Вы находитесь на странице: 1из 14

Situated Cognition and the Culture of Learning

John Seely Brown; Allan Collins; Paul Duguid

Educational Researcher, Vol. 18, No. 1. (Jan. - Feb., 1989), pp. 32-42.

Stable URL:
http://links.jstor.org/sici?sici=0013-189X%28198901%2F02%2918%3A1%3C32%3ASCATCO%3E2.0.CO%3B2-2

Educational Researcher is currently published by American Educational Research Association.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/aera.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org
Tue Jun 26 13:52:03 2007
Situated Cognition
and the Culture of Learning
JOHN SEELY
BROWN ALLANCOLLINS PAULDUGUID

T
he breach between cabulary has often been
learning a n d Many teaching practices iinplicitly assume that conceptual knoiill- is 'low and generally
which is captured by edge can be abstracted froin the situations in ulhich if is learned There is
the categories "know and used. This article argues tkaf this assumption inevitably limits enough time to
what" and "know how'" the effectiveness of suck practices. Drawing on recent research in- teach more than 100 to 200
may be a product of the to cognition as it is inanifest in euenjday activity, the authors argue words per year. Moreovert
structure and practices of Our tkat knozilledge is situated, being in part a product of f k e activity, much of what is taught turns
education 'ystem' Many context, and culture in zclhiclz it is dezleloped and used. They discuss Out to be in
methods of didactic educa- hoz~ithis viez~iof knowledge affects our undewfanding of leanz- practice. give the fol-
assume a separation be- ing, and tkey note that conuentional schooling too often ignores lowing of students'
tween knowing and doing' the influence q'scizool culture on iiihat is learned in school. A s of acquired
treating as an in- an alten~ativeto conventional practices, they propose cognitive this way:
tegral' sub- apprenticeship icollins, Bmoa, O Nezuinan, in press), zuhich M e and iny parents correlate,
stance' indepen- konors the situated nature of knozilledge. They examine tzilo ex- because iclithout t k e m 1
dent of the situations in amples of mathematics instruction that exhibit certain key features wouldn't be here.
which it is learned and used. approach
The primary concern of I zvas metic~llousabout fall-
schools often seems to be the ing off tlze cliff.
transfer of this substance, which com- as cognitive apprenticeship (Collins, Mrs. Morrozi~stiinulated tlze
prises abstract, decontextualized formal Brown, & Newman, in press) that em- Given the method, such mistakes
concepts. The activity and context in bed learning in activity and make delib- seem unavoidable, Teaching from die-
which learning takes place are thus re- erate use of the social and physical con- tionaries assumes that definitions and
garded as merely ancillary to learn- text are more in line with the under- exemplary sentences are
ing-pedagogically useful, of course, standing of learning and cognition that "piecesu of knowledge. But words and
but fundamentally distinct and even is emerging from research. sentences are not islands, entire unto
neutral with respect to what is learned. themselves. Language use would in-
Recent investigations of learning, Situated Knowledge and Learning
volve an unremitting confrontation
however, challenge this separating of Miller and Gildea's (1987) work on with ambiguity, polysemy, nuance,
what is learned from how it is learned vocabulary teaching has shown how metaphor, and so forth were these not
and used.' The activity in which knowl- the assumption that knowing and do- reso1;ed with the extralinguistic help
edge is developed and deployed, it is ing can be separated leads to a teaching that the context of an utterance pro-
now argued, is not separable from or method that ignores the way situations vides (Nunberg, 1978).
ancillary to learning and cognition. Nor structure cognition. Their work has de- Prominent among the intricacies of
is it neutral. Rather, it is an integral part scribed how children are taught words language that depend on extralinguistic
of what is learned. Situations might be from dictionary definitions and a few help are indexical words-words like I,
said to co-produce knowledge through exemplary sentences, and they have here, now, next, tomorrozu, afteni~ards,
activity. Learning and cognition, it is compared this method with the way this. Indexical terms are those that "in-
now possible to argue, are fundamen- vocabulary is normally learned outside dex" or more plainly point to a part of
tally situated. school. the situation in which communication
In this paper, we try to explain in a People generally learn words in the is being c ~ n d u c t e dThey
. ~ are not mere-
deliberately speculative way, why ac- context of ordinary communication. ly context-sensitive; they are completely
tivity and situations are integral to This process is startlingly fast and suc- context-dependent. Words like I or now,
cognition and learning, and how dif- cessful. Miller and Gildea note that by
ferent ideas of what is appropriate listening, talking, and reading, the
learning activity produce very different average 17-year-old has learned vo- JOHN SEELY BROWN and PAUL DUGUID are
results. We suggest that, by ignoring cabulary at a rate of 5,000 words per at the Institute for Research on Leanzing,
the situated nature of cognition, educa- year (13 per day) for over 16 years. By 2550 Hanover Street, Palo Alto, California
tion defeats its own goal of providing contrast, learning words from abstract 94304. ALLAN C OLLINS is at BBN Inc, 10
useable, robust knowledge. And con- definitions and sentences taken out of Moulton Street, Cambridge, Massachusetts
versely, we argue that approaches such the context of normal use, the way vo- 02238.

