American Educational Research

Journal
http://aerj.aera.net

Characteristics of Classroom Mathematics Traditions: An Interactional

Analysis
Paul Cobb, Terry Wood, Erna Yackel and Betsy McNeal
Am Educ Res J 1992 29: 573
DOI: 10.3102/00028312029003573

The online version of this article can be found at:

http://aer.sagepub.com/content/29/3/573

Published on behalf of

American Educational Research Association

and

http://www.sagepublications.com

Additional services and information for American Educational Research Journal can be found
at:

Email Alerts: http://aerj.aera.net/alerts

Subscriptions: http://aerj.aera.net/subscriptions

Reprints: http://www.aera.net/reprints

Permissions: http://www.aera.net/permissions

Citations: http://aer.sagepub.com/content/29/3/573.refs.html

>> Version of Record - Jan 1, 1992

What is This?

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 American Educational Research Journal
Fall 1992, Vol. 29, No 3, pp. 573-604

Characteristics of Classroom Mathematics

Traditions:
An Interactional Analysis
Paul Cobb, Terry Wood, Erna Yackel, and Betsy McNeal
Purdue University
In this paper, we attempt to clarify what it means to teach mathematics
for understanding and to learn mathematics with understanding. To this
end, we present an interactional analysis of transcribed video recordings
of two lessons that occurred in different elementary school classrooms.
The lessons, which are representative of a much larger data corpus, were
selected because both focus on place value numeration and involve the
use of similar manipulative materials. The analysis draws on Much and
Shewder's (1978) identification of five qualitatively distinct types of
classroom norms and pays particular attention to the mathematical ex-
planations and justifications that occurred during the lessons. In one
classroom, the teacher and students appeared consistently to constitute
mathematics as the activity of following procedural instructions in the
course of their moment by moment interactions. The analysis of the other
classroom indicated that the teacher and students constituted mathemati-
cal truths as they coconstructed a mathematical reality populated by ex-
perientially real, manipulable yet abstract mathematical objects. These
and other differences between mathematical activity in the two classrooms
characterize two distinct classroom mathematics traditions, one in which
mathematics was learned with what is typically called understanding and
the other in which it was not.

PAUL COBB is now a professor of mathematics education at Peabody College,

Box 330, Vanderbilt University, Nashville, TN 37203. He specializes in mathematical
learning in the context of classroom social interaction.
TERRY WOOD, an assistant professor of elementary mathematics education in
the Department of Curriculum and Instruction, Purdue University, West Lafayette,
IN 47907, specializes in teacher education and elementary mathematics education.
ERNA YACKEL is an associate professor of mathematics education in the Depart-
ment of Mathematics, Computer Science, and Statistics at Purdue University-
Calumet, Hammond, IN 46323. She specializes in small-group collaborative learn-
ing, and mathematics teaching and learning from a socioconstructivist perspective.
BETSY MCNEAL is now an assistant professor of education at the University of
Pennsylvania, Graduate School of Education, 3700 Walnut St., Philadelphia, PA
19104. She specializes in microsociological analyses of classroom interactions dur-
ing mathematics instruction.

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal

R ecent calls for reform in mathematics education in the United States

have focused on the need to promote instructional practices that facili-
tate what is generally called meaningful learning (e.g., National Council of
Teachers of Mathematics, 1989; National Research Council, 1989). The ex-
plication of what it might mean to teach for understanding and to learn
with understanding is clearly central in this ongoing reform effort. In this
regard, we find considerable value in several of the contrasts that have been
made between instructional situations that encourage meaningful learning
and those that do not. For example, Skemp (1976) discussed the distinc-
tions between relational and instrumental learning; Brown, Collins, and
Duguid (1989) contrasted authentic and unauthentic mathematical activity;
and Richards (1991) developed the notions of inquiry mathematics and
school mathematics. We attempt to explicate further the contrasts between
instructional situations that promote learning with understanding and those
that do not by developing an empirically grounded analysis of the mathe-
matics traditions established in different classrooms.
The approach we propose rests on the assumption that qualitative dif-
ferences in these classroom mathematics traditions can be brought to the
fore by analyzing teachers' and students' mathematical explanations and
justifications during classroom discourse. We illustrate this approach by first
developing our general orientation and clarifying our theoretical constructs.
We then examine transcripts of lessons from two classrooms, only one of
which can be said to foster meaningful learning. These two sample analyses
are subsequently used as a backdrop to reconsider what is meant by the
terms meaning and understanding. In the course of this discussion, we
find it useful to distinguish between the way the terms are used in two con-
texts, the explicitly political context of educational reform and the academic
context of mathematics education research. Finally, we consider claims
about the nature of mathematical development made from both the infor-
mation-processing perspective and a strong sociological perpective. Our
analysis of qualitatively distinct classroom mathematics traditions leads us
to conclude that both claims reflect unduly restricted conceptions of
mathematical activity.

Classroom Mathematics Traditions

Several scholars have investigated the process of learning and teaching
mathematics from a sociological perspective by analyzing what might be
called the mathematical culture of the classroom. In doing so, they have
introduced a variety of closely related theoretical notions including the
classroom discursive practice (Walkerdine, 1988), the tradition of classroom
practice (Solomon, 1989), and the classroom subculture or microculture
(Bauersfeld, Krummheuer, & Voigt, 1988; Voigt, 1985). In contrast to cog-
nitive analyses that focus on qualitative differences in the beliefs and mathe-
matical conceptions of individual teachers or students, sociological analyses
574

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
concentrate on the evolution of taken-as-shared or normative suppositions,
assumptions, and interpretations that make communication possible (Gergen,
1985). It is these often implicit consensual social and mathematical under-
standings that constitute what we will call the classroom mathematics tradi-
tion (Cobb, 1991).
As the term implies, a classroom mathematics tradition is in many ways
analogous to a scientific research tradition (Cobb, Wood, & Yackel, 1991).
For our purposes, it suffices to note that both are created by a community
and that both influence individuals' construction of scientific or mathema-
tical knowledge by constraining what can count as a problem, a solution,
an explanation, and a justification (Barnes, 1982; Knorr-Cetina, 1982; Lampert,
1988; Tymoczko, 1986). We take this as our starting point in that an analysis
of these four aspects of mathematical activity—problems, solutions, explana-
tions, and justifications—offers an empirically grounded way to characterize
the mathematics tradition established in any particular classroom. The work
of Much and Shweder (1978) is relevant in this regard in that they identi-
fied five qualitatively distinct types of normative activity by analyzing how
teachers and students account for their actions. We first outline their analy-
tical approach and attempt to extend it by considering both explanations
and justifications as they pertain specifically to mathematical activity. Next,
we argue that explaining and justifying are joint as well as individual activi-
ties and illustrate the difficulties in communication that can arise unless there
is a taken-as-shared understanding of what it means to explain and justify.
With this initial orientation completed, we then compare and contrast the
quality of the normative mathematical activities established by the teachers
and students in two classrooms.

Classroom Norms
Much and Shweder (1978) identified five types of classroom norms: regula-
tions, conventions, morals, truths, and instructions. They used a variety
of criteria to distinguish between these norms including their historicity,
their source, and the consequences of transgressing them. In this scheme,
regulations are historical and are established by a specifiable authority who
can alter them. The consequence of breaching a regulation is typically a
penalty of some kind. As an example, a teacher might tell students during
small-group work that only one member from each group is allowed to
fetch the manipulative materials that the group decides to use. This direc-
tive is a regulation in that it is established by the teacher who can change it.
Conventions are also historical, but their source is not specifiable and
the consequence of a transgression is social disapproval. The distinction
between regulations and conventions is analogous to that between laws
and customs. As an example, in the context of traditional instruction, it is
customary both for students to attempt to respond to the teacher's known-
answer questions and for the teacher to evaluate their responses (cf. Mehan,

575

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
1979). The potential alterability of these norms is demonstrated by Lampert's
(1990) discussion of a classroom in which interactions did not follow this
traditional elicitation-response-evaluation pattern.
In contrast to regulations and conventions, norms classified as either
morals, truths, or instructions are all treated as being ahistorical by members
of a community. Much and Shweder argue that the consequence of trans-
gressing a moral is moral culpability. An example is the traditional classroom
norm that students should not copy another's answers and pass them off
as the results of their own work. The teacher might attempt to make a stu-
dent who transgresses this norm feel guilty about what he or she has done.
It is this sense of culpability that distinguishes morals from truths and in-
structions. The consequence of transgressing a truth is error per se, whereas
that for transgressing an instruction is ineffectiveness.
A priori, these latter two types of norms appeared to have the greatest
relevance for our analysis given that our focus was on one of the more con-
ceptual aspects of elementary school mathematics: place value numeration.
We would accept the arguments of historians and quasi-empiricist philoso-
phers of mathematics that mathematics is a continually evolving human con-
struction. Nonetheless, in contrast to other aspects of mathematics such
as the use of particular notations, it seemed unlikely that place value would
be viewed as potentially revisable by teachers and students when they did
arithmetic in the classroom. Further, although one can imagine a variety
of consequences for transgressing the norms of place value numeration,
it would seem somewhat bizarre if moral culpability were one of them. Thus,
although we could not dismiss this possibility in advance, our primary mo-
tivation in drawing on Much and Shweder's work was to investigate whether
the distinction they made between truths and instructions might be useful
in characterizing differences between classrooms where meaningful mathe-
matical learning occurs and those where it does not.

