even though DeStefano and Pepinsky engaged in exploration
of a single classroom, these researchers used planned contrast among three
students that reflected systematic differences. This study, then, provides
information about life in classrooms from the perspective of students with
different backgrounds and abilities and explores how these differences not
only affected their learning and participation, they also served as a reference
point for exploring the performance of other students and the teacher.
The findings that will be presented here, therefore, provide information
about both the qualitative and quantitative aspects of life within and across
classrooms. One additional caution must be considered. In reporting the
findings and providing a picture of the evolving processes, it was necessary to
decontextualize these variables; that is, to remove them from context and to
isolate them from other processes to which they are linked. This procedure
was necessary to highlight what we have learned about the nature of
teaching-learning processes as discourse processes. Therefore, it is
important to remember that the discrete processes that appear in the
discussion and on the charts in this section are not independent of the
contexts in which they occurred. To anticipate the question of idiosyncrasy
of the findings, criteria were established when selecting the findings for
discussion. The processes presented here represent processes that were
found across activities and/or across settings, that extend knowledge of
previously identified processes, or that related to one of the outcome
measures (e.g., achievement, acquisition of discourse processes). The
findings in this section, therefore, will focus on a growing understanding of
the nature of teaching-learning processes as linguistic processes.
Discourse Processes: Use, Perception, and

Relationship to Performance Measures
Representative findings in 10 categories will be presented. Each category
reflects a different aspect of the conceptualization of teaching-learning
processes as linguistic processes. The categories are presented from general
discourse processes features to more specific, highly topic-centered features
(e.g., praise, sanctions). While the order of presentation is arbitrary, it was
selected so that the reader could begin with broader processes and then, with
these as a frame, begin to consider specific or more focused discourse
processes and their influence on participation and achievement. The 10
categories of representative findings are:
1. Nonverbal discourse features
2. Samples of context constraints on language use
3. Patterns of language use
4. Rules for speaking
5. Peer learning and language use
6. Asking questions and perception of questions
7. Perception of general language use for teachers and students
8. The nature of attention
9. Teachers' use of praise
10. Teachers' use of sanctions
The findings are presented in table form, with each table representing a
different cluster of findings. The 95 findings represent approximately half of
all discrete findings. Those selected are not directly redundant with the
findings discussed in the previous sections. They are those that can be readily
understood without further elaboration of specialized topics such as
narrative structure (Cook-Gumperz et al.), cohesion and linguistic analysis
(DeStefano & Pepinsky), and cognitive psychology (Cole, Griffin, &
Newman, 1978-1981, 1979). Each of these fields has a theoretical base with a
long history and to do them justice further clarification and identification of
underlying constructs would need to be presented to provide an adequate
frame for the findings. Readers interested in the specific findings in these
areas are referred to the original sources. The findings in this section,
therefore, are general and indicative of the nature of teaching-learning
processes as linguistic processes.
Nonverbal discourse features. Communication in classrooms is a complex

process that occurs in different channels. In all instances, verbal and
nonverbal messages co-occur. In some instances, nonverbal messages help
clarify verbal messages and, in other instances, they can be used to carry a
second message that is delivered simultaneously with the first message.
Table V presents a series of findings about how nonverbal behaviors are
used, their general characteristics, the purposes they serve, how a student's
awareness of the nonverbal channel is signaled, ways students learn about
the use of these signals, and other related areas and concepts.
Table V can be read in two ways. Each row provides a different type of
knowledge about one specific characteristic or aspect of the process. In a
new field, researchers must specify the various types of evidence used to
validate the construct. Each column within a row represents different
attributes of the construct. One of the underlying premises of this body of
research is that processes of interaction used by participants as they
communicate can be captured. Therefore, the columns on Table V
summarize the discourse process from different perspectives (the re-
searcher, the teacher, and the student). In addition, information about how
these processes are acquired by participants and identified by researchers is
provided. The information in each column provides an evolving picture of
the different aspects of the process.
As indicated in Table V, there is an "official" channel of communication
which tends to be dominated by the verbal messages. As presented in this
table, "unofficial" does not mean "unimportant." The reader will note that
what occurs in the unofficial channel are messages related to managing the
flow of lessons across different vectors of activity (e.g., monitoring the
groups that are working independently while the teacher works with one
group) and with providing individual help to students who are not directly
working with the teacher. The unofficial channel, therefore, becomes a
distinct resource for teachers who can consciously gain control over the
mechanisms involved in orchestrating two channels of information

simultaneously.
Samples of context constraints on language use. Table VI provides a
description of a broad range of attributes related to the general finding that
classrooms and activities within classrooms are differentiated learning
environments. As indicated in this table, researchers using a linguistic or
discourse approach have been able to identify multiple levels and types of
contexts, rules for participation, descriptions of how contexts and rules
within the contexts are constructed and the interrelationship of the academic
task structure and the participation structure. This work, therefore, defines
"linguistic environment" as both the academic task and the social
participation task structures.
The picture of the classroom as a differentiated linguistic environment
that results from the work reported in Table VI is one of a rule-governed
environment. This work suggests that we can no longer assume that
language used by teachers and students is the same across all content and
contexts. The evolving picture of classroom life is one of shifting demands
within and across lessons. The work reported in this table demonstrates the
complex features of classroom life and teaching-learning contexts that are
orchestrated by teachers as part of their day-to-day activities. In addition,
this work suggests that certain aspects of classroom life that are observable
are developmental in nature. Also, the systematic approaches developed
within these projects provide a vehicle for both researchers and teachers to
observe how these developmental processes evolve over time within and
across educational activities and institutional borders (e.g., grade level,
organizational levels, building levels, schooling levels). Such work will
extend our knowledge about the interrelationships among developmental
processes, environmental constraints, and instructional processes.
Patterns of language use. This section moves to a more global level in the
presentation of findings. In the Appendix, Section A on representative
findings on patterns of language use, specific patterns indicative of
teacher-student talk are presented. These findings represent types of
patterns that were explored and general factors related to each pattern.
What this set of findings and those which follow provide is an emerging
language for describing and talking about strategies (i.e., linguistic devices)
teachers and students use as part of teaching-learning processes. These
findings, therefore, represent a "sampler" of the patterns that can be and
have been identified.
Rules for speaking. In previous sections, classroom discourse was
characterized as rule governed. The Appendix, Section B on representative
findings on rules for speaking, provides an overview of salient features
associated with the development and signaling of rules for speaking and
participation. As indicated in this section, the process of learning rules
TABLE V
Representative Nonverbal Discourse Features (:o
Focus: The Process of identification
Specific Observed How Awareness Way Learned

General Finding Characteristics Purpose is Signaled or Obtained Related to el
messages can be "official" teacher maintaining the flow students who need observe behaviors; differentiation of
sent across two involvement is in and instruction access from other behaviors indicate classroom commu-
channels of com- verbal channel (Merritt vectors use nonver- awareness of pat- nicative environ-
munication (nonver- & Humphrey, p. 136) bal channel terns expected ment
bal and verbal) si- el
multaneously el
unofficial (non-verbal managing the flow of observe behavior; tacit rule is learned obtaining help
modality) channel can activity when more patterns indicate by observing inappropriate participa-
be used without offi- than one group is awareness of expect- stances of successful tion
ciaily interrupting (Mer- used ed and appropriate and unsuccessful ac- slotting-in
ritt & Humphrey, p. behaviors cess gaining slotting-out (1
136)
o
behavior indicates observing instance of providing help
.=
teacher collaborates to 1. provide aid
carry out events in the 2. give directions on awareness of appro- collaboration by managing flow of ac- ,-,I.
secondary vector (Mer- next phase of activ- priate patterns of be- teacher and others tivity
ritt & Humphrey, p. ity havior
106) 3. prevent crisis/emer-
gency
4. check work
5. display of misinfor-
mation
6. brief inquiry
7. student can't contin-
ue without help
verbal channel sanc- focus attention, man- verbally signaled and overt rule tacit strate- locus of attention
tioned in favor of non- agement nonverbally adhered gy: focus on verbal guiding activity
verbal (kindergarten) to
(Merritt & Humphrey,
p. 171)
nonverbal modality is initiation of individu- response to sum- observation of con- managing flow of ac-
preferred for summon- al/private teacher and mons ventionalized signal tivity
ing student child interaction individualization
private interactions
directing students to focus on verbal mean- attention to verbal developmental learning to depend Q
verbal modality (nur- ing (more decontex- over nonverbal cues ability (Cook-Gum- on verbal cues
sery level specified) tualized language) (e.g., student, when perz, 1981c) (adult model)
information in two signaled by verbal ability to decontex- .=
channels, uses infor- actions of others; tualize meaning
mation in verbal can be overtly sig- (Cook-Gumperz,
channel over visual) naled or extracted 1981 a)
(Griffin, 1981) from behavior pat-
terns observed
==
r-
(1
Io
rid
TABLE Vl
Reprseentatlve Samples of Context Constralnta on Language Use 0
Focus: The Process of IdenUflcatlon

General Finding Characteristics Purpose is Signaled or Obtained Related to
context constrains participation or event manage the organiza- appropriate behavior inferred from behav- event organization
language use structure is signaled tion of an event while used during different iors of others and appropriate participa-
from actions of partici- simultaneously pre- phases of lesson, verbal and nonverbal tion o
pants (Erickson, senting content activity, event behaviors of teacher acquiring strategies
1982a; Erickson & for use in activities Q=
Shultz, 1981; Philips, managing or guiding
1972, 1974) activities
norms or expectations manage the organiza- appropriate behavior inferred from behav- appropriate participa-
for behavior are sig- tion of events over
5"
is used during differ- iors of others, feed- tion
naled during activity-- time ent phases of les- back from teacher stability of teaching
signals co-occur with establish expectations sons and over les- (verbal and nonver- patterns }
content (Borman & for participation for ac- sons of similar type bal) managing or guiding e~
Barrett, 1981; Cook- tivity being constructed an activity
Gumperz et al.; Erick- develop conventiona-
son, 1982b; Griffin, lized patterns or cues
Cole, & Newman,
1982; Merritt &
Humphrey)
context mediates levels of context exist 1. /oca/context sig- appropriate participa- inferred from behav- appropriate participa-
meaning simultaneously nals what the im- tion ior patterns tion
(Bloome & Green, mediate activity is--
1982; Erickson & what is occurring appropriate participa-
Shultz, 1981; Merritt & now tion
Humphrey) 2. event context sig- appropriate participa- 1. inferred from be- (Bloome, 1981;
nals the type of tion havior patterns Collins, 1981 ;
event (e.g., group 2. obtained from Cole, Griffin, & New-
lesson) verbal information man, 1979;
3. setting context is observed organiza- 3. differentiated Eder, 1982;
signaled by the tion events, physical McDermott, 1978
physical context description by group areas Stoffan-Roth, 1981b
member
4. mutual biographical signaled by patterns over time frames for surface level differ-
context is history of of behavior used by behavior are estab- ences in instruc-
teacher & child over teacher and child as lished from inter- tion (Erickson et
time they collaborate to action al.)
meet goals
rule-governed (all organize activities recurring patterns of inferred from clash shared context con-
studies) conventions serve to behavior between expected struction Q
support participation
underlying operating
types of behaviors
become more pre-
behavior and ac-
tual behavior
orderliness of activity
stability of expecta-
i
principles (Erickson, dictable (Cook- tion
1982a) Gumperz et al.) (Erickson, 1982b;
}
Guzman, 1980;
Merritt & Humphrey;
Shultz & Florio,
1978;
Wallat & Green,
1979, 1982) r-"
students transform meet felt needs or observed patterns of inferred from

context to own use or goals of students behaviors sequences of behavior
own goals (Hrybyk & Farnham-
Oiggory)
share what is being
read (Bloome & O
Green, 1981; m
Hymes)
(Continued on next page) re