32 EDUCATIONAL RESEARCHER
for instance, can only be interpreted in share several significant features with appropriately without understanding
the context of their use. surprisingly, knowledge: They can only be fully un- the community or culture in which it is
all words can be seen as at least partially derstood through use, and using them used.
indexical (Barwise & Perry, 1983). entails both changing the user's view Conceptual tools similarly reflect the
Experienced readers implicitly under- of the world and adopting the belief cumulative wisdom of the culture in
stand that words are situated. They, system of the culture in which they are which they are used and the insights
therefore, ask for the rest of the used. and experience of individuals. Their
sentence or the context before commit- First, if knowledge is thought of as meaning is not invariant but a product
ting themselves to an interpretation of tools, we can illustrate Whitehead's of negotiation within the community.
a word. And they go to dictionaries (1929) distinction between the mere ac- Again, appropriate use is not simply a
with situated examples of usage in quisition of inert concepts and the de- function of the abstract concept alone.
mind. The situation as well as the dic- velopment of useful, robust knowl- It is a function of the culture and the
tionary supports the interpretation. But edge. It is quite possible to acquire a activities in which the concept has been
the students who produced the sen- tool but to be unable to use it. Similar- developed. Just as carpenters and cab-
tences listed had no support from a nor- ly, it is common for students to acquire inet makers use chisels differently, so
mal communicative situation. In tasks algorithms, routines, and decontex- physicists and engineers use mathemat-
like theirs, dictionary definitions are tualized definitions that they cannot use ical formulae differently. Activity, con-
assumed to be self-sufficient. The ex- and that, therefore, lie inert. Unfor- cept, and culture are interdependent.
trahnguistic props that would structure, tunately, this problem is not always ap- No one can be totally understood with-
constrain, and ultimately allow inter- parent. Old-fashioned pocket knives, out the other two. Learning must in-
pretation in normal communication are for example, have a device for remov- volve all three. Teaching methods often
ignored. ing stones from horses' hooves. People try to impart abstracted concepts as
Learning from dictionaries, like any with this device may know its use and fixed, well-defined, independent en-
method that tries to teach abstract con- be able to talk wisely about horses, tities that can be explored in proto-
cepts independently of authentic situa- hooves, and stones. But they may typical examples and textbook exer-
tions, overlooks the way understand- never betray-or even recognize-that cises. But such exemplification cannot
ing is developed through continued, they would not begin to know how to provide the important insights into
situated use. This development, which use this implement on a horse. Similar- either the culture or the authentic ac-
involves complex social negotiations, ly, students can often manipulate tivities of members of that culture that
does not crystallize into a categorical algorithms, routines, and definitions learners need.
definition. Because it is dependent on they have acquired with apparent com- To talk about academic disciplines,
situations and negotiations, the mean- petence and yet not reveal, to their professions, or even manual trades as
ing of a word cannot, in principle, be teachers or themselves, that they would communities or cultures will perhaps
captured by a definition, even when the have no idea what to do if they came seem strange. Yet communities of prac-
definition is supported by a couple of upon the domain equivalent of a limp- titioners are connected by more than
exemplary sentences. ing horse. their ostensible tasks. Thev are bound
A11 knowledge is, we believe, like lan- People who use tools actively rather by intricate, socially constructed webs
guage. Its constituent parts index the than just acquire them, by contrast, of belief, which are essential to under-
world and so are inextricably a product build an increasingly rich implicit standing what they do (Geertz, 1983).