Situations for Explanation and Justification

The distinctions Much and Shweder made between the various types of
classroom norms were not developed in the abstract, but instead emerged
as these researchers analyzed classroom interactions. In their analysis, they
focused on those moments in which one member of the classroom com-
munity accuses another of having transgressed a norm. For example, in the
setting of small-group work, one child might accuse another of interpreting
the tens digit of a two-digit number as so many ones rather than so many
tens. Much and Shweder call moments of this type situations of account-
ability and term the two basic acts that constitute them as accusations and
accounts. Because we are interested in accountability as it occurs during
the negotiation of mathematical meanings and practices, we will talk of
situations for justification and will call the two basic acts challenges and
justifications.
576

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
The perception of a departure from normative mathematical activity
distinguishes situations for justification from those that we call situations
for explanation. In the latter case, the legitimacy of mathematical activity
is not the issue. Instead, the teacher and students give explanations in an
attempt to communicate aspects of their mathematical thinking that they
think are not readily apparent to others, thereby contributing to the develop-
ment of a basis for communication. At times, the teacher or students simply
infer that aspects of their activity are not taken as shared whereas, on other
occasions, an explanation is given in response to an explicit request. Thus,
if the theme of situations for justification is accountability, then that of situa-
tions for explanation is clarification.

Explaining and Justifying as Joint Activities

We can further develop the notions of a situation for explanation and for
justification by contrasting our use of the term situation with that found
in Brousseau's (1984) theory of didactical situations as it has been explicated
by Balacheff (1990) and Laborde (1989). Brousseau's theory is motivated
by didactical considerations and the various types of situations he distin-
guishes can be characterized by the obligations that the teacher hopes stu-
dents will accept for their mathematical activity. For example, situations
for action require students to search for a solution to a given task, situa-
tions for formulation require students to make their interpretations and
conceptualizations explicit, situations for validation require students to
justify what they have made explicit, and situations for institutionaliza-
tion require students to accept the teacher's legitimation of mathematical
constructions selected from those that have been developed in the course
of classroom activity. This analysis is of value in that it provides a theoretical
framework that can guide the development of didactical experiments which
take account of both the specific mathematical tasks posed to students and
the social settings within which students should, ideally, attempt to com-
plete these tasks.
Our development of the notions of situation for justification and situa-
tion for explanation was motivated by a different concern, namely, to
analyze the nature of classroom mathematics traditions rather than to for-
mulate didactical situations. Consequently, there is no direct correspondence
between the two sets of constructs. For example, a situation for explana-
tion or for justification can occur as students solve a given task, explicate
their interpretations, validate what they have made explicit, or discuss the
legitimacy of particular mathematical constructions. Further, whereas the
various types of didactical situations can be characterized by the obliga-
tions that the researcher or teacher hopes students will accept, situations
for justification and for explanation are interactively constituted by the
teacher and students, or by a group of students, as they attempt to coor-
dinate their mathematical activity (Bauersfeld, 1980; Bruner, 1986; Mehan,

577

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
1979; Voigt, 1989). From this point of view, explaining and justifying are
considered to be collective or joint activities (cf. Blumer, 1969; Mead, 1934).
We illustrate this aspect of situations for justification and explanation
with reference to a sample episode in which two girls attempted to solve
a task that involved finding the sum of three nine teens:

Karen: I know how to do it. You guys won't listen to me though.

Marie: Just a minute!
Karen: It's 10, 20, 30—(she then counts her fingers) 31,32,33... 55,
56, 57.
Marie: It's not that far.

Here, Karen explained her solution without a request to do so, presumably

because she inferred that her approach was not readily apparent to Marie.
Marie for her part contributed to the interactive constitution of a situation
for explanation by attempting to make sense of Karen's solution before she
challenged its legitimacy. In order for this to evolve into a situation for jus-
tification, Karen would have to offer a justification in response to Marie's
challenge. This she did, saying, "I added up all the numbers." Marie, how-
ever, ignored Karen's justification and proposed that they resolve the con-
flict by counting on a hundreds board:

Marie: Let's get the hun. . .

Karen: What's 18 plus 30?
Marie: (Backing away) I know what... we can get the hundreds
board. . .
Karen: What's 18 plus 30?
Marie: Hundreds board. . .
Karen: Marie!
Marie: I'm gonna get the hundreds board. . . then we can look for
it. . .
Karen: Marie, I know the answer.

Karen probably reconceptualized her activity for counting three nines on

from 30 by ones as 30 plus 18 rather than 30 plus 27. She repeatedly posed
the question, "What's 18 plus 30?" in an attempt to persuade Marie of the
legitimacy of her solution after her initial justification had been ignored.
Marie, however, did not attempt to answer Karen's question but continued
to propose that they use the hundreds board. As a consequence, the situa-
tion for justification that they briefly established dissolved without reaching
a conclusion about the legitimacy of Karen's solution.
As this short episode illustrates, the individual acts of challenging or
justifying do not by themselves constitute a situation for justification. The
person whose mathematical activity is questioned must interpret this as a
challenge and respond accordingly. Similarly, the person to whom a justifica-
tion is given must either accept it or say why it is inadequate, thus making
a further challenge (cf. Mehan & Wood, 1975). Much the same can be said
578

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
in the case of situations for explanation. We would not say that a situation
for explanation had been interactively constituted unless someone attempted
to interpret the explanation offered. In short, a situation for explanation
or for justification is interactively constituted only when the understand-
ing that an interpretation or solution should be explained or justified is taken
as shared.
Mathematical Communication
Thus far, we have illustrated that mathematical communication can break
down unless the need to explain or justify an interpretation or solution is
taken as shared. Further difficulties in communication can arise even when
a situation for explanation or for justification is interactively constituted
unless there is a taken-as-shared understanding of what counts as an ex-
planation or a justification. This point can be clarified by considering an
episode in which two children worked together to solve a task that involved
finding the sum of five twelves. One child had written five twelves in ver-
tical, column format and had used the standard addition algorithm to ar-
rive at her answer of 60. She explained her solution to her partner as follows:

Ann: 2, 4, 6, 8, 10 (adds the five twos in the ones column). You

put the "Oh" there (writes "o" in the ones place), the one
there (writes " 1 " at the top of the tens column). 1, 2, 3, 4,
5, 6 (counts the six ones in the tens column). Do you agree
that's 60?

In the view of her partner, Jack, "That's impossible." The exchange

continued:

Jack: Hold on. This is 10. Look. 1, 2,. . .8, 9, 10 (adds the five
twos in the ones column). 10.
Anne: One there (points to the " 1 " she has written at the top of
the tens column). 1, 2, 3, 4, 5, 6. 60.
Jack: 610?
Anne: You don't listen.
Jack: I don't get this.

In this exchange, the children interactively constituted a situation for jus-

tification, the theme being the legitimacy of Anne's solution (cf. Voigt, 1989).
Jack repeatedly challenged Anne's justifications because he did not under-
stand how ten, the sum of the ones column, could become an "o" and a one:

Anne: 2, 4, 6, 8, 10. Now listen. Put that "o" there.

Jack: Yeah.
Anne: The one up there.
Jack: That isn't "o."
Anne: 1, 2, 3, 4, 5, 6.
Jack: That's not "o." That's 10.
Anne: No.

579

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
Jack: That isn't "o" though. That's 10. 1, 2,. . .8, 9, 10.
Anne: Now listen.

As the episode continued, Anne repeatedly described the steps of the stan-
dard algorithm in response to Jack's challenges. In the setting of arithmetical
computation, a procedural description of this type seemed to constitute
a justification for her, and she concluded that Jack's failure to accept her
justification indicated that he was not listening. On the other hand, from
Jack's point of view, Anne had not given an adequate justification for her
acts of writing " o " and " 1 . "
At this point, the children had reached an impasse and Anne sought
the teacher's assistance. When she joined the group, both children explained
their interpretations of the conflict to her:

Anne: He doesn't—he doesn't get. He doesn't get that right.

Jack: (Simultaneously) I don't understand. All the two's add up
to ten. She takes one away from the ten.

Jack's last statement indicates that he interpreted Anne's action of writing

one at the top of the tens column as taking away one from the ten. As a
consequence, Anne's act of writing the " o " in the ones place was irrational
to him. Anne responded to his challenge by again repeating her procedural
description:

Anne: Two plus two. 2, 4, 6, 8, 10. I know that's 10.

Jack: It is 10.
Anne: You put the darn "oh" there. 1, 2, 3, 4, 5, 6.
Jack: (Simultaneously) Show me how you are doing it. I don't
understand.

Jack's comment, "Show me how you are doing it," was an explicit
request for an explanation. He now seemed to accept that Anne's solution
was legitimate, possibly because the teacher had not challenged it. However,
from Anne's point of view, she has justified her solution four times in
response to Jack's challenges. For her, a step-by-step description of a se-
quence of symbol-manipulation acts should have been sufficient to legitimize
her solution and she concluded that Jack was being uncooperative. In con-
trast, an analysis of Jack's mathematical activity both here and in 20 other
lessons consistently indicated that he attempted to learn with what is typical-
ly called understanding (Cobb, 1992). From his point of view, Anne's
descriptions of her acts of manipulating numerals were not adequate as either
explanations or justifications. In another episode, he in fact characterized
her computational methods as "mixing up a bunch of numbers." An ex-
planation or justification seemed to be acceptable to him only if he could
interpret it in terms of mental actions on numbers experienced as manipul-
able arithmetical objects. These actions include those of combining num-

580

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
bers, decomposing or partitioning numbers, and transforming numbers (e.g.,
transforming ten units of one into one unit of ten, itself composed of units
of one). In this exchange with Anne, he was unable to give numerical
significance to her procedural descriptions and she was unable to do more
than specify how she manipulated numerals. As a consequence, the two
children arrived at an impasse that they were unable to resolve even with
the teacher's assistance. Their diffferent understandings of what counted
as an explanation and a justification made it impossible for them to com-
municate effectively and negotiate a taken-as-shared interpretation of Anne's
solution in general, and of the symbol "10" in particular. An interpersonal
conflict that might have given rise to learning opportunities for both children
instead resulted in mutual feelings of frustration. In this regard, their dialogue
reminds one of adherents to incommensurable paradigms who, in Kuhn's
(1970) terms, talk past each other.
As we turn to illustrate the relevance of the constructs discussed above
to analyses of classroom life, the reader might in fact come to view Anne
and Jack as representatives of different classroom mathematics traditions.
The two lessons we will examine in detail make it possible to clarify dif-
ferences between classrooms in which meaningful mathematics learning is
encouraged and those in which it is not. In the course of the analyses, we
will consider both the quality of the mathematical activity in each classroom
and the processes by which that activity was interactively constituted.