..&
o
TABLE Vl---Contlnued
Representative Samples of Context Constraints on Language Use
Focus: The Process of Identlflcetlon

General Finding Characteristics Purpose is Signaled or Obtained Related to
teachers transform teacher appropriates observed teacher ac- inferred from se- task performance =E
evolving task to meet student behavior to tions during instruc- quences of behaviors intentional teaching
o
goals and complete complete task (Cole, tion in response to (Cole, Griffin, &
task Griffin, & Newman)- student actions Newman, 1979;
use what student Griffin, Cole & New-
brings to task to help man, 1982; =
him/her complete the Griffin, Newman, &
task. Cole, 1981 ;
Newman, Griffin, & 5"
Cole, 1981 )
context constrains oral signals who talks, patterns of appropri- inferred from actions communicative com- (3
a,
production (Carey, how, when, and to ate talk and lan- and verbal behaviors petencies (Hymes,
Harste, & Smith, 1981; whom guage use of participants within 1974) O
Cole, Dore, Hall, & (Hymes, 1974) and across different
Dowling, 1978; Geni- contexts
shi, 1979; Gumperz et
al., 1981; Steinberg &
Cazden, 1979)
academic task struc- provides students with 1. sequence of be- inferred from se- content presentation
tures provide different academic content in haviors (verbal quences of informa-
information than does organized ways and nonverbal) tion presented how tasks are con-
the participation struc- 1. presents logic of 2. steps in content structed
ture (see below); these subject matter presentation (Cole, Griffin, &
structures co-occur 2. presents information Newman, 1979;
(Cole, Griffin, & New- content in various Cole, Hood, &
man, 1979, 1981; steps McDermott, 1978;
Erickson, 1982b; Grif- 3. "meta-content" Green & Harker,
fin, Newman, & Cole, cues, steps to, and 1982)
1981) strategies for com-
pleting task task analysis
4. physical materials (Griffin, Cole, &
through which tasks Newman, 1982;
are accomplished Griffin, Newman &
Cole, 1981; Gear-
hart, & Newman,
198O)
social participation guides activity sequences of behav- inferred from expec- participant structure
Q
structure involves 1. governs sequencing iors used and re- tations for and be-
1. whole pattern of in- and articulation of ceived haviors used during organization (b
teractions interaction interaction~how structure .=
2. configuration of 2. defines communica- people hold each
roles tion roles other accountable D
}
3. distribution of rights and construct the ac-
and obligations tivity
(Erickson, 1982a; Phi-
lips, 1972, 1974) m
m
m
Academic task structure & social participation structure = "linguistic environment"
B
O
Q
(o
IO
IO
requires conversational inferencing because rules are signaled both tacitly

and explicitly. When considered in light of the findings on the differentiated
nature of classrooms, the inferencing process required, which is cognitive in
nature, is ongoing; that is, while routines are established and activities begin
to recur, demands for participation shift frequently both within and across
lessons. Therefore, for a student to participate appropriately at any given
point in a lesson, he or she must actively interpret the meaning being
signaled in the official and unofficial channels of communication.
Peer learning and language use. In this section, the lens of the microscope
shifts from a focus on the teacher-child interaction to peer interactions for
instructional purposes. The evidence outlined in Section C of the Appendix
indicates that peer teaching situations are contexts for intellectual
development, The findings presented in this section of the Appendix
provide a picture of the resources students have available in both informal
and formal peer learning/teaching situations. This work indicates that
children in primary grades are aware of both the academic and social task
requirements of teaching and learning, and this awareness becomes an
active resource in guiding the student in the role of peer teacher.
Asking questions and perception of questions. One of the major areas of
past work on the study of teaching has been the area of questioning. Findings
on both question asking and perceptions of questions by teachers and
students are highlighted in Section D of the Appendix. This work diverges
from most of the early work by considering the functions or purposes of
questions. Therefore, this work provides new information and identifies a
viable new direction to the study of questioning and its role in instructional
processes. For example, both teachers and students perceived questions as
serving more than one purpose. The finding that questioning is an area in
which the rules for use and/or response are not always clear to students
provides further information about the nature of the demands for
information processing required of students. It also helps identify another
attribute of the linguistic differentiation process in classrooms. Further-
more, this set of findings suggests that exploring the nature and purpose of
questions from a linguistic base will help clarify one of the most frequently
used and salient linguistic devices available to teachers for instructional
purposes.
Perception of general language use for teachers and students. The findings
on student and teacher perceptions of language use suggest that both
students and teachers are aware of the differentiated nature of the classroom
as a linguistic environment. As indicated in Section E of the Appendix, at
times teachers and students view events from similar perspectives and at
other times they view events from different perspectives. In addition,
student perceptions of language functions and different types of language
use were found to influence student participation. This work reflects an
approach that explores the nature of processes from the participants'

perspective. Additional work needs to be undertaken to explore the ways in
which teachers can use strategies for obtaining students' perspectives as
instructional resources.
The nature of attention. The findings on the nature of attention (section F)
are more complete than those overviewed previously. The reason for the
density of findings in this area is that attention was the focus of a substudy of
one of the projects; that is, it became a topic for specific study for one study
(May, 1981). In addition, findings were generated by work on other
processes in other studies (e.g., Cooper, Ayers-Lopez, & Marquis;
Morine-Dershimer & Tenenberg). Central to the topic-centered study on
attention was the question of what counts as attention. In systematically
exploring this question, May (1981) found that attention, like other aspects
of face-to-face behavior in instructional activities, varies across contexts, by
student, by teacher, and by goal or intent of the activity. In addition, May's
work also shows that teachers are aware of instances of attention and
inattention, use various indicators to determine attention and inattention,
and make moment-by-moment decisions about how to react to or sanction
instances of inattention.
Other work reported in this section of the Appendix indicates that
students attended differentially to other students, to teacher talk, and to
different types of language processes. Developmental differences were
found with regard to what aspects of linguistic performance of teachers and
students became the focus of attention. In addition, relationships between
student status and achievement and attention to specific linguistic features
of teaching-learning discourse were found. No differences, however, were
found with regard to teacher questions and student responses for ethnic
groups.
Teachers' use ofpraise. Findings on both teacher use of praise and student
perception of praise are presented in Section G of the Appendix. Praise was
shown to be a differentiated linguistic phenomena that served a variety of
purposes; it functioned as an attention focuser, a confirming device, a
framing device, as well as a reward for appropriate performance. Praise was
significantly related to student participation, student status, and student
achievement. This was especially true of higher status students or
high-achieving students (as determined by entering reading achievement).
On the whole, students perceived praise as deserved (Morine-Dershimer &
Tenenberg).
Teachers' use of sanctions. Two distinct levels of analysis of sanctions and
two different approaches were identified: verbal sanctions and behavioral
sanctions. Linguistic analysis of the use of sanctions in classrooms is
presented in Section H of the Appendix. This work demonstrates that the
most frequently occurring type of sanction is not the behavioral sanction for
inappropriate behavior or behavior problems, but a linguistic sanction