of the activity and situations in which understanding of the world in which The activities of many communities are
they are produced. A concept, for ex- they use the tools and of the tools unfathomable, unless they are viewed
ample, will continually evolve with themselves. The understanding, both from within the culture. The culture
each new occasion of use, because new of the world and of the tool, continual- and the use of a tool act together to
situations, negotiations, and activities ly changes as a result of their interac- determine the way practitioners see the
inevitably recast it in a new, more tion. Learning and acting are interest- world; and the way the world appears
densely textured form. So a concept, ingly indistinct, learning being a con- to them determines the culture's under-
like the meaning of a word, is always tinuous, life-long process resulting from standing of the world and of the tools.
under construction. This would also ap- acting in situations. Unfortunately, students are too often
pear to be true of apparently well-de- Learning how to use a tool involves asked to use the tools of a discipline
fined, abstract technical concepts. Even far more than can be accounted for in without being able to adopt its culture.
these are not wholly definable and defy any set of explicit rules. The occasions To learn to use tools as practitioners use
categorical description; part of their and conditions for use arise directly out them, a student, like an apprentice,
meaning is always inherited from the of the context of activities of each com- must enter that community and its cul-
context of use. munity that uses the tool, framed by ture. Thus, in a significant way, learn-
the way members of that community ing is, we believe, a process of encul-
Learning and tools. To explore the idea see the world. The community and its turation.
that concepts are both situated and pro- viewpoint, quite as much as the tool
gressively developed through activity, itself, determine how a tool is used. Learning and enculturation. Encul-
we should abandon any notion that Thus, carpenters and cabinet makers turating may, at first, appear to have lit-
they are abstract, self-contained entities. use chisels differently. Because tools tle to do with learning. But it is, in fact,
Instead, it may be more useful to con- and the way they are used reflect the what people do in learning to speak,
sider conceptual knowledge as, in some particular accumulated insights of com- read, and write, or becoming school
ways, similar to a set of tools.4 Tools munities, it is not possible to use a tool children, office workers, researchers,
and so on. From a very early age and appear informal, but it is nonetheless Archetypal school activity is very dif-
throughout their lives, people, con- full-blooded, authentic activity that can ferent from what we have in mind
sciously or unconsciously, adopt the be deeply informative-in a way that when we talk of authentic activity, be-
behavior and belief svstems of new textbook examples and declarative ex- cause it is very different from what
social groups. Given the chance to planations are not. authentic ~ractitioners do. When
observe and practice in situ the behavior authentic activities are transferred to the
Authentic Activity classroom, their context is inevitably
of members of a culture, people pick up
relevant jargon, imitate behavior, and Our case so far rests on an undefined transmuted: thev become classroom
gradually start to act in accordance with distinction between authentic and tasks and part of the school culture.
its norms. These cultural practices are school activity. If we take learning to be Classroom procedures, as a result, are
often recondite and extremely complex. a process of enculturation, it is possible then applied to what have become
Nonetheless, given the opportunity to to clarify this distinction and to explain classroom tasks. The system of learn-
observe and practice them, people why much school work is inauthentic ing and using (and, of course, testing)
adopt them with great success. Stu- and thus not fully productive of useful thereafter remains hermetically sealed
dents, for instance, can quickly get an learning. within the self-confirmine " culture of the
imulicit sense of what is suitable dic- The activities of a domain are framed school. Consequently, contrary to the
tion, what makes a relevant question, by its culture. Their meaning and pur- aim of schooling, success within this
what is legitimate or illegitimate pose are socially constructed through culture often has little bearing on per-
behavior in a particular activity. The negotiations among present and past formance elsewhere.
ease and success with which people do members. Activities thus cohere in a Math word problems, for instance,
this (as opposed to the intricacy of way that is, in theory, if not always in are generally encoded in a syntax and
describing what it entails) belie the im- practice, accessible to members who diction that is common only to other
mense importance of the process and move within the social framework. math problems. Thus the word prob-
obscures the fact that what they pick up These coherent, meaningful, and pur- lems of a textbook of 1478 are instantly
is a product of the ambient culture poseful activities are authentic, accord- recognizable today (Lave, 1988~).But
rather than of explicit teaching. ing to the definition of the term we use word problems are as foreign to
Too often the practices of contem- here. Authentic activities then, are most authentic math practice as Miller and
porary schooling deny students the simply defined as the ordinary practices Gildea's example of dictionary learning
chance to engage the relevant domain of the culture. is to the practices of readers and
culture, because that culture is not in This is not to say that authentic ac- writers. By participating in such ersatz
evidence. Although students are shown tivity can only be pursued by experts. activities students are likelv to miscon-
the tools of many academic cultures in Apprentice tailors (Lave, 1988a), for in- ceive entirely what practitioners actual-
the course of a school career, the per- stance, begin by ironing finished gar- ly do. As a result, students can easily
vasive cultures that they observe, in ments (which tacitly teaches them a lot be introduced to a formalistic, intimi-
which they participate, and which about cutting and sewing). Ironing is dating view of math that encourages a
some enter quite effectively are the simple, valuable, and absolutely culture of math phobia rather than one
cultures of school life itself. These authentic. Students of Palincsar and of authentic math activitv.