Mathematical Activity as Procedural Instructions

The first lesson we will consider occurred near the beginning of the school
year in a third-grade classroom when the teacher and students were review-
ing the place value interpretation of two-digit numerals and number words.
It is one of 28 lessons that were video recorded in this classroom over a
6-week period (McNeal, 1991). We should stress at the outset that our goal
is not to criticize a well-intentioned teacher. The teacher was, in fact, a sin-
cere professional who held the intellectual and social welfare of her students
paramount. By all traditional criteria including those derived from the ef-
fective schools research, she would be judged as highly competent. The
central issue is, instead, to clarify how students can come to view mathe-
matics in school as an activity that involves manipulating symbols that do
not necessarily signify anything when their teachers would honestly like
them to learn with understanding.
As the lesson commenced, the teacher placed a drawing of 45 tally
marks on an overhead projector and asked the children, "Without count-
ing, just without counting, how many sticks do you think are there? Just
(snaps her fingers), just estimate. How many do you think there are?" The
children gave a variety of estimates ranging from 25 to 53. At this point,
the teacher initiated a change in the theme of the discourse, saying, "OK.
Now, those are just estimates. Now let me show you a real easy way to

581

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
figure out how many are on that board there." The teacher, in effect, of-
fered an unsolicited rationale for the procedure that she then demonstrated,
that of repeatedly counting and circling groups of ten sticks. One can
speculate that this rationale was not convincing to many of the children
given that they would have found it both easier and more efficient simply
to count the sticks by one. Importantly, none of the children challenged
the teacher's rationale and, as a consequence, she did not have to justify
her procedure. Possibly, the children understood that the teacher's rationale
should not be taken seriously because she was actually demonstrating the
procedure for didactical purposes. Krummheuer (1983) called situations of
this type, in which students demonstrate a willingness to cooperate even
if they do not comprehend, working interims.
As the lesson continued, the teacher twice asked, "How many groups
of ten do we have?" and "How many left over?" Very few children re-
sponded, suggesting that most did not understand what was expected of
them. These children continued to participate and did not request an ex-
planation. The teacher then asked,

Teacher: Now, can anybody tell me what number that would be?
We have four tens and five ones. What is that number?
Stephanie?
Stephanie: (No response).
Teacher: If we have four tens and five ones, what is that number?
Stephanie: (Quietly) Nine.
Teacher: Pardon me?. . .

From the teacher's perspective, Stephanie's additive interpretation of the

situation transgressed a mathematical norm. Importantly, although she chal-
lenged Stephanie's interpretation, she did not seem to expect her to justify
it and Stephanie did not attempt to do so:

Teacher: Look at how many we have there (points to the four

groups of ten).
Sandra: (Whispers to Stephanie) 45.
Stephanie: 45.
Teacher: And five ones. If we have four tens and five ones, we
have. . .45 (students join in on "45").

It seems reasonable to suggest that the implicit and unintended lesson of

such exchanges is that mathematical interpretations do not need to be justi-
fied. Stephanie's quick acceptance of Sandra's suggestion indicates that she
had learned this lesson well. Notice that the teacher did not feel obliged
to offer an explanation at any point during this exchange. Instead, she con-
cluded the episode by saying, "If we have four tens and five ones, we have
. . .45." The children could reasonably have interpreted this statement as
a directive about how to respond effectively. We will see that this was the
case for many students as we consider subsequent episodes.

582
Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014
 Classroom Mathematics Traditions
In the next phase of the lesson, the teacher first wrote "tens" and
"ones" as the headings of two columns. She then wrote "7, 3 " and "6,
9" and " 5 , 8" as successive rows of the table and asked the children to
"write the numbers." She moved around the room as the children com-
pleted the tasks until she reached Stephanie's desk and said, "Oh, no. I just
want you to write the number. I want you to write the number. I want
you to write [inaudible]. I need you to write it like it should be. It would
not be six plus nine. It is. . .69, OK?" It would seem that Stephanie had
not modified her interpretation of the tasks even after her first exchange
with the teacher. The teacher's intervention in this second exchange ap-
peared to preclude the interactive constitution of a situation in which
Stephanie would be obliged to explain or justify her answer. As a conse-
quence, Stephanie again could have reasonably concluded that mathematical
activity need not be explainable.
Once the above tasks had been completed, the teacher passed out
Cuisenaire rods of two sizes and established with the children that there
were ten little blocks in each of the longer blocks. We can note that both
this use of manipulative materials and the first instructional activity in which
the teacher drew and circled tally marks indicate that she genuinely wanted
her students to learn with understanding. Thus, our focus is not on her
intent, which is above reproach, but on the characteristics of mathematical
activity as it was realized in her classroom.
The teacher introduced the first task that involved the use of Cuisenaire
rods as follows:

Alright. I'm going to put this [card] on the board. One of them is
going to tell you how many tens I want you to find and the other
one is going to tell you how many ones. I want you to use your
ten rod and your one rod to show me the number that I put on the,
on the board over here... All right? The first thing I want you to
show me is three tens and five ones.

The teacher posed eight tasks of this type, each time asking the children
to say what number they had made once they had put out an appropriate
collection of rods. The children were able to give the desired responses
when called upon and the discourse generally proceeded smoothly. The
teacher introduced the third task by stating, "Now, be careful here, guys.
I want you to show me seven tens and no ones." The children responded,
"Seventy," and she then requested an explanation:

Teacher: Now, boys and girls, look up here (writes "7" on the chalk-
board). Why didn't we just put seven? Because you have
seven of these (holds up a long rod), you have nothing
else. Why didn't we just put seven?
Stephen: Because there's 70 um blocks.

583

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
In posing this question, the teacher expressed her professional knowledge
in action as she anticipated a possible misinterpretation on the part of her
students:

Because there's 70 blocks. That's right. This tells us (underlines the

seven) that there were seven tens. It has to have a zero here (writes
"0" after the "7" and underlines it twice) to tell us that there are
no ones. OK? Very good.

We infer from Stephen's explanation that he had solved the task by

constructing a numerical relationship. In particular, he appeared to inter-
pret the seven long rods as composite units of ten and to transform them
mentally into seventy units of one (i.e., seventy smaller rods). The teacher
acknowledged that his explanation was valid but then gave an alternative
explanation, which stressed a rule for mapping between the digits of the
numeral and the number of each type of rod. In doing so, her actions im-
plicitly served to delegitimize Stephen's interpretation of the task as a rela-
tionship between numbers experienced as arithmetical objects. Her action
is reasonable if one grants that she treated place value numeration as a set
of specific procedures with the intention of facilitating her students' learn-
ing (cf. Brousseau, 1984; Mason, 1989; Steinbring, 1989). For the teacher,
who had constructed a relatively sophisticated understanding of place value
numeration, these procedures specified how to manipulate numbers.
However, many students, like Jack when he interacted with Anne, could
not give them the numerical significance that she intended. Consequently,
her implicit delegitimization of Stephen's solution can be viewed as one
brief moment in the process by which she unknowingly inducted the
children into the view that arithmetic involves following procedures that
specify how to manipulate numerals and number words per se. The analysis
of both the remainder of this lesson and the other 27 lessons that were
video recorded indicates that this stance towards arithmetic became increas-
ingly taken as shared in the classroom as the children learned how to par-
ticipate in the smooth flow of classroom discourse by acting in accord with
her expectations (McNeal, 1991).
If we accept that the teacher acted as though place value numeration
was a set of procedures, then her request for an explanation can be seen
as an attempt to ensure that the children did not misinterpret the specific
procedure that she had in mind. In this case, the instruction was to read
the first numeral and put out that number of long rods, and then read the
second numeral and put out that number of small rods. The teacher again
stated this instruction after the children had completed the last of the eight
tasks, zero tens and two ones. "What, when we write this number, boys
and girls (points to '2' on the chalkboard), it shows us is that we have two
ones and we have no tens. We don't have to put anything there (points
to the left of the '2'). All right?"
584

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
Our contention that the teacher unknowingly initiated and guided the
interactive constitution of place value numeration as a set of taken-as-shared
procedures is consistent with several other observations. For example, the
teacher frequently used the phrase "I want you to. . . " when she posed
the tasks. This suggests that she implicitly assumed that the children's pur-
pose when they engaged in mathematical activity was to act in accordance
with her expectations. There is every indication that this purpose was taken
as shared. We have noted that the children demonstrated a willingness to
cooperate even when they did not know what the teacher expected of them.
Further, Stephanie answered the teacher's question on one occasion by ac-
cepting another child's prompt despite her apparent lack of comprehen-
sion. In short, the children acted as pragmatists and strove to be effective
in the social setting of the classroom by attempting to complete tasks and
answer questions in ways that were compatible with the teacher's expecta-
tions (cf. Balacheff, 1986).
The analysis we have presented thus far indicates that normative
mathematical activities in this classroom had the quality of instructions in
Much and Shweder's terms. As they observed, the consequence of trans-
gressing an instruction is ineffectiveness in that the means are inadequate
to achieve the end. In this regard, we would not merely want to say that
Stephanie made mistakes when she twice interpreted place value tasks as
additive situations. She was ineffective in that she could not achieve the
end of acting in accord with the teacher's expectations and, consequently,
could not participate in the routinized pattern of classroom discourse that
involves teacher elicitation, student response, and teacher evaluation (cf.
Mehan, 1979). At these moments, she was temporarily an outsider with re-
gard to the classroom community. It is, therefore, not surprising that she
accepted another child's prompt to give what seemed to have been to her
a completely inexplicable answer.
The children's activity in the remainder of the lesson further substan-
tiates the claim that their overriding goal was to be effective by following
specific procedural instructions. For the last whole class activity, the teacher
and children worked through four textbook tasks that each involved a pic-
tured collection of base ten longs and individual cubes. In each case, the
teacher asked, "How many tens do you see?" and "How many ones do
you see?" and "What number is that?" The children answered these ques-
tions appropriately for the first two tasks. The picture for the third task
showed three longs and six cubes:

Teacher: How many tens do you see? Monica?