which tacitly indicates to students when and where it is appropriate to talk.
The dramatic contrast between placement sanction (250 out of 297
sanctions) and all other sanction provides clarification of how rules for
speaking, for participating, and for appropriate language use are signaled.
In addition, this work provides another area of clarification of information-
processing demands on students.
The work using the linguistic approach generates new terms for often
unnoticed processes and helps to explain how teachers manage the flow of
interaction while simultaneously presenting and managing the content of
lessons. The concept of sanctions provided by the linguistic approach also
extends our knowledge of communicative competence required in and
across the differentiated activities. The refinement of our knowledge of this
construct provides a point of departure for further exploration of how
teachers orchestrate and manage instructional conversations, construct
rules for appropriate performance, and teach children about the strategies
for appropriate behavior in the differentiated learning environments that
make up daily life in classrooms.
Summary
The sections above presented an overview of the nature of teaching-
learning processes and the relationship of select processes to student
participation and academic performance. This information produced a
picture of the classroom as a linguistic environment and teaching-learning
processes as linguistic processes. The descriptions and constructs generated
from the microanalytic studies provide a language that can be used to
explore how teachers fine tune instruction within and across the
differentiated environments of classrooms and other educational settings.
This work demonstrates that tasks and activities within and across
educational settings do not structure learning; rather, teachers and students
acting on these tasks and on each others' messages and behaviors construct
or create these tasks. Curriculum, then, within this perspective is an evolving
process that occurs within general frameworks that may or may not be static
(e.g., lesson plans are guides; lessons evolve or are created through
interactions). In addition, classrooms were defined as environments in
which teachers and students develop shared meanings for activities, and
teaching-learning processes are often developmental in nature.
Work in this emerging field has identified demands on both teachers and
students for participation and learning. As shown above, teachers must
attend to activities in more than one vector of activity; that is, they must
manage both the primary activity for the group with which they are working
and the flow of other activities for students not directly involved with the
teacher at the moment. Within an activity, teachers were found to attend to
both the academic and social participation structure; that is, as the teacher
delivered academic content, he or she also provided information about how
to participate, what behavior was appropriate, and when to talk. Teachers
also were found to differentiate conditions for learning within and across
classroom activities and to use linguistic cues to evaluate student
performance in both academic and participation areas.
Student knowledge required for appropriate participation also was
identified. Microanalysis of student participation requirements showed that
students must know academic information and how and when to display this
information. Being accurate or right was not enough; students needed to
present information in appropriate form at the appropriate time. In other
words, students had to know both the form and the content required.
In addition, peers were shown to be effective teachers, and peer
learning/teaching situations were shown to be a source of intellectual
development. Peer interactions occurred in formal and informal situations
within and across classrooms; this work also showed that teachers cannot use
all informal peer learning activities as "formal" learning activities, since
these do not meet student expectations for "school" activities. Further-
more, work in this area found that discontinuities between home and school
existed, some of which were not a problem for teacher or student and some
of which were. If the discontinuity intruded on the "formal" world of the
classroom, it became problematic for teachers. If the teacher was unaware of
linguistic differences between home and school use of processes such as
narrative structure, discontinuity became a problem for students because it
often led to negative evaluation of student ability. This work suggests that
teachers will be better able to fine tune their instructional practices if they
have information about the unofficial world of learning for children, the
official world of learning, and the mesh between these worlds and the world
of the school curriculum.
Future directions: Research. The findings presented in these sections are
representative of the types explored to date. However, the multifaceted
nature of the collection procedures and the studies themselves and the
systematic indexing of this information means that secondary analysis
grounded in the primary analysis is viable. The work by Merritt and
H u m p h r e y demonstrates both the feasibility and the value of such analysis.
Potential areas for further analysis and synthesis within the core studies
include instruction in reading (Cook-Gumperz et al.; DeStefano &
Pepinsky; Erickson et al.) in writing (Cook-Gumperz et al.; Hymes); in
speaking (Cole, Griffin, & Newman; Cook-Gumperz et ai.; Erickson et al. ;
Hymes; Morine-Dershimer & Tenenberg); in math (Cole, Griffin, &
Newman; Erickson et al.); and in social studies and science (Cole, Griffin, &
Newman; Cook-Gumperz et al.).
A second area for potential future research using the data banks from
these studies includes work on effective organizational structures (Cole,
Griffin, & Newman; Cook-Gumperz et al.; Cooper, Ayers-Lopez, &
Marquis; Erickson et al.; Hrybyk & Farnham-Diggory; Hymes; Merritt &
Humphrey; Morine-Dershimer & Tenenberg); on management (all projects
in the core contain information relevant to this area of exploration); on
technology and education (Cole, Griffin, & Newman); on home-school-
community interaction (Cole, Griffin, & Newman; Cook-Gumperz et al.,
Erickson et al. ; Hrybyk & Farnham-Diggory; Hymes; Morine-Dershimer &
Tenenberg); an on evaluating student performance (Cole, Griffin, &
Newman; Cook-Gumperz et al.; Cooper, Ayers-Lopez, & Lopez;
DeStefano & Pepinsky; Erickson et al.; Hymes; Merritt & Humphrey;
Morine-Dershimer & Tenenberg).
With tighter funding resources, secondary analyses such as Merritt and
Humphrey's may become more frequent. The data banks listed above
provide one source for such potential analysis. The areas listed above do not
necessarily represent those that form the primary foci of the core studies;
rather, these are areas in which systematic data exist and for which
additional information can be obtained about the nature of teaching-learn-
ing processes as linguistic processes and the relationship of these processes
to cognitive, linguistic, and social knowledge required for successful
participation in schooling and learning activities. A word of caution is
needed. Inclusion of a data bank in this list of areas for potential secondary
analysis, however, does not automatically mean they are open for general
access. Rather, access to this information is governed by the contract
negotiated by these researchers with the schools and populations studied.
However, secondary analysis consistent with the negotiated contract and the
theoretical frame is possible, as Merritt and Humphrey demonstrated.
Future directions: Training protocols for practice and research. In addition
to directions for research, these projects also provide a rich resource for the
development of protocols for training researchers and for helping teachers
learn about the strategies and constructs involved in conceptualizing
teaching-learning processes as linguistic processes. A set of protocol tapes
have recently been developed to help teachers and trainers learn about the
nature of classrooms as a linguistic environment and teaching-learning
processes as linguistic processes (Cahir & Kovacs, 1981). The protocol
materials include a training manual, a student participant's book, and
videotapes with illustrative classroom events. Other such materials can be
developed from these projects to illustrate the different constructs and to
help people learn to look at and explore classrooms from a linguistic
perspective.
The second type of protocol that might be developed is one that provides
researchers with a picture of the data bank, information about the
procedures used to collect this data, an index to the corpus and background
information available, and examples of how to proceed with such research
Green: Teaching as e Linguistic Process 229
a n d h o w to d o analysis using the p r o c e d u r e s d e v e l o p e d in this e m e r g i n g

field. This t a p e , t h e n , is b o t h a t r a i n i n g t a p e a n d a r e s e a r c h " s a m p l e r . " Such
a t a p e c o u l d also facilitate d e c i s i o n s a b o u t s e c o n d a r y analysis studies.
In a c t u a l i t y , the use of t h e s e m a t e r i a l s a n d d a t a b a n k s has o n l y just b e g u n
to b e e x p l o r e d . D i a l o g u e m u s t c o n t i n u e a m o n g i n d i v i d u a l s c o n c e r n e d with
t e a c h i n g ; for e x a m p l e , t e a c h e r s , r e s e a r c h e r s , t r a i n e r s , a n d o t h e r e d u c a -
t i o n a l p r o f e s s i o n a l s w h o a r e i n t e r e s t e d in the p o t e n t i a l use o f t h e i n - d e p t h
d e s c r i p t i o n s g e n e r a t e d f r o m t h e studies o f t e a c h i n g as a linguistic p r o c e s s .
T o illustrate t h e v a l u e a n d use of this w o r k in t h e i m p r o v e m e n t of p r a c t i c e ,
t h e last s e c t i o n o f this s y n t h e s i s will p r o v i d e a s u m m a r y o f o n e such
application.
Future directions: Application to practice. T h i s s e c t i o n d e s c r i b e s b r i e f l y
h o w the c o n s t r u c t s , findings, a n d m e t h o d o l o g i e s can be u s e d to i m p r o v e
p r a c t i c e a n d as i n s t r u c t i o n a l s t r a t e g i e s . T h o s e c o n c e r n e d with a p p l i c a t i o n o f
r e s e a r c h to p r a c t i c e will find this s e c t i o n of special i n t e r e s t . This discussion is
n o t an i s o l a t e d i n s t a n c e ; r e c e n t l y t h e A m e r i c a n F e d e r a t i o n of T e a c h e r s
s e l e c t e d the findings in this a r e a for inclusion in a d i s s e m i n a t i o n p r o j e c t for
N I E a l o n g with the w o r k of r e s e a r c h e r s such as K o u n i n , B r o p h y , a n d
E v e r t s o n o n effective t e a c h i n g , m a n a g e m e n t a n d o r g a n i z a t i o n .
The discussion of constructs presented here are the reflections of one
teacher/supervisor (Deborah Smith) who incorporated the constructs and
findings p r e s e n t e d a b o v e in h e r d a i l y w o r k a n d h a d a p r e v i e w of t h e s e
c o n s t r u c t s a n d findings b e c a u s e she w o r k e d on t h e synthesis. She
s p o n t a n e o u s l y e x t r a c t e d a n d a p p l i e d t h e c o n s t r u c t s f r o m t h e s e studies a n d
a p p l i e d h e r s u p e r v i s o r y w o r k with p r e s e r v i c e m e t h o d s s t u d e n t s ( J u n i o r s ) .
S m i t h (1982) f o u n d the following:
One of the main reasons I used so much of what we had worked on over the summer was that
it just struck me, as an experienced teacher, as being so powerful. I recognized things 1 had
done--strategies and language 1had used, almost unconsciously, I think--and saw them for the
first time as purposeful, cognitive entities. Notions like flagging, slotting in, redundant channels
evoked a shock of recognition; my guess is that many "good" teachers evolve, or pick them up
from other teachers but never examine them consciously as part of their teaching. Like many of
the preservice methods students, I had thought of my teaching as mainly "doing content"; the
form I acquired was merely an effective means to an end.
So when I began observing and conferencing with methods students and reading their
comments in their participation logs, all of the ideas from the summer work were fresh in my
mind. I was excited about them, and it seemed natural to share them with students. I have to
admit that, at first, I shared them as content; as the semester progressed, I began to consciously
employ them as the form of what I did, also.
In the r e m a i n d e r o f this s e c t i o n e x a m p l e s o f S m i t h ' s specific a p p l i c a t i o n of

c o n s t r u c t s a n d findings r e l a t e d to p a r t i c i p a t i o n rules, rule shifts within
lessons, e v a l u a t i o n o f s t u d e n t p e r f o r m a n c e , m e a n i n g as c o n t e x t specific,
c o n t e x t as c o n s t r u c t e d , c l a s s r o o m s as c o m m u n i c a t i v e e n v i r o n m e n t s , a n d
l a n g u a g e use are d e s c r i b e d .
Participation rules. Participation rules offered the most difficulty for