cultures can be unintentionally anti- Brown's (1984) reciprocal teaching of In the creation of classroom tasks, ap-
thetical to useful domain learning. The reading may read elementary texts, but parently peripheral features of authen-
ways schools use dictionaries, or math they develop authentic strategies that tic tasks-like the extralinguistic sup-
formulae, or historical analysis are very are recognized by all readers. The stu- ports involved in the interpretation of
different from the ways practitioners dents in Miller and Gildea's study, by communication-are often dismissed as
use them (Schoenfeld, in press). Thus, contrast, were given a strategy that is "noise" from which salient features can
students may pass exams-(a distinctive a poor extrapolation of experienced be abstracted for the purpose of teach-
part of school cultures) but still not be readers' situated use of dictionaries. ing. But the context of activity is an ex-
able to use a domain's conceptual tools School activity too often tends to be traordinarily complex network from
in authentic practice. hybrid, implicitly framed by one cul- which practitioners draw essential sup-
This is not to suggest that all students ture, but explicitly attributed to another. port. The source of such support is
of math or history must be expected to Classroom activity very much takes often only tacitly recognized by practi-
become professional mathematicians or place within the culture of schools, al- tioners, or even by teachers or de-
historians, but to claim that in order to though it is attributed to the culture of signers of simulations. Classroom tasks,
learn these subjects (and not just to readers, writers, mathematicians, his- therefore, can completely fail to provide
learn about them) students need much torians, economists, geographers, and the contextual features that allow
more than abstract conceuts and self- so forth. Many of the activities students authentic activity. At the same time,
contained examples. They need to be undertake are simply not the activities students may come to rely, in impor-
exposed to the use of a domain's con- of practitioners and would not make tant but little noticed ways, on features
ceptual tools in authentic activity-to sense or be endorsed by the cultures to of the classroom context. in which the
teachers acting as practitioners and us- which they are attributed. This hybrid task is now embedded, that are wholly
ing these tools in wrestling with prob- activity, furthermore, limits students' absent from and alien to authentic ac-
lems of the world. Such activity can access to the important structuring and tivity. Thus, much of what is learned
tease out the way a mathematician or supporting cues that arise from the con- in school may apply only to the ersatz
historian looks a t the world and solves text. What students do tends to be er- activity, if it was learned through such
emergent problems. The process may satz activity. activity.

34 EDUCATIONAL RESEARCHER
Activities of students, practitioners, features of JPF, practitioner, and setting. A t no time did the Weight
and just plain folks. The idea that most putative student behavior. Watcher check his procedure against
school activity exists in a culture of its This Table is intended mainly to a paper and pencil algorithm, which
own is central to understanding many make apparent that, in our terms, there would have produced Yi cup x 2/3 cup
of the difficulties of learning in school. is a great similarity between JPFs' and = ?hC U P lnstead, the coincidence of
Jean Lave's ethnographic studies of practitioners' activity. Both have their the problern, setting, and enactment
learning and everyday activity (1988b) activities situated in the cultures in zvas the means by zihich checking
reveal how different schooling is from which they work, within which they took place. (p. 165)
the activities and culture that give negotiate meanings and construct un-
The dieter's solution path was ex-
meaning and purpose to what students derstanding. The issues and problems
learn elsewhere. Lave focuses on the that they face arise out of, are defined tremely expedient and drew on the sort
behavior of JPFs (just plain folks) and of inventiveness that characterizes the
by, and are resolved within the con-
activity of both JPFs and practitioners.
records that the ways they learn are straints of the activity they are pursuing.
quite distinct from what students are Lave's work (198%) provides a good It reflected the nature of the activity, the
asked to do. resources available, and the sort of
example of a JPF engaged in authentic
resolution required in a way that prob-
Three categories primarily concern us activity using the context in which an
lem solving that relies on abstracted
here: JPFs, students, and practitioners. issue emerged to help find a resolution.