Monica: [No response].
Teacher: [Problem] number three.
Monica: Three.
Teacher: How many ones do you see? James?
James: Four.
Teacher: Not in number three, James.

585

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
James: [Inaudible].
Teacher: Six. And what number is that, James?
James: 60.
Both the teacher's and James's subsequent actions indicate that the trans-
gressed mathematical norm was for them an instruction. They did not in-
teractively constitute a situation for justification but, instead, the teacher
began to specify the instruction as James attempted to infer how she ex-
pected him to act. In the process, they constituted a recurrent pattern of
classroom interaction identified by Bauersfeld (1980) and called the funnel
pattern:

Teacher: Look at number three, James. How many tens do we

see? (moves towards him).
James: Three.
Teacher: And how many ones?
James: Six.
Teacher: And what number is that?
James: 63.
Teacher: OK, let's look. James. Look. James, look. How many
tens do we see? Three tens. How many ones? Six ones.
James: 36.
Teacher: 36. Good.
What did James learn in the course of this exchange? Did he construct
numerical relationships and thus conceptualize thirty-six as composed of
units of tens and of ones? It seems more likely that he found a way to pro-
duce an answer acceptable to the teacher by focusing on the number words
"three" and "six." When he found that "sixty-three" was incorrect, he tried
"thirty-six" as an alternative. Like Stephanie, his goal in this social situation
appeared to be to find a way to produce the desired answer. Also like
Stephanie, he was ineffective until he did so in that the smooth flow of
the classroom discourse characterized by the elicitation-response-evaluation
pattern broke down. In this and other exchanges, the possibility that
arithmetic could be anything other than procedural instructions did not arise.
The lesson concluded with seat work as the children completed as-
signed textbook tasks. This portion of the lesson was also notable for the
children's frequent discussion of what they "are supposed to do." This con-
cern was quite reasonable given that they conceptualize their task to be
that of following procedural instructions. For example, the following ex-
change occurred when the teacher stopped at Christopher's desk:

Christopher: What are we supposed, Mrs. Jackson, what are we sup-

posed to do over in this part? What are we supposed
to do down here? (points to a task in his textbook).
Teacher: You write the number, OK? Number three says six tens
and eight; you did that just fine. Let's look at number
four. Four tens and zero ones. But look at number five,
Christopher; it says three ones and one ten.

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
In response to the teacher's warning, Christopher proposed an instruction,
which the teacher corrected:

Christopher: I just have to turn everything around.

Teacher: Not everything. Check. You better look at 'em very
carefully.
Christopher: Yeah, not everything, though.

The teacher continued to move around the room and stopped at

Christopher's desk for a second time when she noticed that he had trans-
gressed a procedural instruction. The funnel pattern that was constituted
in the exchanges involving Stephanie and James occurred again, although
in curtailed form:

Teacher: No, no, Christopher. How many tens do we see?

Christopher: Two.
Teacher: Two.
Christopher: (Counts the cubes in the picture). Six. 26.

Here, Christopher anticipated and answered the teacher's next two ques-
tions (How many ones do we see? What number is that?). As a consequence,
a previously established pattern of interaction had, in curtailed form, become
a personal mathematical routine for him. In other words, he had learned
by actively constructing instructions compatible with those that the teacher
had in mind. This individual constructive activity in turn contributed to
the classroom community's constitution of place value numeration as a set
of procedural instructions.

School Mathematics
In analyzing the third-grade lesson, we focused both on those moments
when one member of the classroom community accused another of trans-
gressing a mathematical norm and on those occasions when the discourse
proceeded smoothly. Every challenge identified was made by the teacher,
and, in this sense, she acted as the sole validator of what could count as
legitimate mathematical activity. We noted that none of the teacher's chal-
lenges initiated the interactive constitution of a situation for justification.
Further, only one situation for explanation occurred, the exchange in which
the teacher asked why seven tens and zero ones would not be seven and
Stephen explained, "Because there's um seventy blocks." This was perhaps
the most significant incident of the entire lesson in that it revealed that the
teacher's actions unintentionally discouraged an interpretation that involved
the construction of a numerical relationship from becoming taken as shared.
If relationships of this sort are excluded, then mathematics is reduced to
an activity that involves constructing associations between signifiers that
do not necessarily signify anything beyond themselves. In particular, the
children did not have to construct and act on units of ten and of one as

587

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
arithmetical objects in order to be effective. Instead, it was sufficient for
them to treat verbal expressions such as "seven tens and zero ones" as en-
tities in and of themselves, and to transform them into other expressions
as directed by instructions (e.g., "seven tens and zero ones" transformed
into "seventy" and "70"; "6 9" transformed into "sixty-nine" and "69").
It is also apparent that instructions were the theme of the discussion
even when the instructional activities involved the use of manipulative
materials. In this case, the instructions specified how numerals and number
words should be linked with configurations of physical objects. This con-
clusion is consistent with the findings of Voigt's (1990) microanalysis of
classroom discussions in which the tasks involved pictures and objects. His
analysis indicates that the teacher and students used number words and
numerals to refer to specific aspects of the pictures and objects that were
incidental from a mathematical perspective. As a consequence, the students
could be effective by making these specific associations and did not have
to construct relationships between numbers experienced as mathematical
objects.
The conclusion that the mathematical terms used in the third-grade
classroom did not necessarily signify mathematical objects for students is
further substantiated by the observation that it is extremely difficult to make
any inferences about the sophistication of their conceptions of place value
numeration. Detailed cognitive models, such as those developed by Steffe,
Cobb, and von Glasersfeld (1988) that specify children's construction of
increasingly sophisticated units of ten, are largely irrelevant to the analysis
of this lesson. Like Stephen, some of the children might have been covert-
ly creating and acting on numbers experienced as arithmetical objects when
they interpreted and solved the tasks, but neither we nor the teacher could
have known that. Such mathematical activity, would, in fact, have been
conducted in the face of instruction that implicity discouraged it, and as
such would have been incompatible with the mathematical tradition estab-
lished in this classroom.

Folk Belief About Mathematics

Procedural instructions, as we have described them, might appear to be
synonymous with regulations or with conventions. As we have noted, the
distinction that Much and Shweder (1978) make between regulations and
conventions, on the one hand, and instructions, on the other, is that regula-
tions and conventions, but not instructions, are treated as being potential-
ly alterable. The way in which the teacher and students interacted in both
the sample episodes we presented and in the other 27 lessons suggests that
the taken-as-shared procedures that they established were viewed as ahis-
torical, unalterable norms that did not have a specifiable source. To be sure,
the children learned as they succeeded in acting in accord with the teacher's
expectations. However, the teacher's instructional routines indicate that she

588

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
regarded the mathematical procedures the children were to learn as both
fixed and self-evident. For example, she urged James, "OK, let's look. James.
Look. James, look," and said to Stephanie, "I need you to write it like it
should be," after each had transgressed a mathematical norm. More general-
ly, she only began to give cues as to the response she had in mind when
she judged that a child had transgressed a norm. For the most part, she
asked tightly sequenced series of questions (e.g, "How many tens do you
see?" "How many ones do you see?" "What number is that?") and expected
the children to answer correctly without explicit instruction. This expec-
tation only makes sense if she considered that she was drawing self-evident,
objective rules to the children's attention. The manner in which the children
routinely cooperated even when the mathematical rule was not immediately
evident to them indicated that, at a minimum, they had learned to act as
though they believed that mathematics consists of fixed, objective rules.
Thus, despite the teacher's intentions, the manner in which she initiated
and guided classroom interactions served to facilitate her students' encultura-
tion into what Lave (1988) called the folk beliefs about mathematics. These
beliefs include the conviction that it is impermissible to use any methods
other than the standard procedures taught in school to solve school-like
tasks and that the use of these procedures is the rational and objective way
to solve mathematical tasks in any situation whatsoever. For example, Lave
found that adults are typically defensive about their use of efficient and ef-
fective nonstandard methods in out-of-school situations, such as making
price comparisons while shopping for groceries in a supermarket.
We should stress that the process of guiding the students' encultura-
tion into the folk beliefs was implicit in that we have no indication that
it was a consciously formulated part of the teacher's instructional agenda.
Further, our comments should not be read as a criticism of the teacher;
she was merely doing her job to the best of her ability. Our point is, in-
stead, that the folk beliefs about mathematics are held by administrators,
parents, and (apparently) textbook writers as well as by many elementary
school teachers. In order words, these beliefs are deeply ingrained in our
society and in the culture of elementary school mathematics in particular,
and mathematics instruction as it is traditionally characterized is a process
of inducting the next generation into this interpretive stance towards mathe-
matical activity. The analysis we have presented offers a glimpse of this
process as it is realized moment by moment in the course of classroom
interactions.
Transmission and Construction
Our contention that the teacher acted as though place value numeration
was a set of instructions when she interacted with her students does not
imply that she merely transmitted these rules to them. In fact, the observa-
tion that her primary instructional routines involved questioning rather than
explicit telling or demonstrating indicates that the frequently used trans-
589
Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014
Cobb, Wood, Yackel, and McNeal
mission metaphor inadequately characterized her beliefs in action. If we
were to attribute an implicit instructional theory to her, it would be that
of sequencing tasks and asking questions in order to draw self-evident
mathematical relationships to students' attention. If should also be clear from
the sample episode that the children were not passive recipients of
mathematical knowledge. As in any instructional setting, they actively con-
structed their mathematical ways of knowing as they interacted with the
teacher and each other in the classroom. Thus, one does not need to employ
behaviorist theory to account for learning in the setting of typical textbook
instruction and constructivist theory to account for learning in classrooms
more compatible with recent reform recommendations. Learning in both
settings is susceptible to a constructivist analysis. The difference between
the two resides in the contrasts between qualitatively distinct classroom
mathematics traditions. This becomes apparent when we consider a second
lesson.