students. They seemed to expect that children would continue following the
same ground rules that the cooperating teacher had established, no matter
who was teaching, and thus rarely stated any rules for their own lessons.
Children initially spent a good deal of time testing the old rules, in order to
establish, through interactional sequences, whether this context was, in fact,
the same as their usual classroom context. From the student teachers' frame
of reference, this behavior was an affront, rather than an active attempt to
infer rules. In the first 4 days in the classroom, 16 of the 19 of my written
comments on the students' teaching concerned the lack of clear rules for
participation in lessons.
When students did begin stating participation rules clearly, they often
failed to monitor their enforcement of those rules, as they continued with the
academic task at hand. For example, one student stated the "hands up for a
turn" rule, then alternated calling on children with hands up and allowing
children to respond spontaneously. In the middle of the lesson, a child
volunteered a comment and was told, "L., from now on, you wait until I ask
y o u . " In conference with me, the student was chagrined to learn that her
effort to enforce the rule had resulted in the negative evaluation of a child's
active attempt to participate as others had in the same lesson. In addition,
she realized her own inconsistency had contributed to the children's
confusion and the breakdown of the lesson.
Rule shifts within lessons. Students sometimes encountered trouble when
their own behavior was taken by the children as a signal that a new set of
participation rules was in effect. In gathering the class around her for a story,
a student began by clearly establishing rules for listening and sitting during
the story. However, a few pages into the story, she paused to ask a general
question to the class, about the colors in the picture. She did not ask a
specific child to respond, nor did she require a raised hand for a turn, thus
children eagerly volunteered answers; then, on each succeeding page, they
continued to volunteer comments about the colors. The student teacher
sensed that something had gone awry, but was unaware that her own
question and acceptance of spontaneous replies had initiated a new rule:
anyone can spontaneously comment on colors in the picture. In focusing on
the academic task, and forgetting to monitor the effects of sequences of her
own behavior on already established participation rules, the student
inadvertently sabotaged both.
Decisions to shift rules in the middle of a lesson sometimes left children
confused as to which rules were in operation. For example, a student teacher
began accepting spontaneous comments consistently, and thus signaled that
hands were no longer required for a turn. However, this covert message was
missed by one child, who continued raising his hand, while others
volunteered answers. In contrast, later in the semester, after much
discussion with me about rules for getting a turn, another student started her
lesson by stating the "hands up for a turn" rule; halfway through, she
decided that this procedure was taking too much time and inhibiting the flow
of children's responses to the poetry she was reading. She verbalized a
change in participation rules, "You can just tell me now," and children
smoothly shifted to taking turns when a pause occurred.
Evaluation of student performance. Two of the education students also
adopted their cooperating teacher's negative evaluation of several black
children's participation styles during lessons. The classroom teacher had
decided the children's bantering and chatting was "naughty" and clearly
conveyed this to both the children and the students. Through discussion of
alternative cultural explanations for differences in participation styles, the
students gradually questioned whether these children meant to be naughty
or were merely perceived to be naughty from the teacher's point of view.
This incident confirms findings by Michaels and Cook-Gumperz (1979) and
Hymes on the importance of children's correct participation in lessons and
the consequences when they fail.
Meaning as context specific. The importance of monitoring behavior
sequences in making judgments about a child's participation in a lesson was
highlighted for one student in a kindergarten class. She had given directions
for the children to choose a blue crayon and color in the biggest balloon.
However, as she turned her head to monitor other children, she missed one
child's interaction with a nearby student teacher. The latter had noticed that
the child had no blue crayon, gone to get him one, and placed it in his box.
When the student teacher looked at him, she saw that he was behind the
others, only now reaching for his crayon, and gazing at the other student
teacher. She reprimanded him for not following directions and not paying
attention. In this case, if we consider the four levels of context identified by
Merritt and Humphrey, we can see that, in the local context, she failed to
monitor all the events which resulted in the behavior; in terms of
biographical context, she adopted the cooperating teacher's frame of
reference (the child was bilingual and often scolded for inattention); and
within the event context, she incorrectly inferred that the child was not
attempting to comply with the requirements. In conference, she was upset to
learn what had really happened, and the effect her frame of reference had
had on her evaluation of the child.
Context as constructed: Routines and rituals. One student experienced
firsthand the power of established routines in a first-grade classroom. Her
cooperating teacher had a set of opening exercises which never varied in
kind or sequence, only in particular content (e.g., pledge to the flag, song,
today's date, etc.). On the student's last day, her cooperating teacher was
absent and the substitute asked her to lead the opening exercises. The
children cheerfully and cooperatively followed her lead through the entire
15-minute sequence, much to the student's surprise and delight. Because she
knew the rituals in that classroom, she was able to step into the leadership
role; the more experienced substitute teacher clearly expected that her own
lack of knowledge in the same area might have led to a less smooth
beginning.
Classrooms as c o m m u n i c a t i v e environments: Contextualization cues.
Student teachers had difficulty monitoring the ways in which their nonverbal
messages sometimes thwarted their academic lessons. They often failed to
notice when, in spite of saying the correct words, their voices, facial
expressions, body positions, or distance from the group made contradictory
statements to children. For example, a student who was sitting on the floor
and writing a story with children placed her chin in her hand and rested her
elbow on the floor. Her face showed little expression and her words were
garbled by the hand on her chin. In response, children began lying down,
talking among themselves, and volunteering fewer ideas. Although the
student continued asking questions about the story, her nonverbal cues had
changed the context for the children. Their behavior reflected the student
teacher's passive posture; the lesson ground to a halt. In conference, the
student ~as surprised that her posture had had such an effect. In subsequent
lessons, she was careful to face the group, lean forward and remain erect.
Once students became aware of the need to monitor all the channels which
carried messages, they consciously used cues to support their teaching task.
One student decided in the middle of her lesson that the group was too far
away; she asked them to move closer and explained why. Children's
responses and involvement increased after the move. Another beginning
teacher noticed that she had shifted body position and gaze while asking one
child to explain an answer; as a result, she was facing that child, with the rest
of the group to her side. The other children began talking among themselves
and moving in their seats. As soon as she had shifted back so that her body
and gaze included the whole group, the other children resumed listening and
looking at the student teacher.
L a n g u a g e use. A s students became more familiar with their classrooms,

and gained confidence, some were able to begin monitoring their choice of
language in lessons. Although language such as "Stop talking," and "Sit
d o w n , " sent clear verbal messages, students were puzzled by children's
negative responses to these directives. In conferences, we discussed many
ways to say the same thing; students then experimented with language in
their lessons. For example, they learned to use indirect messages, such as
"Sharon has her hand up" to say the same thing as "Stop talking," thus
signaling the preferred behavior and providing specific information about
" h o w " to participate.
As the term progressed, the students also began personalizing their

questioning; in contrast, their earlier questions had been what Mehan
(1979b) and Morine-Dershimer and T e n e n b e r g have called pseudo-ques-
tions, e.g., " A good listener--what must he or she d o ? " Students modified
questions to ask such things as, " Y o u ' r e the first little pig; what will you
say?" As they noticed the increased participation that resulted from these
changes in questioning styles, they realized how language could play a
powerful role in the success of their lessons.
Summary. This brief account of one supervisor's efforts to use the
research on teaching and learning as linguistic processes suggests some
potential directions for future applications. Experience showed that
students benefited both in m o r e successful teaching and in their own
understanding of that success. Students who gain a different conceptual
perspective on their own teaching may be able to continue teaching
themselves, long after they leave colleges of education. In other words, work
on teaching as a linguistic process provides (a) ways to describe teaching,
and (b) a language for this description that has its theoretical base in the
nature of face-to-face interactions and their consequences.
FOOTNOTES
' Because these studies will be referred to throughout the text and in the tables, the dates will
not always be included hereafter. These citations are representative of the 10 projects. The
reader is referred to the reference list for additional citations.
2The analogy of the microscope was suggested by Deborah Smith (University of Delaware),
who was a research assistant on this project.
APPENDIX. Representative Findings
A. Representative Findings on Patterns of Language Use

General Finding Source from Core Studies
1. Nominations: Differences in use by context DeStefano & Pepinsky
1. overall
(a) nominations to respond = 77%
(b) nominations in opening bids = 23%
2. nominations during reading = 90%
3. few opening bids occurred that were to
initiate talk in classrooms (teacher).
2. Discourse was found to be dominated by Morine-Dershimer &
teacher-initiated exchanges with a relati- Tenenberg
vely high density of speech acts in opening
moves.
1. opening moves by teachers were do- Morine-Dershimer &

minated by "directive management" Tenenberg
and "non-participant informants"
2. dual purposes served Guzman, 1980
(a) enforce order and structure
(b) illustrate and make clear lesson or
exercise that follows
3. Teacher talk centered on states or actions DeStefano & Pepinsky
with a relatively heavy accompanying
references to objects. There was little
teacher talk about processes.
4. Information giving in classrooms was Morine-Dershimer &
usually in response to a teacher's question Tenenberg
and rarely resulted in complete sentences.
In play settings, information giving was
usually volunteered and mostly given in
"complete sentences."
5. Teachers evaluate student language ability Cook-Gumperz & Worsley,
using an adult model or an "ideal" model. 1981
This evaluation process is often uncon- Cook-Gumperz & Green,
sciously done but has an influence on in press
students who have developed a sense of Michaels & Cook-Gumperz,
narrative different from the "ideal" that 1979
teacher has or the adult model.
6. Students of different language traditions Collins & Michaels, 1980
use language for the same purpose but use Hymes
different forms and cohesive structures. Michaels, 1981a, 1981b
Michaels & Cook-Gumperz,
1979
B. Representative Findings on Rules for Speaking

Salient Feature Source from Core Studies
1. Explicit signals are found in literal meaning. Erickson, 1982b
2. Implicit features signal Erickson, 1982b
1. nonverbal changes in distance
2. changes in posture Erickson, 1982b
3. contextualization cues include Erickson & Shultz, 1981
(a) syntax Cook-Gumperz & Gumperz,
(b) lexicon 1976
(c) stylistic register Gumperz & Tannen, 1980
(d) prosody (pitch, stress, intonation,
etc.)
(e) gaze
(f) body movement
(g) postural position
(h) interpersonal distance
(i) tying between messages
(j) rhythm
These cues point to meaning, provide
redundancy of information and/or addition-
al messages.
3. Tacit: students and observers must actively All researchers (e.g., see
interpret meaning from interactions. DeStefano & Pepinsky)
4. Rights and obligations to speak for any Merritt & Humphrey
individual related to the question "about
what?"
5. Conversational accessibility is regulated. Merritt & Humphrey
Michaels & Cook-Gumperz,
1979
6. Norms or expectations for participation are Borman & Barrett, 1981
constructed as part of the ongoing activity. DeStefano & Pepinsky
Erickson, 1982c
Erickson & Shultz, 1981
Hymes, 1981a, 1981b
Merritt & Humphrey
C. Representative Findings on Peer Learning