The example comes from a study of a knowledge cannot.
Put most simply, when JPFs aspire to
learn a particular set of practices, they Weight Watchers class, whose partici- This inventive resolution depended
have two apparent options. First, they pants were preparing their carefully on the dieter seeing the problem in the
can enculturate through apprentice- regulated meals under instruction. particular context, which itself was
ship. Becoming an apprentice doesn't embedded in ongoing activity. And this
In this case they were to fix a serv-
involve a qualitative change from what again is characteristic of both JPFs and
ing of cottage cheese, supposing the experts. The dieter's position gave him
JPFs normally do. People enculturate amount laid out for the meal zvas
into different communities all the time. privileged access to the solution path he
three-quarters of the tzvo-thirds cup
The apprentices' behavior and the JPFs' chose. (This probably accounts for the
the program allozoed. The problem
behavior can thus be thought of as pret- certainty he expressed before beginning
solver in this example began the task
ty much the same.5 his calculation.)He was thus able to see
muttering that he had taken a cal-
the problem and its resolution in terms
The second, and now more conven- culus course in college. . . . Then after
tional, option is to enter a school as a of the measuring cup, cutting board,
a pause he suddenly announced that
and knife. Activity-tool-culture
student. Schools, however, do seem to he had "got it!" From then on he ap-
(cooking-kitchen utensils-dieting)
demand a qualitative change in be- peared certain he zms correct, even
havior. What the student is expected to moved in step throughout this pro-
before carrying out the procedure. He
do and what a JPF does are sigruficantly cedure because of the way the problem
filled a measuring-cup tzilo-thirds full
different. The student enters the school was seen and the task was performed.
of cottage cheese, dumped it out on
culture while ostensibly being taught The whole micro-routine simply be-
the cutting board, patted it into a cir-
something else. And the general came one more step on the road to a
cle, marked a cross on it, s c q e d azuay
meal.6Knowing and doing were inter-
strategies for intuitive reasoning, re- one quadrant, and semed the rest.
solving issues, and negotiating mean- locked and inseparable.
Thus, "take t h ree-quarters of tzvo-
ing that people develop through every- thirds of a cup of cottage cheese" was This sort of problem solving is carried
day activity are superseded by the pre- not just the problem statement but out in conjunction with the environ-
cise, well-defined problems, formal also the solution to the problern and ment and is quite distinct from the pro-
definitions, and symbol manipulation the procedure for solving it. The set- cessing solely inside heads that many
of much school activity. ting zilas part of the calculating pro- teaching practices implicitly endorse. By
We try to represent this discontinui- cess and the solution zijas simply the off-loading part,of the cognitive task on-
ty in Table 1, which compares salient problem statement, enacted with the to the environment, the dieter auto-
matically used his environment to help
solve the problem. His actions were not
in any way exceptional; they resemble
TABLE I . many ordinary working practices.
JPF, Practitioner, and Student Activity Scribner (1984) records, for instance,
how complex calculations can be per-
JPFs Students Practitioners formed by practitioners using their en-
vironment directly. In the case she
reasoning with: causal stories laws causal models studied, dairy loaders used the con-
acting on: situations symbols conceptual situations figuration of crates they were filling and
resolving: emergent problems well-defined ill-defined emptying almost like an elaborate
and dilemmas problems problems abacus. Nor are such problem solving
producing: negotiable meaning fixed meaning negotiable strategies limited to the physical or
& socially & immutable meaning social environment. This sort of reliance
constructed concepts & socially on situations can be seen in the work
understanding constructed
understanding
of physicists, who see "through" for-
mulae by envisioning a physical situa-

JANUARY-FEBRUARY 1989 35
tion, which then provides support for thus implicitly devalues not just indi- their reference (see, for instance, Rubin,
inferences and approximations (deKleer vidual heuristics, which may be fragile, 1980, on the difference between speech
& Brown, 1984). Hutchins' (in press) but the whole process of inventive and writing).
studv of intricate collaborative -naval problem solving. Lave (1988~)describes Perhaps the best way to discover the
navigation records the way people dis- how some students feel it necessary to importance and efficiency of indexical
tribute the burden across the environ- disguise effective strategies so that terms and their embedding context is
ment and the group as well. The result- teachers believe the problems have to imagine discourse without them.