Mathematical Truths
The lesson we will analyze occurred midway though the school year in
a second-grade classroom in which every mathematics lesson was video
recorded for the entire school year. The lesson was selected because, like
the third-grade lesson, it occurred in a whole class setting, because it focused
on place value, and because it involved the use of similar manipulatives.
The teacher introduced the first instructional activity by placing two
longs and 11 individual cubes on an overhead projector. She then showed
this array to her students for approximately 3 seconds and asked them to
describe what they saw. One child said that he had seen 31 when the teacher
called upon him. Some students agreed while others disagreed. The teacher
said, "All right. Let's take a look. How are we going to figure this out?"
In contrast to the third-grade classroom, the students took it for granted
that they could challenge each other's interpretations, solutions, and
answers. The teacher did not merely tolerate this behavior. At the begin-
ning of the school year, she had actively encouraged her students to make
challenges, and the children were fulfilling their obligations to her by do-
ing so (Cobb, Yackel, & Wood, 1989). We can also note that, in contrast
to the third-grade classroom, none of the children said, for example, " 1 3 , "
when the task was presented. There was a taken-as-shared understanding
of the task that reflects a prior history of explicit negotiations of meanings
and interpretations in this classroom. In concert with our analysis of the
third-grade lesson, Voigt (1990) found that tasks are frequently ambiguous
in traditional classrooms. As he observed, teaching as it is typically con-
ceptualized does not acknowledge the plurality of possible interpretations
that students might make, and teachers do not initiate and guide the ex-
plicit negotiation of routines of interpretation.
590

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
In the case of the task involving two longs and 11 individual cubes,
it soon became apparent that most, but not all, of the children agreed with
the initial answer of 31:

Michael: I . . . 30. I think it's 30 (walks to the screen).

John: How can he get 30?
Teacher: Let's listen to his explanation.
Michael: 'Cause 10, 20, 21, 22, 23,. . ., 31. I missed one.
Teacher: You're just off by one. Sure, that's OK.

The task might seem straightforward, merely specifying the cardinality of

a collection of base ten blocks. Nonetheless, it gave rise to a discussion in
the course of which a situation for justification was interactively constituted.
We note that Michael went to the screen without being asked, indicating
that he took it for granted that he was obliged to justify why he disagreed
with the view of the majority. The teacher legitimized his interpretation
of this as a situation for justification, saying, "Let's listen to his explana-
tion." In doing so, she initiated the children into her interpretive stance
in which mathematical activity should be explained and, when necessary,
justified.
The episode also illustrates a further point of contrast with the
classroom mathematics tradition established in the third-grade classroom.
It is apparent that Michael was effective in the classroom even though his
initial answer 30 proved to be incorrect. Whereas children who made er-
rors in the third-grade classroom temporarily became outsiders to the
classroom community, Michael participated in the constitution of a routine
pattern of interaction when he gave a justification in response to a challenge.
The teacher's final comment, "Sure, that's OK," in fact explicitly legitim-
ized his activity. Thus, the consequence of his failure to enumerate the col-
lection appropriately was error per se rather than ineffectiveness.
The distinction between error and ineffectiveness is important in light
of Much and Shweder's analysis in that it corresponds to the distinction
they make between truths and instructions. At first glance, truths might not
seem normative. The truths of the communities to which we belong ap-
pear to tell us how the world is, not how it ought to be. This view overlooks
the fact that we are typically challenged by another member of our com-
munity if our actions are not in accord with accepted truths. If our actions
continue to conflict with a truth and we cannot justify our conduct in a
way that satisfies the standards of argumentation of the community, we
eventually cease to be members of that community. Thus, in this view,
members of a community interactively constitute the truths that constrain
their individual activities. Because truths tell the members of a community
for which they are truths how the world is or ought to be, the consequence
is intrinsic to the transgressing action—one has got things wrong or has
made an error. Michael, for example, acknowledged that his initial answer
of 30 was untrue by explaining, "I missed one." We have seen that, in con-
591

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
trast, the consequence of transgressing an instruction is at least partly ex-
trinsic to the action—one fails to achieve the end for which the instruction
is the means. It might be argued that Michael did, in fact, transgress an in-
struction, that which specifies how to count a collection of base ten blocks.
However, there is no indication that the teacher and students interpreted
the situation in this way. Instead, there is every indication that they interac-
tively constituted the cardinality of the collection, rather than the legitimacy
of particular counting procedures, as the theme of their discussion. As Barnes
(1982) observed, a result is acceptable if the process by which the result
is obtained is acceptable. This is the case in both scientific communities
and classroom mathematics communities (Cobb, Wood, & Yackel, 1991).
Michael counted by way of justification to demonstrate the process by which
he had arrived at his answer of 30, not to demonstrate that he had followed
instructions that specify how to count correctly. His counting activity and
his observation that he had missed one indicate that, as is the case with
scientists' activity, the context in which a result is obtained could not be
separated from the context of justification in this second-grade classroom.
This analysis of the initial exchange between Michael and the teacher
suggests that the classroom tradition was one in which the teacher and
children constituted mathematical truths in the course of their social in-
teractions and that the acts of explaining and justifying were central to this
process. We can estabish whether these conclusions have any generality
as we consider the remainder of the lesson. As the discussion continued,
Brenda indicated that she had an insight that she wanted to share:

(Goes to the screen). One thing is. . . you know there is two ten bars
that make 20 and then you count. You count these (indicates the
individual cubes), you count these 1, 2, 3,. . ., 11. Since there's 11
[inaudible] you see then you one, two (points to the two ten bars),
then take 10 and that makes 30. And then you put that extra one
for 11 and you get 31.

Although it had never been explicitly discussed, the teacher and students
seemed to have a taken-as-shared understanding of what it meant to have
a mathematical insight. As Brenda's explanation indicates, an insight typically
involves constructing numerical relationships that made it possible to solve
a task in a more sophisticated or efficient manner. In addition, her behavior
provides us with a further indication that the teacher and students took
it for granted that mathematical activity is intrinsically explainable. We can
also note that cognitive models of children's conceptual understandings,
such as those developed by Steffe et al. (1988), are relevant as we attempt
to analyze their interactions. Given that our focus is on the quality of the
classroom mathematics tradition, rather than the sophistication of individual
children's mathematical understanding, we will not present detailed analyses
of their individual task interpretations. Nonetheless, it is readily apparent
592

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions

that the units often that Brenda constructed were more sophisticated than
those that Michael constructed and counted. At least in this instance,
Michael's units of ten were perceptually based whereas the units Brenda
constructed transcended perceptual constraints. The fact that we can make
inferences of this type when we examine the classroom discourse indicates
that the children were constructing and acting on numbers conceptualized
as arithmetical objects that had a manipulable quality. Brenda said, for ex-
ample, "You one, two (points to the two ten bars), then take 10 and that
makes 30. And then you put that extra one for 11 and you get 31." It would
therefore seem that the teacher and students were constituting a taken-as-
shared mathematical reality as they discussed their interpretations and
solutions.
This contention is reminiscent of the observation of Davis and Hersh
(1981): "Mathematicians know that they are studying an objective reality.
To an outsider, they seem to be engaged in an esoteric communication with
themselves and a small clique of friends" (pp. 43-44). The distinction be-
tween studying a mathematical reality and engaging in esoteric communica-
tion succinctly captures the contrast between the mathematics traditions
established in the two classroom. If the teacher and children in the second-
grade classroom were mutually orienting each other in a taken-as-shared
mathematical reality when they engaged in mathematical discussions, then,
by definition, the mathematical norms they established while doing so were
experienced as truths. This claim that the norms were taken-as-shared
assumptions about how mathematical reality is or ought to be, in turn, im-
mediately indicates why they routinely constituted situations for explana-
tion and justification. In the third-grade classroom, we saw that a solution
or answer was acceptable if the child had followed the mathematical in-
structions correctly. Explanations and justifications were almost redundant
because the instructions were constituted as fixed rules for linking expres-
sions that did not have to signify anything beyond themselves. In contrast,
a solution or answer was acceptable in the second-grade classroom only
if the child's acts of creating and acting on mathematical objects were ac-
ceptable. Consequently, the child was obliged to describe these acts if he
or she expected others to agree that the solution or answer was legitimate.
To give such a description is to explain or to justify, depending on the in-
teractional context.
Our claim that the teacher and children were interactively constituting
a taken-as-shared mathematical reality is consistent with the teacher's im-
mediate response to Brenda's explanation:

That's right. She is telling me that if I take all of these ones and
squeeze them all together (aligns 10 cubes) I'm going to make a ten
bar or a ten, and so that's how she figured out that if you take all
those ones and make a ten out of it then you have a ten bar. It's
the same thing.