and Language Use
1. Peer culture constrains student participa- Schafer, 1981
tion in classroom interactions by fourth
grade.
2. Peer learning/teaching is a context for Cooper, Ayers-Lopez, &
intellectual development. Marquis
Morine-Dershimer &
Tenenberg
3. In play groups, the most frequently oc- Morine-Dershimer &
curring language, in order, was: Tenenberg
1. information giving
2. directing/influencing.
4. The most reported language in interviews, Morine-Dershimer &
in order, was: Tenenberg
1. directing/influencing
2. information giving.
5. In didactic situations, the students used a Cooper, Ayers-Lopez, &
systematic set of strategic interaction Marquis
patterns that were associated with effec-
tiveness in peer learning that was also
developmental (kindergarten and second
grade)
1. orient partner
2. manage behavior
3. instrumental statements
4. responses
5. evaluation and feedback
6. Directives were more effective in coopera- Cooper, Ayers-Lopez, &
tive peer learning tasks. No such effect Marquis
was found for the didactic task.
7. In cooperative tasks, effectiveness was Cooper, Ayers-Lopez, &
achieved when Marquis
1. children pointed
2. children showed blocks to partners
3. children used directives
4. children labeled blocks
5. children made evaluative comments.
8. Low frequencies occurred in the use of Cooper, Ayers-Lopez, &
praising or critical remarks to partners in Marquis
peer learning situations.
9. Seven-year-olds used pacing and man- Cooper, Ayers-Lopez, &
aging embedded in teaching and learning Marquis
sequences to meet their goals.
10. The use of relevant comments and Cooper, Ayers-Lopez, &
specificity were high ly associated with the Marquis
success of teaching episodes.
11. Even five-year-olds seemed able to repair Cooper, Ayers-Lopez, &
the ambiguity of nonspecific bids. Marquis
12. Repeats and reformulations are discourse Merritt & Humphrey
resources that children use extensively in
their replayings (done when no response
is given to initiations).
13. At first, second-, third-grade levels, repeti- Merritt & Humphrey
tions and reformulations occur more
frequently when students are trying to
engage other students than when trying to
engage the teacher.
14. Students report watching other students Morine-Dershimer, Galluzzo,
to & Tully, 1981
1. find out the answer when they don't Morine-Dershimer, Galluzzo,
know the answer & Tully, 1981
2. check their answer when they are not Borman, 1979
sure 8orman et al.
3. in informal situations (e.g., playground)
to learn how to play games and what to
do.
D. Representative Findings of Question Asking and Perception
of Questions
1. Pupils and teachers agreed that teachers Morine-Dershimer &
asked questions in order to teach or to tell. Tenenberg
2. Pupils reported that mothers asked ques- Morine-Dershimer &
tions because they wanted to know, and Tenenberg
that they (students) answered questions in
school and home settings because "some-
one asked."
3. Questions and responses to questions Morine-Dershimer &
occurred infrequently in play-group set- Tenenberg
tings, and when they did occur appeared to
serve an attention-getting function rather
than an informing function.
4. Among both age groups (kindergarten and Cooper, Ayers-Lopez, &
second grade), successful learners used Marquis
questions and referred to materials on the
problem involved in peer teaching situa-
tions.
5. Classroom questioning is one area (see MorineDershimer &
attention for another) where rules are least Tenenberg
clear.
6. In teacher initiated exchanges, pupil re- Morine-Dershimer &
sponses were more salient to both pupils Tenenberg
and teachers, even though teacher ques-
tions occurred somewhat more frequently.
7. Pupil responses to questions were reported Morine-Dershimer &
more often when Tenenberg
1. questions occurred in conjunctive
cycles (p < .0017) (conjunctive cycles
are two cycles which are tied together
because the same question is asked of
more than one student--horizontal
questions)
2. question cycles that occurred in em-
bedded cycles (p < .00013) (embedded
cycles are those in which one student is
asked a series of questions for clarifica-
tion or evaluation within a general
cycle).
E. Representative Findings on the Perception of General

Language Use for Teachers and Students
1. Teachers tended to report a question in Morine-Dershimer &
conjunction with a pupil response, or a Tenenberg
series of responses, while pupils tended to
report pupil responses in isolation from the
question that it elicited when asked to
report what they remembered hearing or
what was said.
2. Pupils perceived clear differences for rules Morine-Dershimer &
of discourse across the settings of home, Tenenberg
play, and classroom lesson.
3. It is clear that pupil perceptions of language Morine-Dershimer &
shifted as the social context changed from Tenenberg
the formal setting of the lesson to the
informal settings of families and play
groups.
4. Frequency of participation in class discus- Morine-Dershimer &
sions contributed significantly (p < .0001) Tenenberg
to the explained variance in final reading
achievement.
5. Defining question (students) as informative Morine-Dershimer &
contributed significantly to the explained Tenenberg
variance in pupil participation in classroom
discourse (p < .01).
6. There were strong classroom differences in Morine-Dershimer &
pupil perceptions of the functions of teacher Tenenberg
questions (p < .001) and of teacher praise
(p < .05) in lessons, and these differences
correspond to differences in teachers' use
of questions and praises, as identified in
analysis of classroom language.
7. There was a significant relationship be- Morine-Dershimer &
tween pupil perceptions of the functions of Tenenberg
questions in lessons and their "composite
concurrent classroom status" (p < .025).
F. Representatlve Findings on the Nature of Attentlon

1. Attention requirements differ within an May, 1981
activity, for example
1. teaching
2. direction-giving
3. independent work
4. tests
5. free time
2. Students' style of verbal interaction can May, 1981
mask their attention to task.
3. Teacher bases his or her inference about May, 1981
attention on the basis of observed cues
1. orientation of students' bodies
(a) eye gaze
(b) body movement
(c) body orientation
2. movement while teacher is talking
3. interference with activities of others
4. talking
5. inability to respond.
4. Teachers feel that they have the ability to May, 1981
determine when students are paying
attention, but they do not claim to always
be correct.
5. Teachers rank various evidence of inat- May, 1981
tention and place greater faith in this
evidence than in test results
1. ability to follow attention of students--
this ability is based on more than
behavior of students
2. on moment-to-moment indications of
attention.
6. Not all learning requires constant atten- May, 1981
tion
7. Students can give attention without un- May, 1981
derstanding the content or activity.
8. The amount of attention required depends May, 1981
on the degree the teacher feels is
necessary; therefore, the rule for attention
giving is not constant.
9. Teachers let attention go and pull it in May, 1981
which leads to different tolerance accord-
ing to
1. activity
2. individual child
3. participation structure
4. teacher
10. Teachers ignore two types of inattention May, 1981
1. whatever can be ignored
2. whether or not the teachers want the
instance of inattention to be raised to
the public level.
11. Types of inattention behaviors likely to be May, 1981

raised to the public level
1. loud voice
2. lasts a long time (duration)
3. draws others in
4. one student's voice can be identified
5. student moves from one place to the
next without positive sanctions.
12. A difference exists between perception of May, 1981
inattention and public notice; therefore
consideration of what counts as attention
or inattention depends on teacher per-
ception, teacher goal for lesson, group, or
individual.
13. Teacher and students do not begin with a DeStefano & Pepinsky
shared understanding of what constitutes Merritt, 1982
task or rules. The teacher's role is to direct
the focus of attention on task and rules
and to lay out format expectations
1. if the format expectations are not laid
out, students may "borrow" formats
from another learning environment as
a frame for the current activity and
therefore perform inappropriately
2. problems can also exist when students
read only part of the cues (one
modality--e.g., verbal and not both
verbal and nonverbal)
14. Fourth graders gave more attention to Morine-Dershimer &
comments of high achievers, frequent Tenenberg
participants, and pupils high in status with
the teachers. Third graders showed no
significant differences in ratios of attention
based on any of these variables.
15. Boys were not attended to more than girls. Morine-Dershimer &
Tenenberg
16. High peer status pupils were not attended Morine-Dershimer &
to more than low peer status pupils. Tenenberg
17. Ratios of attention were highest for pupils Morine-Dershimer &
who participated frequently in class dis- Tenenberg
cussions.
18. Ratios of attention were higher for com- Morine-Dershimer &
ments of pupils who were high in entering Tenenberg
reading achievement and lower for pupils
low in entering reading achievement,
19. There were no ethnic differences in Morine-Dershimer &

attention to either teacher questions or Tenenberg
pupil responses.
20. Pupils high in peer status attended more Morine-Dershimer &
to pupil response than pupils of middle or Tenenberg
low peer status.
21. High achievers in reading attended more Morine-Dershimer &
to teacher questions than low achievers (p Tenenberg
< .001).
22. Fourth graders attended less than second Morine-Dershimer &
and third graders to teacher questions (p Tenenberg
< .05) and attended more than second
and third graders to pupil answers (p <
,02).
23. Ratios of attention to pupil responses Morine-Dershimer &
were significantly related to the type of Tenenberg
question being responded to (p < .001).
Pupils reported hearing responses to
lower convergent and higher divergent
questions most frequently.
24. Pairing attention-focusing statements Cooper, Ayers-Lopez,
with other moves during peer instruction & Marquis
was highly associated with positive out-
comes. Second graders used attention-
focusing to precede their questions.
G. Representative Findings on Teachers' Use of Praise

1. The teacher used positive evaluations 2 to DeStefano & Pepinsky
1 over negative ones.
2. Students' replies were evaluated or ac- DeStefano & Pepinsky
cepted, but not ignored.
3. Pupil responses which drew teacher Morine-Dershimer &
praise were reported as heard more Tenenberg
frequently than responses which did not.
4. Teachers' strong praise was heard more Morine-Dershimer &
frequently by students even though its Tenenberg
frequency of occurrence was less than
other forms of talk.
5. Teachers use the same response forms to Merritt & Humphrey
sanction students during individual work
time as they do in whole group lessons.
6. Pupils and teachers agreed that teacher Morine-Dershimer &

praise was given because it was deserved Tenenberg
(e.g., "for good ideas").
7. Defining praise as deserved (students) Morine-Dershimer &
was significantly related to higher partici- Tenenberg
pation in class discussion (p < .025).
8. Verbal praise occurred rarely in video- Morine-Dershimer &
tapes of both family conversations and Tenenberg
play groups.
9. There were significant relationships be- Morine-Dershimer &
tween pupil perceptions of praise and Tenenberg
each of the concurrent classroom status
measures separately:
1. entering reading achievement (p <
.01 )
2. peer status (p < .05)
3. status with teacher (p < .005)
10. Students of the higher classroom status Morine-Dershimer &
viewed questions as instructional and Tenenberg
praise as deserved.
11. Students of lower classroom status tend- Morine-Dershimer &
ed not to provide any definition of the Tenenberg
functions of these language events (see
item 10).
H. Representative Findings on Teacher's Use of Sanctions