ing cognitive activity can then only be been solved in the approved way. Authors of a collaborative work such as
explained in relation to its context. this one will recognize the problem if
"[wlhen the context of cognition is ig- Structuring activity. Authentic activity, they have ever discussed the paper
nored," Hutchins observes, "it is im- as we have argued, is important for over the phone. "What you say here"
possible to see the contribution of struc- learners, because it is the only way they is not a very useful remark. Here in this
ture in the environment, in artifacts, gain access to the standpoint that en- setting needs an elaborate description
and in other people to the organization ables practitioners to act meaningfully (such as "page 3, second full para-
of mental processes. and purposefully. It is activity that graph, fifth sentence," beginning. . . )
Instead of taking problems out of the shapes or hones their tools. How and and can often lead to conversations at
context of their creation and providing why remain to be explained. Activity cross purposes. The problem gets
them with an extraneous framework, also provides experience, which is harder in conference calls when you be-
JPFs seem particularly adept at solving plainly important for subsequent ac- comes as ambiguous as here is unclear.
them within the framework of the con- tion. Here, we try to explain some of The contents of a shared environment
text that produced them. This allows the products of activity in terms of make a central contribution to conver-
JPFs to share the burdens of both de- idiosyncratic "indexicalized" represen- sation.
fining and solving the problem with the tations. When the immediacy of indexical
task environment as they respond in Representations arising out of activi- terms is replaced by descriptions, the
"real time." The adequacy of the solu- ty cannot easily (or perhaps at all) be nature of discourse changes and under-
tion they reach becomes apparent in replaced by descriptions. Plans, as standing becomes much more proble-
relation to the role it must play in allow- Suchman argues (1987), are distinct matic. Indexical terms are virtually
ing activity to continue. The problem, from situated actions. Most people will transparent. They draw little or no at-
the solution, and the cognition involved agree that a picture of a complex tention to themselves. They do not
in getting between the two cannot be machine in a manual is distinctly dif- necessarily add significantly to the dif-
isolated from the context in which they ferent from how the machine actually ficulty of understanding a proposition
are embedded. looks. (In an intriguing way you need in which they occur, but simply point
Even though students are expected to the machine to understand the manual, to the subject under discussion, which
behave differently, they inevitably do as much as the manual to understand then provides essential structure for the
behave like the JPFs they are and solve the machine.) The perceptions resulting discourse. Descriptions, by comparison,
most of their problems in their own from actions are a central feature in are at best translucent and at worst
situated way. Schoenfeld (in press) de- both learning and activity. How a per- opaque, intruding emphatically be-
scribes mathematics students using son perceives activity may be deter- tween speakers and their subjects. The
well-known but unacknowledged strat- mined by tools and their appropriated audience has first to focus on the de-
egies, such as the position of a problem use. What they perceive, however, con- scriptions and try to interpret them and
in a particular section of the book (e.g., tributes to how they act and learn. Dif- find what they might refer to. Only
the first questions at the end of chapters ferent activities produce different index- then can the proposition in which they
are always simple ones, and the last icalized representations not equivalent, are embedded be understood. (How-
usually demand concepts from earlier universal ones. And, thus, the activity ever elaborate, a description does not
chapters) or the occurrence of a par- that led to those representations plays merely replace the indexical word.) The
ticular word in the problem (e.g., "left" a central role in learning. more elaborate the description is in an
signals a subtraction problem), to find Representations are, we suggest, in- attempt to be unambiguous, the more
solutions quickly and efficiently. Such dexicalized rather in the way that lan- opaque it is in danger of becoming.
ploys indicate how thoroughly learners guage is. That is to say, they are depen- And in some circumstances, the index-
really are situated, and how they al- dent on context. In face-to-face conver- ical term simply cannot be replaced
ways lean on whatever context is avail- sations, people can interpret indexical (Perry, 1979).