593

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
In redescribing the solution in a way that was acceptable to Brenda, the
teacher presumably judged that it was relatively sophisticated and that she
could capitalize on it as she pursued her pedagogical agenda of facilitating
the children's enculturation into the mathematical ways of knowing of the
wider society. Thus, in redescribing the solution, she was initiating and guid-
ing the classroom community's constitution of a normative mathematical
interpretation that would be compatible with those institutionalized by the
wider community. We can also note that the teacher exemplified a process
that Newman, Griffin, and Cole (1989), following Leont'ev (1978), called
appropriation. Brenda had said that you count the individual cubes and
"then take 10 and that makes 30." In contrast, the teacher, guided by her
pedagogical intentions, explained that Brenda had solved the task by squeez-
ing ones together to make a ten bar. In doing so, the teacher emphasized
a selected aspect of Brenda's activity and formulated it in a way that she
believed might be comprehensible to other children and yet was accept-
able to Brenda as a description of what she had actually done. It seems essen-
tial that the teacher make interventions of this type in that she was the on-
ly member of the classroom community who could take the mathematical
norms of the wider society as a reference and judge whether the children's
constructions would be productive with regard to their further learning and
mathematical enculturation.
In general, it would be a misrepresentation to say that the third-grade,
but not the second-grade, teacher was an authority in the classroom. The
distinction between the two teachers' instructional approaches resides in
the manner in which they expressed their institutionalized authority in ac-
tion. The third-grade teacher acted as the sole validator of her students'
interpretations and solutions as she initiated them into a realm of mathe-
matical instructions that were, for the most part, beyond explanation and
justification. It was precisely because the mathematical practices interac-
tively constituted in this classroom could not be explained or justified that
learning was synonymous with acting in accord with the teacher's expec-
tations. In contrast, the teacher and children acted together as a community
of validators in the second-grade classroom. This was apparent, for exam-
ple, during the initial exchange when Michael contended that the cardinality
of the collection was 30. The children's developing ability to assess the
legitimacy of each others' mathematical activity is closely related to the
observation that the activity was intrinsically explainable and justifiable. In
particular, they developed this ability as they participated in the interactive
constitution of situations for explanation and justification under the teacher's
guidance (cf. Haste, 1987). As a consequence, a primary way in which the
teacher expressed her authority in action was by initiating and guiding the
interactive constitution of these situations and by redescribing certain aspects
of the children's mathematical activity but not others. Thus, the children
were not obliged to attempt to act in ways that the teacher had in mind
all along. In fact, the teacher actively encouraged them to say how they
594

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
had actually interpreted and solved tasks (Cobb, Yackel, & Wood, 1989;
Wood, Cobb, & Yackel, 1990). Their enculturation into the mathematical
ways of knowing occurred as the teacher capitalized on their relatively
autonomous constructions to guide the constitution of taken-as-shared
mathematical meanings and practices.
We note in passing that the teacher's response to Brenda's explana-
tion is consistent with our contention that the teacher and children were
mutually orienting each other in a taken-as-shared mathematical reality. The
teacher's redescription took this form: "She is telling me that if I perform
these mathematical actions then I will make this mathematical object." Al-
though the teacher actually moved pieces of wood, she intended that her
physical actions would be interpreted as signifying conceptual mathematical
actions. She explicitly attempted to indicate this at one point, saying, "I'm
going to make a ten bar or a ten." More generally, the teacher's and chil-
dren's language in the remainder of this lesson, and in other lessons through-
out the school year that had arithmetical activity as their focus, typically
expressed the metaphor of acting in physical reality (Bloor, 1983; Cobb,
1985, 1989). Analyses of other lessons indicate that this was the case even
when manipulative materials were not used (Cobb, 1991).
As a further point, we also note that the exchanges involving Michael
and Brenda indicate that the teacher and children took it for granted that
there could be at least two alternative mathematical interpretations of a situa-
tion. This observation is significant in that it indicates how the teacher at-
tempted to cope with the qualitative differences in her students' develop-
ing conceptions of place value numeration. In particular, less conceptually
advanced students such as Michael could actively participate in the classroom
community's constitution of mathematical truths.
In the remaining 31 minutes of the lesson, the teacher posed four ad-
ditional tasks, during which time she and the children interactively con-
stituted 20 situations for explanation and three situations for justification.
In the course of the discussion, the mathematical act of creating concep-
tual units of ten became an increasingly explicit topic of conversation. For
example, the teacher briefly showed the children a collection of four ten
bars and 16 cubes when she posed the third task. The cubes were arranged
in collections of three, four, and five and were separated by the ten bars.
She first asked Joan to explain her answer of 56. Joan said, "Well, I know
four ten bars make 40," and then continued as follows:

Joan: Take all four of these (points to a collection of four cubes)

and one of these (points to one cube of the second col-
lection of four).
Teacher: OK. She said take off all of these and one of these (removes
five cubes), and now what do you want me to do with
them?
Joan: Put them right here (points to collection of five cubes).
Teacher: Put them right here (she places the five cubes as Joan di-

595
Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014
Cobb, Wood, Yackel, and McNeal
rected). OK. Alright, let's see what happens. OK, what
happens! Now what do we have Joan? What did we make?
Joan: That makes 10.
Teacher: That makes 10. Now, what are we going to do?
Joan: Well, I know I have five tens and we have six right there
(points to the remaining six cubes), so we should have 56.
Teacher: So it should be 56. Right. Do you agree with her?
Students: Yeah. Sure.

In the course of this exchange, Joan and the teacher interactively constituted
the explanation of Joan's solution. The teacher's role in this process was
not limited to that of expressing Joan's conceptual acts as physical acts on
the cubes. Note, for example, how she attempted to indicate the significance
of Joan's act of creating a unit of 10 by saying, "Alright, let's see what hap-
pens. OK, what happens! Now what do we have Joan? What did we make?"
She was, in effect, providing a running commentary from the perspective
of one who could judge which aspects of Joan's activity might be poten-
tially productive with respect to the children's enculturation into the math-
ematical practices of the wider society. More generally, she acted in this
way to initiate and guide the interactive constitution of selected aspects of
individual children's interpretations and solutions as taken-as-shared mathe-
matical practices by the classroom community.
Shortly after Joan had given her explanation, another child, Paul, asked
the teacher to return the cubes to their original positions. Once she had
done so, he began his explanation as follows:

Paul: I squished all these together over here (points to the two
collections of four cubes).
Teacher: What two?
Paul: These two fours.
Teacher: These two fours. OK, I like that word. I think I will have
. . . Webster add that word to his dictionary. Squish. I like
that word.

In this case, the teacher judged that Paul's use of the term squish might
be potentially productive. Her appropriation of this term can be viewed
as an attempt to guide the development of a common signifier for the con-
ceptual act of integrating individuals units into a single entity. This develop-
ment would further facilitate the interactive constitution of the act of creating
composite units as a taken-as-shared mathematical practice in the classroom
(cf. Feldman, 1987).
The exchanges involving Joan and Paul both exemplify a more general
process by which the teacher capitalized on the children's autonomous con-
structions to guide the constitution of mathematical truths while simulta-
neously initiating them into her interpretive stance towards mathematical
activity. As the school year progressed, the view that mathematics is an ac-
tivity that involves the creation and conceptual manipulation of
596

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
mathematical objects became increasingly taken as shared (Cobb, Yackel,
& Wood, 1989).

School Mathematics and Inquiry Mathematics

We have analyzed social interactions in the two classrooms in order to com-
pare and contrast two qualitatively distinct classroom mathematics tradi-
tions. As in any case study analysis, we hope that the sample episodes might
serve as paradigm cases that will aid the reader as he or she attempts to
interpret life in mathematics classrooms. With regard to previous analyses
of classroom situations, our work is most compatible with that of Richards
(1991) and consequently we adopt his terminology and call the two ex-
emplified classroom mathematics traditions school mathematics and inquiry
mathematics.
In the course of analyzing these two traditions, it became apparent that
the characterization of school mathematics as the transmission of knowledge
from the teacher to passive students is inadequate. In our view, one need
not create a theoretical dualism by appealing to two incommensurable
epistemologies to account for learning in the two classroom mathematics
traditions. We have attempted to demonstrate that communication in both
traditions can be treated as a process in which the teacher and students
mutually orient their own and each other's activity within a consensual do-
main of taken-as-shared mathematical meanings and practices (cf. Maturana,
1980). We have further suggested that both teachers were authorities in
the sense that they initiated their students into particular interpretive stances.
In the course of these interactions, their students learned which mathemat-
ical activities were acceptable, which needed to be explained or justified,
and what counted as a legitimate explanation or justification.
The differences between the two traditions concern the quality of the
taken-as-shared or normative meanings and practices constituted in the two
classrooms. Consequently, students' experiences of meaningfulness as they
engaged in mathematical activity differed in the two classroom. More
specifically, we suggest that meaningfulness was relative to the classroom
mathematics traditions. This contention reflects the view that students
generally experienced an activity as meaningful if they could cope in the
interactively constituted instructional situation. In saying this, we do not
deny that the students interpreted mathematical activities as meaningful with
reference to their individual beliefs, values, and purposes. Rather, we are
arguing that they constructed these interpretive schemes in the course of
their initiation into a particular interpretive stance (cf. Toulmin, 1983).
Taken-as-shared interpretations of this type are, in fact, integral aspects of
a classroom mathematics tradition. In this regard, we note that Skemp's
(1976) discussion of instrumental learning and relational learning can be
interpreted as a characterization of normative mathematical activity in the
school mathematics and inquiry mathematics traditions, respectively. In our
597