1. Teachers' sanctions in whole group les- Merritt & Humphrey
sons fall into six types:
1. placement sanctions: placement of ut-
terances by students occurs in wrong
place in lessons (e.g., within another
student's turn)
2. defivery santions: manner (volume or
tempo) of utterance is inappropriate
3. responsive sanctions: placement is in-
correct but teacher's response indicates
awareness of content of student's utter-
ance
4. double-takes: teacher first sanctions
utterance because placement is incor-
rect, then revises action and responds
due to emergency signaled in utterance
5. curt responses: teacher responds to

content of student's utterance, but curt-
ness indicates that he or she is unhappy
with its placement
6. behavioralsanctions: teacher is unhap-
py with the student's behavior and not
with any particular utterance
2. Placement sanctions were overwhelmingly Merritt & Humphrey
more frequent than any other kind
(250 placement, 17 behavioral, 11 respon-
sive, 8 curt, 6 double-takes, 5 delivery).
3. During individual worktime, four categories
of sanctions can be distinguished:
1. squelch sanctions: "SHHH!"
2. attention deferrals: "Just a minute, let
her finish her sentence."
3. general behavior sanctions: "You are
behind; you have not paid attention. Sit
down."
4. rechannellings: "If you would get your
math checked with Connie, you'd get out
faster."
4. Four main ways of sanctioning students Hrybyk & Farnham-Diggory
were found in second grade:
1. behavior modification point system with
prizes for daily and monthly good
behavior (as a rule, the students kept
close watch on who were "pumpkin
persons")
2. loss of center privileges
3. banishment to a desk at the back of the
room
4. penalty points in classroom games
5. biased turn-giving along dimensions of
perceived intelligence and cooperation,
ability to listen and follow directions
5. When school personnel noted a peer group Hrybyk & Farnham-Diggory
(e.g., a clique), they set out to control it.
REFERENCES
Agar, M. The professional stranger. New York: Academic Press, 1980.
Au, K. Participation structures in a reading lesson with Hawaiian children.
Anthropology and Education Quarterly, 1980, 11(2), 91-115.
Bales, R., & Strodtbeck, R. Phases in group problem-solving. In E. J. Amidon & J.
B. Hough (Eds.), Interaction analysis: Theory, research and application. Reading,
Mass.: Addison-Wesley, 1967.
Bellack, A., Kliebard, H., Hyman, R., & Smith, F., Jr. The language of the
classroom. New York: Teachers College Press, 1966.
Bloome, D. An ethnographic approach to the study of reading activities among black
junior high school students. Unpublished doctoral dissertation, Kent State
University, 1981.
Bloome, D., & G ree n, J. L. Capturing social contexts of reading for urban junior high
school youth in home, school, and comrnunity settings (Final Report, NIE
G-80-0128). Washington, D.C.: National Institute of Education, 1981.
Bloome, D., & Green, J. L. The social contexts of reading: A multidisciplinary
perspective. In B. Hutson (Ed.), Advances in reading~language research (Vol. 1).
Greenwich, Conn.: JAI Press, 1982.
Borman, K. Children's interactions on playgrounds. Theory Into Practice, 1979,
18(4), 251-257.
Borman, K., & Barrett, D. Negotiating playground games (Final Report, NIE
Borman, K., Piazza, S., Barrett, D., & Sheoran, P. Children's interpersonal
relationships, playground games, and social cognitive skills (Final Report, NIE
G-79-0123). Washington, D.C.: National Institute of Education, 1981. (a)
Borman, K., & Piazza, S., Barrett, D., & Sheoran, P. The ecology of children's play
(Final Report, NIE G-79-0123). Washington, D.C.: National Institute of
Education, 1981. (b)
Brophy, J.. & Evertson, C. Learning from teaching: A developmental perspective.
Boston: Allyn & Bacon, 1976.
Cahir, S. R. Activities between and within activities: Transitions. Unpublished
doctoral dissertation, Georgetown University, 1978.
Cahir, S. R., & Kovacs, C. Exploring functional language. Washington, D. C.:
Center for Applied Linguistics, 1981.
Carey, R.. Harste, J., & Smith, S. L. Contextual constraints and discourse
processes: A replication study. Reading Research Quarterly, 1981,16(2), 201-212.
Carrasco. R. Expanded awareness of student performance: A case study in applied
ethnographic monitoring in a bilingual classroom. In H. T. Trueba, G. T.
Gutherie. & K. Au (Eds.), Culture and the bilingual classroom: Studies in
classroom ethnography. Rawley, Mass.: Newbury House, 1981.
Carrasco, R., Acosta, C., De La Torre-Spencer, S. Language use, lesson engagement
and participation structures: A micro-ethnographic analysis of two language arts
lessons in a bilingual first grade classroom with policy and practical implications.
Paper presented at the annual meeting of the American Educational Research
Association, Los Angeles, April 1981.
Carrasco, R., Vera, A., & Cazden, C. Aspects of bilingual students' communicative
competence in the classroom: A case study. In R. Duran (Ed.), Reading latino
language and communicative behavior. Norwood, N.J.: Ablex, 1981.
Cazden, C. B. Learning to read in classroom interaction. In J. T. Guthrie (Ed.),
Reading comprehension and education. Newark, Del.: International Reading
Association, 1981.
Cazden, C. B. Language in education: Variation in the teacher-talk register, In J.