able for help. Within the practices of expressions (containing such words as Knowledge, we suggest, similarly in-
schooling this can obviously be very ef- 1, you, here, r~ozc?,that, etc.), because they dexes the situation in which it arises
fective. But the school situation is ex- have access to the indexed features of and is used. The embedding circum-
tremely specialized. Viewed from out- the situation, though people rarely stances efficiently provide essential
side, where problems do not come in notice the significance of the surround- parts of its structure and meaning. So
textbooks, a dependency on such ings to their understanding. The impor- knowledge, which comes coded by and
school-based cues makes the learning tance of the surroundings becomes ap- connected to the activity and environ-
extremely fragile. parent, however, when they try to hold ment in which it is developed, is spread
Furthermore, though schooling seeks similar conversations at a distance. across its component parts, some of
to encourage problem solving, it disre- Then indexical expressions become which are in the mind and some in the
gards most of the inventive heuristics problematic until ways are found to world much as the final picture on a jig-
that students bring to the classroom. It secure their interpretation by situating saw is spread across its component

36 EDUCATIONAL RESEARCHER
pieces. in a way similar to that evident-and students with the opportunity to enter
As Hutchins (in press), Pea (1988), evidently successful-in craft appren- the culture of mathematical practice.
and others point out, the structure of ticeship. In this section, we examine Schoenfeld's students bring problems
cognition is widely distributed across briefly two examples of mathematics to class that he and they investigate
the environment, both social and teaching in an attempt to illustrate how mathematically. His students can wit-
physical. And we suggest that the en- some of the characteristics of learning ness and participate in spontaneous
vironment, therefore, contributes im- that we have discussed can be honored mathematical thinking and see mathe-
portantly to indexical representations in the classroom. We use examples matics as a sense-making pursuit. This
people form in activity. These represen- from mathematics in part because that approach is distinctive because, before
tations, in turn, contribute to future ac- is where some of the most innovative graduate school, few students get the
tivity. Indexical representations de- work in teaching can be found. But we opportunity to see their teachers en-
veloped through engagement in a task firmly believe that this sort of teaching gaged in mathematical practice, yet the
may greatly increase the efficiency with is not just possible in mathematics. students are expected to understand
which subsequent tasks can be done,
if part of the-environment that struc-
tures the representations remains in-
variant. This is evident in the ability to
perform tasks that cannot be described
or remembered in the absence of the
situation. Recurring features of the en-
vironment may thus afford recurrent
sequences of actions. Memory and sub-
sequent actions, as knots in handker-
chiefs and other aides memoires reveal,
are not context-independent processes.
Routines (Agre, 1985) may well be a
product of this sort of indexicalization.
Thus, authentic activity becomes a cen-
tral component of learning.
One of the key points of the concept
of indexicality is that it indicates that
knowledge, and not just learning, is
situated. A corollary of this is that learn-
ing methods that are embedded in
authentic situations are not merely use-
ful; they are essential.
Learning Through Cognitive
Apprenticeship
We have been working toward a con-
ception of human learning and reason-
ing that, we feel, it is important for
school practices to honor. Though there
are many innovative teachers, schools,
and programs that act otherwise, pre-
valent school practices assume, more
often than not, that knowledge is indi-
vidual and self-structured, that schools
are neutral with respect to what is The tentmakers and the apprentice
learned, that concepts are abstract, rela-
tively fixed, and unaffected by the ac-
tivity through which they are acquired Schoenfeld's teaching of problem solv- the nature of that practice.
and used, and that JPF behavior should ing. Schoenfeld's teaching of problem In one case (Schoenfeld, in press), he
be discouraged. solving (1985, in press) deliberately at- and his class faced the problem of the
Cognitive apprenticeship (Collins, tempts to generate mathematical prac- magic square (see Figure 1).Though the
Brown, & Newman, in press), whose tice and to show college students how problem is relatively straightforward,
mechanisms we have, to some extent, to think mathematically about the the collaborative work involved in solv-
been trying to elucidate, embraces world, how to see the world through ing it and, importantly, in analyzing the
methods that stand in contradistinction mathematicians' eyes, and, thus, how solution helped reveal to the class the
to these practices. Cognitive appren- to use the mathematician's tools. His way mathematicians look at problems.
ticeship methods try to enculturate approach goes well beyond simply giv- The class worked collectively through
students into authentic practices ing students problem-solving strategies. a number of strategies, which, on re-
through activity and social interaction Much more importantly, it provides flection, they recognized as more gen-

JANUARY-FEBRUARY 1989 37
working forwards from an initial solu- in their community. The students' pro-
FIGURE 1 tion; using systematic generating pro- cedure parallels the story problems they
The Magic Square Problem cedures; having more than one way to had created. Eventually they find ways
solve a problem. Schoenfeld is con- to shorten the process, and they usually
Can you place the digits 1, 2, 3, 4, 5, 6, 7, 8, sistently careful to emphasize that all arrive at the standard algorithm, justi-
9 in the box below, so that the sum of the digits
along each row, each column, and each such strategies are illustrated in action, fying their findings with the stories they
diagonal is the same? The completed box is developed by the class, not declared by created earlier.