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
terms, students who participate in a school mathematics tradition typically
experience mathematical understanding when they can follow procedural
instructions successfully. In contrast, students who participate in an inquiry
mathematics tradition typically experience understanding when they can
create and manipulate mathematical objects in ways that they can explain
and, when necessary, justify.
The arguments we have made concerning the meaningfulness of stu-
dents' mathematical activity appear, at first glance, to be out of step with
the way that phrases such as "meaningful learning" and "learning with un-
derstanding" are often used in mathematics education. In particular, our
contention that the activity of following procedural instructions can be
meaningful for students seems to be at odds with mathematics educators'
frequent characterization of this activity as meaningless. In our view, the
primary concern when making attributions of this latter type is not to infer
the quality of students' mathematical experiences but instead to advocate
a particular vision of mathematics teaching and learning. In other words,
the language of the political arena is used in an attempt to persuade others
of the value of a particular vision. Thus, when we argue in support of pro-
posed reforms in mathematics education, we are engaging in a debate that
has at its heart the issue of whose interpretation of the process of learning
and teaching mathematics will become the official, taken-as-shared view.
In this explicitly political situation, we would be naive to restrict ourselves
to the language of cognitive or instructional theory if we expect to initiate
those outside the mathematics education community into our interpretive
stance.
Research and the School Mathematics Tradition
The comparisons and contrasts we have made between school mathematics
and inquiry mathematics have helped us clarify some of our underlying
assumptions when we characterize students' activity in the school mathe-
matics tradition as meaningless. We again use the analysis of classroom math-
ematics traditions as a backdrop to focus on implicit assumptions, this time
as they relate to two methodological claims concerning the nature of math-
ematical development. The first of these claims, which we will discuss only
briefly, is that models of students' mathematical cognitions should be writ-
ten in formal computer languages. The second is that mathematical learn-
ing can be accounted for adequately in solely sociological terms.
The first of these claims reduces to the contention that students' math-
ematical cognitions are rule-following processes. In particular, it is argued
that students complete mathematical tasks by applying instructions to mani-
pulate symbols. As has been noted elsewhere (Cobb, 1990; Dreyfus &
Dreyfus, 1988; Searle, 1984), these symbols do not signify anything beyond
themselves in the context of the program. It is the human interpreter stand-
ing outside the computational system who gives the symbols meaning in
terms of his or her own mathematical ways of knowing. One can therefore
598

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
conclude that computer-based models account for the cognitive behavior
of a student participating in an extreme version of the school mathematics
tradition. The relevance of such models to mathematics educators who want
to facilitate teachers' and students' interactive constitution of classroom tradi-
tions compatible with inquiry mathematics is, therefore, not readily
apparent.
The second methodological claim, namely, that cognitive models are
generally irrelevant to analyses of students' learning in the classroom, has
been made most forcefully by Solomon (1989) and Walkerdine (1988) in
two closely related critiques of Piagetian psychology. Both reject accounts
of cognitive development that characterize it as a process of constructing
increasingly sophisticated systems of thought. The alternative accounts they
propose are premised on the claim that mathematical activity should be
viewed as a social or discursive practice rather than a cognitive process.
Thus, as Solomon puts it, "Understanding is intrinsically social; knowing
about numbers entails knowing how and when to use and respond to num-
bers according to the context in which they appear" (1989, p. 7). One clear
point of contact between their proposals and the analysis we have presented
is that both Solomon and Walkerdine use the notion of a tradition of math-
ematical practice when they analyze classroom social interactions. However,
there are two major issues of disagreement between their proposals and
our perspective that stem from their explicit rejection of a cognitive perspec-
tive on mathematical activity. The first issue concerns the view of mathe-
matical activity that emerges from their analyses. For Solomon, it is an ac-
tivity in which students learn to follow situated mathematical rules that,
in Much and Shweder's (1978) terms, have the quality of instructions.
Walkerdine, for her part, is quite explicit in her argument that the purpose
of doing mathematics in school is to produce formal statements that do
not signify anything beyond themselves. In these accounts, there is no room
for what Davis and Hersh (1981) take to be the key feature of mathematical
experience—the creation and mental manipulation of abstract mathematical
objects. It would seem that, like the proponents of computer-based models,
Solomon and Walkerdine assume that there is no alternative to the school
mathematics tradition. As we have seen, it was virtually impossible to infer
the mathematical conceptions of any individual child when we analyzed
the third-grade lesson. In this sense, Solomon and Walkerdine have a point;
models that account for students' construction of mathematical objects are
largely irrelevant when we attempt to explain their activity in this instruc-
tional setting. However, their conclusion that cognitive psychology has
nothing to contribute to analyses of the learning and teaching of mathematics
is, in our view, an overgeneralization. On the one hand, we have attempted
to illustrate that models of students' construction of mathematical objects
can make an important contribution to analyses of the learning-teaching
process as it occurs in the inquiry mathematics tradition. Individual chil-
dren's mathematical interpretations of the task at hand and of others' ac-

599
Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014
Cobb, Wood, Yackel, and McNeal
tivity profoundly influence the nature of the social interactions in which
they participate (Yackel, Cobb, & Wood, 1991). On the other hand, we
note that Solomon and Walkerdine do implicitly appeal to cognitive
psychology in the course of their analyses when they talk of students "enter-
ing into particular social practices and thus learning how to act appropriately
in various contexts" (Solomon, 1989, p. iv). It is a psychology that ignores
individual experience and views human activity as situated rule-following
behavior.
Our second point of contention concerns Solomon's (1989) and
Walkerdine's (1988) assumption that students do not actively contribute
to the development of the classroom mathematics tradition. Instead, they
argue that the tradition exists independently of students' activity. Walker-
dine, for example, talks of students being "embedded" in the tradition. In
this view, learning is said to have occurred when students know how to
act in accord with the normative rules of the tradition. We, in contrast,
have argued that students as well as teachers actively contribute to the in-
teractive constitution of the mathematics traditions established in their
classroom (cf. Bauersfeld, 1980). This contrast can be clarified by considering
an example that Walkerdine discusses in some detail—a lesson in which
the focus is on place value numeration. In the course of the lesson, the
teacher and children engage in a variety of activities that involve the group-
ing of match sticks into bundles of ten. Walkerdine summarizes what she
considers to be the central aspect of the lesson by observing, "The teacher
is constantly, and often literally, pointing out that a numeral in a particular
place represents a different pile [of match sticks]. She manufactures a cor-
respondence between the value of a numeral and a pile of match sticks"
(pp. 170-171). Those of us who, like Walkerdine, have already constructed
a relatively sophisticated understanding of place value numeration can cer-
tainly interpret the teacher's activity in this way. We would argue that it
is also important to consider how the children interpreted the teacher's
linguistic acts. Walkerdine comments that the teacher "manufactures a cor-
respondence . . . hoping that this correspondence will be obvious to the
children" (p. 171). In the absence of an analysis of either the children's in-
terpretations or their contributions to the interaction, we, like Walkerdine
and the teacher, can only hope. The analysis of Steffe et al. (1988) indicates
that the correspondence the teacher had in mind is not at all obvious to
most young children. This and the observation that the teacher attempted
to steer the students' responses toward those she had in mind all along lead
us to speculate that many of the students might well have interpreted the
teacher's actions as instructions that they were expected to follow. Cognitive
analyses of individual children's mathematical conceptions would, of course,
clarify the situation. However, in the absence of such analyses, it seems
reasonable to suggest that the teacher and students were together interac-
tively constituting a school mathematics, rather than an inquiry mathematics,
tradition in their classroom.
600

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
The two points over which we have taken issue, namely, that the school
mathematics tradition is paradigmatic and that this tradition exists in-
dependently of children's activity, together underpin Solomon's (1989) and
Walkerdine's (1988) claim that cognitive psychology is irrelevant to analyses
of students' learning in the classroom. We have argued that both the teacher
and students actively contribute to the development of their classroom
mathematics tradition. In addition, we contend that cognitive models which
document students' construction of increasingly sophisticated mathematical
objects are essential to analyses of their activity as they participate in the
interactive constitution of an inquiry mathematics tradition.

Conclusion
In our view, one way that research can contribute to the current reform
effort in mathematics education is by developing empirically grounded
analyses of the learning-teaching process as it is interactively constituted
in classrooms. For us, the overriding goal should be to infer and account
for the development of the meanings that individual and collective mathe-
matical activities have for teachers and their students. We also suggested
that such analyses involve cognitive as well as sociological perspectives.
On the one hand, a classroom mathematics tradition does not exist apart
from the individual mathematical activities that constitute it. Cognitive ana-
lyses of the teacher's and students' interpretations of tasks and of each
other's activity are, therefore, critical. On the other hand, the classroom
mathematics tradition, as it is currently constituted, both enables and con-
strains the teacher's and students' individual mathematical interpretations
and actions. Sociological analyses are, therefore, essential even if the primary
focus is on the teacher's and students' cognitions. In this respect, cognitive
and sociological analyses are complementary in that each informs the other.
At first glance, this line of research might appear to lack practical rele-
vance. We should, therefore, stress again that the analyses must be empirical-
ly grounded. The primary reason for conducting such an analysis is, in fact,
to advance our understanding of what might be happening in mathematics
classrooms. This activity of developing theoretical constructs to make sense
of both teachers' and students' activities and the classroom mathematics
traditions they constitute is a process of reflecting on practice. In our ex-
perience, opportunities to develop insights that can contribute to the refine-
ment of classroom practices arise as one engages in activity of this type.
Consequently, we contend that detailed theoretical analyses have the poten-
tial to be of great practical relevance to the current reform effort in mathe-
matics education.