Alatis & R. Tucker (Eds.),'Language in public life (Georgetown University
Round Table on Languages and Linguistics, 1979). Washington, D.C.:
Georgetown University Press, 1979.
Cazden, C. B., Carrasco, R., Guzman, A. A., & Erickson, F. The contribution of
ethnographic research to bicultural bilingual education. In J. Alatis (Ed.), Current
issues in bilingual education (Georgetown University Round Table on Language
and Linguistics, 1980). Washington, D.C.: Georgetown University Press, 1980.
Cazden, C., John, V. & Hymes, D. Functions of language in the classroom. New
York: Teachers College Press, 1972.
Cherry Wilkinson, L, Analysis of teacher-student interaction-expectations com-
municated by conversational structure. In J. L. Green & C. Wallat (Eds.),
Ethnography and language in educational settings. Norwood, N.J. : Ablex, 1981.
Cherry Wilkinson, L. (Ed.). Communicating in classrooms. New York: Academic
Press, 1982.
Cherry Wilkinson, L., & Calculator, S. Effective speakers: Students' use of language
to request and obtain information and action in the classroom. In L. Cherry
Wilkinson (Ed.), Communicating in classrooms. New York: Academic Press,
1982. (a)
Cherry Wilkinson, L., & Calculator, S. Requests and responses in peer-directed
reading groups. American Educational Research Journal, 1982, 19, 107-120. (b)
Cherry Wilkinson, L., & Dollaghan, C. Peer communication in first grade reading
groups. Theory Into Practice, 1979, 18(4), 267-274.
Cole, M., Dore, J., Hall, W., & Dowley, G. Situation and task in children's talk.
Discourse Processes, 1978, 1, 119-176.
Cole, M., Griffin, P., & Newman, D. Mid-quarter reports (NIE G-78-0159).
Washington, D. C.: National Institute of Education, 1978-81.
Cole, M., Griffin, P., & Newman, D. Personal communication, October 1981.
Cole, M., Griffin, P., & Newman, D. They're all the same in their own way.
(Mid-quarter report, NIE G-78-0159). Washington, D.C.: National Institute of
Education, 1979.
Cole, M., Hood, L., & McDermott, R. P. Ecological niche picking: Ecological
invalidity as an axiom of experimental cognitive psychology. Unpublished
manuscript, University of California, San Diego, and The Rockefeller University,
1978.
Collins, J. Differential treatment in reading instruction. In J. Cook-Gumperz, J.
Gumperz, & H. Simons, School-Home Ethnography Project (Final Report, NIE
Collins, J., & Michaels, S. The importance of conversational discourse strategies in
the acquisition of literacy. Berkeley, Calif.: Berkeley Linguistic Society, 1980.
Cook-Gumperz, J. Interactive styles in instructional talk. In J. Cook-Gumperz, J.
Gumperz, & H. Simons, School-Home Ethnography Project (Final report, NIE
G-78-0082). Washington, D.C.: National Institute of Education, 1981. (a)
Cook-Gumperz, J. Language at home and at school: Focusing on discourse. In J.
Cook-Gumperz, J. Gumperz, & H. Simons, School-Home Ethnography Project
(Final Report, NIE G-78-0082). Washington, D.C.: National Institute of
Education, 1981. (b)
Cook-Gumperz, J. Persuasive talk: The social organization of children's talk. In J. L.
Green & C. Wallat (Eds.), Ethnography and language in educational settings.
Norwood, N.J.: Ablex Publishing, 1981. (c)
Cook-Gumperz, J. & Corsaro, W. Social-ecological constraints on children's
communicative strategies. In J. Gumperz & J. Cook-Gumperz, Papers on
Language and Context (Working Paper No. 46). Berkeley: Language Behavior
Research Laboratory, University of California, 1976.
Cook-Gumperz, J., & Green, J. L. Children's sense of story: Influences on
children's storytelling ability. In D. Tannen (Ed.), Coherence in spoken and
written language. Norwood, N.J.: Ablex, in press.
Cook-Gumperz, J., & Gumperz, J. Context in children's speech. In J. Gumperz & J.
Cook-Gumperz, Papers on Language and Context (Working Paper No. 46).
Berkeley: Language Behavior Research Laboratory, University of California,
1976.
Cook-Gumperz, J., & Gumperz, J. Communicative competence in educational
perspective. In L. Cherry Wilkinson (Ed.), Communicating in classrooms. New
York: Academic Press, 1982.
Cook-Gumperz, J., Gumperz, J., & Simons, H. D. Language at school and home:
Theory, methods, and preliminary findings (Mid-quarter Report, NIE G-78-
0082). Washington, D. C.: National Institute of Education, 1979.
Cook-Gumperz, J., Gumperz, J., & Simons, H. D. School-home ethnography
project (Final Report, NIE G-78-0082). Washington, D.C.: National Institute of
Education, 1981.
Cook-Gumperz, J., & Worsley, L. Report on the narrative discourse study. In J.
Cook-Gumperz, J. Gumperz, & H. D. Simons, School-Home Ethnography
Project (Final Report, NIE G-78-0082). Washington, D.C.: National Institute of
Education, 1981.
Cooper, C., Ayers-Lopez, S., & Marquis, A. Children's discourse in cooperative and
didactic interaction: Developmental patterns in effective learning (Final Report,
N1E G-78-0098). Washington, D.C.: National Institute of Education, 1981.
Cooper, C., Marquis, A., & Ayers-Lopez, S. Developmental patterns in classroom
peer learning: Characteristics of competence. Unpublished manuscript, University
of Texas, Austin, 1980.
Cooper, C., Marquis, A., & Ayers-Lopez, S. Peer learning in the classroom: Tracing
developmental patterns and consequences of children's spontaneous interaction.
In L. Cherry Wilkinson (Ed.), Communicating in classrooms. New York:
Academic Press, 1982.
Corsaro, W. Entering the child's world: Research strategies for field entry and data
collection. In J. L. Green & C. Wallat (Eds.), Ethnography and language in
educational settings. Norwood, N.J.: Ablex, 1981.
Crystal, D. A first dictionary of linguistics and phonetics. Boulder, Colo. : Westview
Press, 1980.
DeStefano. J., & Pepinsky, H. The learning of discourse rules of culturally different
children in first grade literacy instruction (Final Report, NIE G-79-0032).
Washington, D.C.: National Institute of Education, 1981.
DeStefano, J., Pepinsky, H., & Sanders, T. Discourse rules for literacy learning in a
classroom. In L. Cherry Wilkinson (Ed.), Communicating in classrooms. New
York: Academic Press, 1982.
Dunkin, M. J., & Biddle, B. J. The study of teaching. New York: Holt, Rinehart, &
Winston, 1974.
Eder, D. The social context of learning: A micro-analysis of ability grouping. Paper
presented at a meeting of the American Sociological Association, New York,
August 1980.
Eder, D. Differences in communicative styles across ability groups. In L. Cherry
1982.
Green: Teaching as s Linguistic Process 247
Elkind, D. Child and society. New Jersey: Oxford University Press, 1979.
Elliott, J. Developing hypotheses about classrooms from teachers' practical
constructs. Grand Forks, N.D.: North Dakota Study Group, 1976.
Erickson, F. Some approaches to inquiry in school-community ethnography.
Anthropology and Education Quarterly, 1977, 8(2), 58-69.
Erickson, F. On standards of descriptive validity in studies of classroom activities.
East Lansing, Mich.: Institute for Research on Teaching, University of Michigan,
1979.
Erickson, F. Personal communication, July 1981.
Erickson, F. Audio-visual documentation of everyday life in schools: A handbook of
methods and resources. East Lansing, Mich.: Institute for Research on Teaching,
University of Michigan, 1982. (a)
Erickson, F. Classroom discourse as improvisation: Relationships between
academic task structure and social participation structure in lessons. In L. Cherry
1982. (b)
Erickson, F., Cazden, C., & Carrasco, R., & Guzman, A. Social and cultural
organization of interaction in classrooms of bilingual children (Mid-quarter
Report, NIE G-78-0099). Washington, D.C.: National Institute of Education,
1978-81.
Erickson, F., & Mohatt, G. The social organization of participation structures in two
classrooms of Indian students. In G. Spindler (Ed.), The ethnography of
schooling: Educational anthropology in action. New York: Holt, Rinehart, &
Winston, 1981.
Erickson, F., & Shultz, J. When is a context? Some issues and methods in the
analysis of social competence. In J. L. Green & C. Wallat (Eds.), Ethnography
and language in educational settings. Norwood, N.J. : Ablex Publishing, 1981.
Erickson, F., & Shultz, J. The counselor as gatekeeper. New York: Academic Press,
1982.
Fiering, S. Commodore school: Unofficial writing. In D. Hymes, Ethnographic
monitoring project (Final report, NIE G-78-0038). Washington, D.C.: National
Institute of Education, 1981.
Florio, S. Learning how to go to school: An ethnography of interaction in a
kindergarten/first grade classroom. Unpublished doctoral dissertation, Harvard
University, 1978.
Florio, S., & Shultz, J. Social competence at home and at school. Theory Into
Practice, 1979, 18(4), 234-243.
Florio, S., & Walsh, M. The teacher as colleague in classroom research. Paper
presented at the annual meeting of the American Educational Research
Association, San Francisco, April 1976.
Frederiksen, C. Inference in preschool children's conversations--A cognitive
perspective. In J. L. Green & C. Wallat (Eds.), Ethnography and language in
educational settings. Norwood, N.J.: Ablex Publishing, 1981.
Garnica, O. Social dominance and conversational interaction--the Omega child in
the classroom. In J. L. Green & C. Wallat (Eds.), Ethnography and language in
Gearhart, M., & Newman, D. Learning to draw a picture: The social context of an
individual activity. Discourse Processes, 1980, 3, 169-184.
Genishi, C. Young children communicating in the classroom: Selected research.
Theory Into Practice, 1979, 18(4), 244-250.
Gilmore. P. Shortridge school and community: Attitudes and admissions to literacy.
In D. Hymes, Ethnographic monitoring project (Final report, NIE G-78-0038).
Gilmore. P., & Glatthorn, A. (Eds.). Children in and out of school: Ethnography
and education. Washington, D.C.: Center for Applied Linguistics, 1982.
Green, J. L. Pedagogical style differences as related to comprehension performance:
Grades one through three. Unpublished doctoral dissertation, University of
California, Berkeley, 1977.
Green. J. L.. & Harker, J. O. Gaining access to learning: Conversational, social, and
cognitive demands of group participation. In L. Cherry Wilkinson (Ed.),
Comm,micating in classrooms. New York: Academic Press, 1982.
Green, J. L., & Harker. J. O. Using language in classrooms: Communicative
demands and communicative consequences. In A. Jaggar & M. T. Smith-Burke
(Eds.), Kid-watching. Observing the language learner. Newark, Del.: Interna-
tional Reading Association, in press.
Green, J. L.. & Wallat. C, What is an instructional context? An exploratory analysis
of conversational shifts over time. In O. Garnica & M. King (Eds.), Language,
children and socieo'. New York: Pergamon, 1979.
Green, J. L.. & Wallat, C. Ethnography and language in educational settings.
Norwood, N.J.: Ab[ex Publishing, 1981. (a)
Green, J. L.. & Wallat, C. Mapping instructional conversations. In J. L. Green & C.
Wallat (Eds.), Ethnography and language in educational settings. Norwood, N.J. :
Ablex Publishing, 1981. (b)
Griffin, P. Personal communication, October 1981.
Griffin. P., Cole, M., & Newman, D. Locating tasks in psychology and education.
Discourse Processes, 1982, 5(2), 1 t 1-125.
Griffin, P., Newman, D., Cole, M. Activities, actions and formal operations: A
Vygotskian analysis of a Piagetian task. Paper presented at a meeting of the
International Society for the Study of Behavioral Development, Toronto, August
1981.
Gumperz, J. J. Sociocultural knowing in conversational inference. In M.
Saville-Troike (Ed.), 28th Annual Round Table Monograph Series on Language
and Linguistics. Georgetown. Va.: Georgetown University Press, 1977.
Gumperz, J. J. Conversational inference and classroom learning. In J. L. Green &
C. Wallat (Eds.), Ethnography and language in educational settings. Norwood,
N.J.: Ablex Publishing, 1981. (a)
Gumperz. J. J. Personal communication, July 1981. (b)
Gumperz, J. J., & Cook-Gumperz, J. Beyond ethnography: Some uses of
sociolinguistics for understanding classroom environments. In J. Cook-Gumperz,
J. Gumperz, & H. D. Simons, School-Home Ethnography Project. (Final Report,
NIE G-78-0082). Washington, D. C.: National Institute of Education, 1981. (a)
Gumperz, J. J., & Cook-Gumperz, J, Personal communication, July 1981. (b)
Gumperz, J. J.. Cook-Gumperz, J., & Micbaels, S. Personal communication, July
1981.
Gumperz, J. J., Cook-Gumperz, J.. & Simons, H. D. School-home ethnography
project (Mid-quarter report, NIE G-78-0(182). Washington, D.C.: National
Institute of Education, 1979.
Gumperz, J. J., & Herasimchuk, E~ The conversational analysis of social meaning:
A study of classroom interaction. In R. Shuy (Ed.), Sociolinguistics: Current
trends and prospects. Georgetown Universio, Monographs in Language and
Linguistics. Georgetown, Va.: Georgetown University Press. 1973.
Gumperz. J. J., & Hymes, D. Directions in sociolinguistics: Ethnography of
communication. New York: Holt, Rinehart, & Winston, 1972.

Guzman, A. A. A multidimensional ethnographic framework for studying classroom
organization and interaction. Boston, Mass.: Harvard University, 1981.
Guzman, A. A. Theoretical and methodological issues for the ethnographic study of
teachers' differential treatment of children in bilingual bicultural classrooms. Paper
presented at the annual meeting of the American Anthropological Association,
Washington, D.C., December 1980.
Guzman, A. A. A multidimensional ethnographic framework for studying classroom
organization and interaction. Qualifying Paper, Harvard University, 1981.
Halliday, M. A. K. Three aspects of children's language development: Learning
language, learning through language and learning about language. In Y.
Goodman, M. Haussler, & D. Strickland (Eds.), Oral and written language
development research: Impact on the schools. Urbana, I11.: National Council of
Teachers of English, 1982.
Heap, J. What counts as reading? Limits to certainty in assessment. Curriculum
Inquiry, 1980, 10(3), 265-292.
Hrybyk, M., & Farnham-Diggory, S. Children's groups in school: A developmental
case study (Final Report, NIE G-79-0124). Washington, D.C.: National Institute
of Education, 1981.
Hymes, D. Foundations in sociolinguistics. Philadelphia: University of Pennsylvania
Press, 1974.
Hymes, D. Ethnographic monitoring project (Final Report, N1E G-78-0038).
Washington, D.C.: National Institute of Education, 1981. (a)
Hymes, D. Language in education. Ethnolinguistic essays. Washington, D. C.:
Center for Applied Linguistics, 1981. (b)
Kendon, A., Harris, R., & Key, M. R. (Eds.). The organization of behavior in
face-to-face interaction. The Hague: Mouton, 1976.
Lussier, J. R. Shortridge school and community: Boys' verbal abilities, In D. Hymes
(Ed.), Ethnographic monitoring project (Final Report, NIE G-78-0038).
Lutz, F. Ethnography: The holistic approach to understanding schooling. In J. L.
Green & C. Wallat (Eds.), Ethnography and language in educational settings.
Norwood, N.J.: Ablex Publishing, 1981.
May, L. Spaulding school: Attention and styles of interaction. In D. Hymes (Ed.),
Ethnographic monitoring (Final report, NIE G-78-0038). Washington, D.C.:
National Institute of Education, 1981.
McDermott, R. P. Kids make sense: An ethnographic account of interactional
management of success and failure in one first-grade classroom. Unpublished
doctoral dissertation, Stanford University, 1976.
McDermott, R. P. Relating and learning: An analysis of two classroom reading
groups. In R. Shuy (Ed.), Linguistics and reading. Rawley, Mass.: Newbury
House, 1978.
Mead, M. Blackberry winter: My earl), 3,ears. New York: Pocket Books, 1975.
Mehan, H. Learning lessons. Cambridge, Mass.: Harvard University Press, 1979. (a)
Mehan, H. What time is it Denise?: Asking known information questions in
classroom discourse. Theory Into Practice, 1979, 18(4), 285-294. (b)
Mehan, H., Cazden, C., Coles, L., Fisher, S., & Maroules, N. The social
organization of classroom lessons. San Diego, Calif.: Center for Human
Information Processing, 1976.
Merritt, M. Distributing and directing attention in primary classrooms. L. Cherry
1982.
Merritt, M., & Humphrey, F. Teacher, talk and task: Communicative demands
during individualized instruction time. Theory Into Practice, 1979, 18(4), 298-303.
Merritt, M., & Humphrey, F. Service-like events during individual work time and
their contribution to the nature o f communication in primary classrooms (NIE
Michaels, S. 'Sharing time': Children's narrative styles and differential access to
literacy. In J. Cook-Gumperz, J. Gumperz, & H. D. Simons, School-Home
Ethnography Project (NIE G-78-0082). Washington, D.C.: National Institute of
Education, 1981. (a)
Michaels, S. Sharing time revisited. In J. Cook-Gumperz, J. Gumperz, & H. D.
Simons. School-Home Ethnography Project (NIE G-78-0082). Washington,
D.C.: National Institute of Education, 1981. (b)
Michaels. S., & Cook-Gumperz. J. A study of sharing time with first grade students:
Discourse narratives in the classroom. In Proceedings o f the Berkeley Linguistic
Society'. 1979, 5. 87-103.
Morine-Dershimer. G., Galluzzo, G., & Fagal, F. Rules o f discourse, classroom
status, pupil participation and achievement in reading: A chaining o f relationships
(Final Report, Part 3. NIE G-78-0161). Washington, D.C.: National Institute of
Education, 1981.
Morine-Dershimer, G., Galluzzo, G., & Tully, H. Who hears whom: Classroom
status variables and pupil attention to the comments o f other pupils. Paper
presented at the annual meeting of the American Educational Research
Association, Los Angeles, April 1981.
Morine-Dershimer, G., Lay-Dopyera, M.. & Graham, P. L. Attending to the
discourse ofclassroomates in play settings (Final Report, Part 5, NIE G-78-0161).
Morine-Dershimer. G., Ramirez, A., Shuy, R., & Galluzzo, G. How do we know?
(Final Report, Part 4, NIE G-78-016l). Washington, D.C.: National Institute of
Education, 1981.
Morine-Dershimer, G., & Tenenberg, M. Participant perspectives o f classroom
discourse (Final Report, Executive Summary'. NIE G-78-0161). Washington,
D.C.: National Institute of Education, 1981.
Morine-Dershimer, G., Tenenberg, M., & Shuy, R. Why doyou ask? (Final Report,
Part 2, NIE G-78-0161). Washington, D.C.: National Institute of Education,
1981.
National Institute of Education. Panel 5 report on teaching as a linguistic process in a
cultural setting. Washington, D.C.: Author, 1974.
National Institute of Education. Grants announcement. Washington, D.C. : Author,
1977.
Newman, D., Griffin. P., & Cole, M. Laboratory and classroom tasks: Social
constraints and the evaluation of children's performance. In J. Lave & B. Rogoff
(Eds.), Everyday cognition: Its development in social context. Cambridge, Mass.:
Harvard University Press, in press.
Olmeida-Williams, I. Functions o f code switching as a communicative strategy in a
Spanish-English bifingual classroom. Unpublished doctoral dissertation, Kent
State University, 1980.
Petitto, A. L. Culturally motivated goals and calculating devices: Learning arithmetic
in elementa O' schools. Paper presented at the annual meeting of the American
Educational Research Association, New York, March 1982.
Philips, S. U. Participant structures and communicative competence: Warm Springs
children in community, and classroom. In C. Cazden, V. John, & D. Hymes
(Eds.), Functions of language in the classroom. New York: Teachers College

Press, 1972.
Philips, S. U. The invisible culture." Communication in classroom and community on
the Warm Springs reservations. Unpublished doctoral dissertation, University of
Pennsylvania, 1974).
Piestrup, A. Black dialect interference and accommodation of reading instruction in
first grade. Monographs of the Language-Behavior Research Laboratory, 1973, 4
(Whole issue).
Sacks, H., Jefferson, G,, & Schegloff, E. A simplest systematic for the organization
of turn-taking for conversation. Language, 1974, 50(4), 696-738,
Schafer, J. Situation properties of competence: A case study of one child's behavior in
two educational settings. In J. Cook-Gumperz, J. Gumperz, & H. D. Simons,
School-Home Ethnography Project (NIE G-78-0082). Washington, D.C.:
Scollon, R., & Scollon, S. Cooking it up and boiling it down: Abstracts of
Athabaskan children's story retellings. In D. Tannen (Ed.), Coherence in spoken
and written language. Norwood, N.J.: Ablex Publishing, in press.
Sevigny, M. Triangulated inquiry: A methodology for the analysis of classroom
interaction. In J. Green & C. Wallat (Eds.), Ethnography and language in
Shultz, J., & Florio, S. Stop and freeze: The negotiation of social and physical space
in a kindergarten-first grade classroom. Anthropology and Education Quarterly,
1978, 10(3), 166-181.
Sinclair, J., & Coulthard, M. Towards an analysis of discourse." The English used by
teachers and pupils. England: Oxford University Press, 1975.
Simons, H. D., & Gumperz, J. J. Language and communication in school
performance. In J. Cook-Gumperz, J. Gumperz, & H. D. Simons, School-Home
Ethnography Project (Final Report, NIE G-78-0082). Washington, D.C.:
Simons, H. D., & Murphy, S. Spoken language strategies and reading acquisition. In
J. Cook-Gumperz, J. Gumperz, & H. D. Simons, School-Home Ethnography
Project (Final Report, NIE G-78-0082). Washington, D.C.: National Institute of
Education, 1981.
Smith, D. Personal communication, January 1982.
Smith, B. O., & Meux, M. A study of the logic of teaching. Urbana, I[l,: University of
Illinois Press, 1970.
Spindle r, G. The ethnography of schooling: Educational anthropology in action. New
York: Holt, Rinehart, & Winston, 1981.
Spradley, J. Ethnographic interviewing. New York: Holt, Rinehart, & Winston,
1980. (a)
Spradley, J. Participant observation. New York: Holt. Rinehart, & Winston, 1980.
(b)
Steinberg, Z., & Cazden, C. Children as teachers--of peers and ourselves. Theory
Into Practice, 1979, 18(4), 258-266.
Stoffan-Roth, M. Conversational access gaining strategies in instructional contexts.
Unpublished doctoral dissertation, Kent State University, 1981. (a)
Stoffan-Roth, M. Shhh! The children are watching. Partnership, 1981,6(3), 12. (b)
Tannen, D. What's in a frame? Surface evidence for underlying expectations. In R.
Freed[e (Ed.), Advances in discourse processing (Vol. 2). Norwood, N.J. : Ablex
Publishing, 1979.
Tannen, D. A theory of conversational style. Paper presented for Psycholinguistic
Models of Production: An Interdisciplinary Workshop, University of Kassel,

West Germany, July 1980.
Tenenberg, M., Morine-Dershimer, G., & Shuy, R. What did anybody say? (Final
Report, Part 1, NIE G-78-0161). Washington, D.C.: National Institute of
Education, 1981.
Wallat, C., & Green, J. L. Social rules and communicative contexts in kindergarten.
Theory Into Practice, 1979, 18(4), 275-284.
Wallat, C., & Green, J. L. Construction of social norms. In K. Borman (Ed.), Social
life of children in a changing society. Hillsdale, N.J.: Lawrence Erlbaum
Associates, 1982.
Wallat, C., Green, J. L., Conlin, S. M., & Haramis, M. Issues related to action
research in the classroom: The teacher and researcher as a team. In J. L. Green &
C. Wallat (Eds.), Ethnography and language in educational settings. Norwood,
N.J.: Ablex Publishing, 1981.
Watkins, M. Shortridge school and community: Teacher-parent relationships. In D.
Hymes, Ethnographic Monitoring Project (Final Report, NIE G-78-0038).
Woods-Elliott, C. Commodore school: Teaching writing: One teacher's experience.
In D. Hymes, Ethnographic Monitoring Project (Final Report, NIE G-78-0038).

Green 1983 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Green 1983 PDF

Uploaded by

Copyright:

Available Formats

Chapter 6

Research on Teaching as a Linguistic Process:

In the last decade, several new approaches to the study of teaching-

Project Population General Participant Topic-centered Natural Experiments,'

(Continued on next page)

Project Population General Participant Topic-centered Natural Experiments/

tions tied to literacy written language

Erickson et al. (1978-81) 2 first grades Classroom: Periodic videotaping and

(Continued on next page)

Project Population General Participant Topic-centered Natural Experiments/

Erickson et al. (Con'd) Subset of individual chil- q

Hrybyk & Farnham-Dig- Second grade, private, Observation of Sociometric question-

Project Population General Participant Topic-centered Natural Experiments/

DavisD Two classes each, grades Focus: school-community

Project Population General Participant Topic-centered Natural Experiments/

(Continued on next page)

Project Population General Participant Topic-centered Natural Experiments/

(Continued on next page)

Project Population General Participant Topic-centered Natural Experiments/

Morine-Dershimer & 6. continuities and dis- 4. Organizing 3 x 5

THE EMERGING FIELD: AN HISTORICAL PERSPECTIVE

Consolidation of the Field: NIE's Role

teaching and had its own configuration of participants. Panel 5 focused on

Between Education and Other Disciplines

CONCEPTUAL BASIS OF THE FIELD

Extracting Common Elements

extracted from three sources: published papers, unpublished papers

Constructs that Define and Guide Teaching as a Linguistic

Teachers signal their theory of pedagogy from their X X X X 4

Comprehension as an Inferencing Process, (e) The Classroom as a

Face-to-Face Interaction: A Rule-governed Phenomenon

response be predicted with certainty. For various reasons, an individual can

items and nonverbal aspects such as gesture, facial expression, proxemic

Meaning as Context Specific

from the consideration that no particular conversational behavior exists in a

Comprehension as an Inferencing Process

face-to-face interactions. However, the works of Cole, Newman, and

confirm these products of conversations. These cues are also open to

interested in certain aspects of classroom processes must go beyond the

This section explores the structural aspects of classrooms that help

for general participation, recent research has shown that teachers

The Teacher and the Communicative Process

tion is subject to the rules and expectations of conversation. Classroom

initiate-respond-evaluate sequence (Mehan, 1979a; Mehan et al., 1976), or

METHODOLOGY: A PRODUCT OF THE RESEARCH

Capturing Complex Processes: The General Approach

relationships between language and cognitive processes. This work,

playgrounds, children's interactions on playgrounds, and children's social

General Structural Components of the Research Designs

Cook-Gumperz et al.) utilized all three categories in planning and

FIGURE 1. Design of Study: School~Home

STAGE I: Framing and Planning the Project Components

Mental Frame-Mental Grid Participant Observation

Ethnography Project (Cook-Gumperz et aL)

STAGE I1: Data Collection-General Participant Observation

Participant Observation I: Explicit Definition of Behav-

Participant Observation I1:

(Continued on next page)

STAGE II1: Topic-centered Participant Observation

Participant Observation II1: Participant Observation IV:

STAGE IV: Inferencing/Hypothesis STAGE V: Natural Experiments

Observe discourse strategies Plan and execute natural ex-

V. Home data collection