Note
The research reported in this paper was supported by the National Science Foundation
under grant No. MDR 885-0560 and by the Spencer Foundation. The opinions expressed
do not necessarily reflect the views of these foundations.

601

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

Cobb, Wood, Yackel, and McNeal
Several notions central to this paper were elaborated in the course of discussions with
Heinrich Bauersfeld, Gotz Krummheuer, and Jorg Voigt at the University of Bielefeld,
Germany.

References
Balacheff, N. (1986). Cognitive versus situational analysis of problem-solving behavior.
For the Learning of Mathematics, 6 (3), 10-12.
Balacheff, N. (1990). Towards a problematique for research on mathematics teaching.
Journal for Research in Mathematics Education, 21, 258-272.
Barnes, B. (1982). 775. Kuhn and social science. New York: Columbia University Press.
Bauersfeld, H. (1980). Hidden dimensions in the so-called reality of a mathematics
classroom. Educational Studies in Mathematics, 11, 23-41.
Bauersfeld, H., Krummheuer, G., & Voigt, J. (1988). Interactional theory of learn-
ing and teaching mathematics and related microethnographical studies. In H. G.
Steiner & A. Vermandel (Eds.), Foundations and methodology of the discipline
of mathematics education (pp. 174-188). Antwerp: Proceedings of the TME
Conference.
Bernstein, R.J. (1983). Beyond objectivism and relativism: Science, hermeneutics,
and praxis. Philadelphia: University of Pennsylvania Press.
Bloor, D. (1983). Wittgenstein: A social theory of knowledge. New York: Columbia
University Press.
Blumer, H. (1969). Symbolic interactionism. Englewood Cliffs, NJ: Prentice-Hall.
Brousseau, G. (1984). The crucial role of the didactical contract in the analysis and
construction of situations in teaching and learning mathematics. In H.G. Steiner
(Ed.), Theory of mathematics education (pp. 110-119). Occasional Paper 54.
Bielefeld, Germany: IDM.
Brown, J.S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of
learning. Educational Researcher, 18 (1), 32-42.
Bruner, H. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University
Press.
Cobb, P. (1985). Mathematical actions, mathematical objects, and mathematical sym-
bols. Journal of Mathematical Behavior, 4, 127-134.
Cobb, P., Yackel, E., & Wood, T. (1989). Young children's emotional acts while
doing mathematical problem solving. In D.B. McCleod & V.M. Adams (Eds.),
Affect and mathematical problem solving: A new perspective (pp. 117-148).
New York: Springer-Verlag.
Cobb, P. (1989). Experiential, cognitive, and anthropological perspectives in mathe-
matics education. For the Learning of Mathematics, 9 (2), 32-42.
Cobb, P. (1990). A constructivist perspective on information-processing theories of
mathematics education. International Journal of Educational Research, 14,
67-92.
Cobb, P. (1991). Reconstructing elementary school mathematics. Focus on Learn-
ing Problems in Mathematics, 13 (2), 3-32.
Cobb, P. (1992). Case study—Anne and Jack: January 1$-March23. Unpublished
manuscript, Purdue University, School Mathematics and Science Center, West
Lafayette, IN.
Cobb, P., Wood, T., & Yackel, E. (1991). Analogies from the philosophy and soci-
ology of science for understanding classroom life. Science Education, 75, 23-44.
Davis, P.J., & Hersh, R. (1981). The mathematical experience. Boston: Houghton
Mifflin.
Dreyfus, H.L., & Dreyfus, S.E. (1988). Making a mind versus modeling the brain:
Artificial intelligence back at a branchpoint. In S.R. Graubard (Ed.), The artifi-
602

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014

 Classroom Mathematics Traditions
cial intelligence debate: False starts, real foundations (pp. 15-43). Cambridge:
MIT Press.
Feldman, C.F. (1987). Thought from language: The linguistic construction of cognitive
representations. In J. Bruner and H. Haste (Eds.), Making sense: The child's con-
struction of the world (pp. 131-145). New York: Methuen.
Gergen, KJ. (1985). The social constructionist movement in modern psychology.
American Psychologist, 40, 266-275.
Haste, H. (1987). Growing into rules. In J. Brunner and H. Haste (Eds.), Making sense:
The child's construction of the world (pp. 163-195). London: Methuen.
Knorr-Cetina, K.D. (1982). The manufacture of scientific knowledge. Oxford:
Pergamon Press.
Krummheuer, G. (1983). Das Arbeitsinterim in Mathematikunterricht. In H. Bauersfeld
(Ed.), Lernen und Lehren von Mathematik (pp. 57-106). Cologne, Germany:
Aulis.
Kuhn, T.S. (1970). The structure of scientific revolutions (2nd ed.). Chicago: Univer-
sity of Chicago Press.
Laborde, L. (1989). Audacity and reason: French research in mathematics education.
For the Learning of Mathematics, 9 (3), 31-36.
Lampert, M. (1988). The teacher's role in reinventing the meaning of mathematical
knowing in the classroom. In M. Behr, C. LaCampagne, and M.M. Wheeler (Eds.),
Proceedings of the tenth annual meeting of the North America Chapter of the
International Group for the Psychology of Mathematics Education (pp. 4 3 3 -
480). DeKalb, IL: Northern Illinois University.
Lampert, M. (1990). When the problem is not the question and the solution is not
the answer: Mathematical knowing and teaching. American Educational Re-
search fournal, 27, 29-63.
Lave, J. (1988). Cognition in practice: Mind, mathematics and culture in everyday
life. Cambridge: Cambridge University Press.
Leont'ev, A.N. (1978). Activity, consciousness, and personality. Englewood Cliffs,
NJ: Prentice-Hall.
Maturana, H.R. (1980). Man and society. In F. Benseler, P.M. Hejl, and W.K. Koch
(Eds.), Autopoiesis, communication, and society (pp. 11-32). Frankfurt, Ger-
many: Campus Verlag.
Mason, J. (1989). Mathematical abstraction as the result of a delicate shift of atten-
tion. For the Learning of Mathematics, 9 (2), 2-8.
McNeal, B. (1991). The social context of mathematical development. Unpublished
doctoral dissertation, Purdue University, West Lafayette, IN.
Mead, G.H. (1934). Mind, self and society. Chicago: University of Chicago Press.
Mehan, H. (1979). Learning lessons. Cambridge, MA: Harvard University Press.
Mehan, H., & Wood, H. (1975). The reality of ethnomethodology. New York: John
Wiley.
Much, N.C., & Shweder, R.A. (1978). Speaking of rules: The analysis of culture in
breach. New Directions for Child Development, 2, 19-39.
National Council of Teachers of Mathematics. (1989). Curriculum and evaluation
standards for school mathematics. Reston, VA: Author.
National Research Council. (1989). Everybody counts: A report to the nation on the
future of mathematics education. Washington, DC: National Academy Press.
Newman, D., Griffin, P., & Cole, M. (1989). The construction zone: Working for
cognitive change in school. Cambridge: Cambridge University Press.
Richards, J. (1991). Mathematical discussions. In E. von Glasersfeld (Ed.), Radical
constructivism in mathematics education (pp. 13-52), Dordrecht, Netherlands:
Kluwer.

603
Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014
Cobb, Wood, Yackel, and McNeal

Rorty, R. (1979). Philosophy and the mirror of nature. Princeton: Princeton Univer-
sity Press.
Searle, J. (1984). Minds, brains, and science. Cambridge: Harvard University Press.
Skemp, R. (1976). Relational understanding and instrumental understanding. Mathe-
matical Teaching, 77, 1-7.
Solomon, Y. (1989). The practice of mathematics. London: Routledge.
Steffe, L.P., Cobb, P., & von Glasersfeld, E. (1988). Construction of arithmetical
meanings and strategies. New York: Springer-Verlag.
Steinbring, H. (1989). Routine and meaning in the mathematics classroom. For the
Learning of Mathematics, 8 (1), 24-33.
Toulman, S. (1983). The construal of reality: Criticism in modern and postmodern
science. In WJ.T. Mitchell (Ed.), The politics of interpretation (pp. 219-237).
Chicago: University of Chicago Press.
Tymoczko, T. (1986). Introduction. In T. Tymoczko (Ed.), New directions in the
philosophy of mathematics (pp. xiii-xvii). Boston: Birkhauser.
Voigt, J. (1985). Patterns and routines in classroom interaction. Recherches en Didac-
tique des Mathematiques, 6, 69-118.
Voigt, J. (1989). The social constitution of the mathematics province—A microeth-
nographical study in classroom interaction. The Quarterly Newsletter of the
Laboratory of Comparative Human Cognition, 11 (1 & 2), 27-34.
Voigt, J. (1990, August). The microethnographical investigation of the interactive
constitution of mathematical meaning. Paper presented at the Second Bratislava
Symposium on Mathematics Education, Bratislava, Czechoslovakia.
Walkerdine, V. (1988). The mastery of reason: Cognitive development and the pro-
duction of rationality. London: Routledge.
Wood, T., Cobb, P., & Yackel, E. (1990). The contextual nature of teaching: Mathe-
matics and reading instruction in one second-grade classroom. Elementary
School Journal, 90 (5) 497-514.
Yackel, E., Cobb, P., & Wood, T. (1991). Small group interactions as a source of
learning opportunities in second grade mathematics. Journal for Research in
Mathematics Education, 22, 390-408.

Received February 6, 1991

Accepted September 19, 1991
Final revision received March 8, 1992

604

Downloaded from http://aerj.aera.net at UNIVERSITAETBIBLIOTHEK on May 9, 2014