called a rnagic square. the teacher. In his classes, the belief Through this method, students de-
system is instilled in the only way it can velop a composite understanding of
be, through practice in which the stu- four different kinds of mathematical
dents actively take part. knowledge: (a) intuitive knowledge, the
kind of short cuts people invent when
Lampert's teaching of multiplication. doing multiplication problems in
Lampert (1986) also involves her stu- authentic settings; (b) computational
dents in mathematical exploration, knowledge, the basic algorithms that are
which she tries to make continuous usually taught; (c) concrete knowledge,
with their everyday knowledge. She the kind of concrete models of the
has devised methods for teaching algorithm associated with the stories
mathematics to fourth grade students the students created; and (d) principled
Noti,: From Schoenfeld, In press that lead from students' implicit under- knowledge, the principles such as asso-
standing of the world beyond the class- ciativity and commutativity that under-
room, through activity and social con- lie the algorithmic manipulations of
era1 and more powerful mathematical struction in the culture, to the sort of numbers. Lampert tries to inculcate an
robust learning that direct teaching of inseparable understanding of these
ideas. In discussing whether 9 can go
algorithms usually fails to achieve. kinds of knowledge and the connec-
in the center of the square, they de-
She starts teaching multiplication,for tions between them, and thus to bridge
veloped the ideas of "focusing on key
points that give leverage," and "ex- example, in the context of coin prob- the huge gap that emerges from much
ploiting extreme cases." Although lems, because in the community of conventional teaching between concep-
Schoenfeld may appear to be teaching fourth grade students, there is usually tual knowledge and problem solving
strategy rather than subject matter, he a strong, implicit, shared understand- activity-between, as we characterized
was, more fundamentally, building ing of coins. Next, the students create them at the beginning, knowing and
with his class a mathematical belief stories for multiplication problems, doing.
system around his own and the class's drawing on their implicit knowledge to This approach fosters procedures that
intuitive responses to the problem. delineate different examples of multipli- are characteristic of cognitive appren-
As an indication that Schoenfeld's cation. Then, Lampert helps them ticeship:
class was working in the culture of toward the abstract algorithm that By beginning with a task embedded in a
mathematics, not in the culture of everyone learns for multidigit multipli- fanziliar activity, it shows the students the
schooling, he did not have the students cation, in the context of the coin prob- legitimacy of their implicit knowledge and
stop at what, in culture of school prac- lems and stories the community has its availability as scaffolding in apparently
tice, would mark the end: an answer. created. Thus, the method presents the unfamiliar tasks.
algorithm as one more useful strategy
Are zue done? In most mathematics By pointing to different decompositions,
to help them resolve community prob-
classes the ansziler is "yes. " Early in it stresses that heuristics are not absolute,
lems. but assessed with respect to a particular
the semester, my students all say The first phase of teaching starts with
"yes," expecting me to go on to task-and that even algorithms can be as-
simple coin problems, such as "using sessed in this way.
another problem. M y answer, how- only nickels and pennies, make 82
ever, is a resounding "no. '' In most By allowing students to generate their
cents." With such problems, Lampert o z ~ nsolution paths, it helps make them con-
classes, so-called "problems" are ex- helps her students explore their implicit
ercises; you are done zilhen you've scious, creative members of the culture of
knowledge. Then, in the second phase, problem-solving mathematicians. And, in
shown that you 've mastered the rele- the students create stories for multipli-
vant technique by getting the answer. enculturating through this activity, they ac-
cation problems (see Figure 2). They quire some of the culture's tools-a shared
(Schoenfeld, in press) perform a series of decompositions and vocabulary and the means to discuss, reflect
His class's goal, by contrast, was to discover that there is no one, magical- upon, evaluate, and validate community
understand the mathematical nature of ly "right" decomposition decreed by
procedures in a collaborative process.
magic square, and it was in part by do- authority, just more and less useful
ing this that the belief system was ex- decompositions whose use is judged in Schoenfeld's approach differs prin-
emplified. The class explored other pos- the context of the problem to be solved cipally in its strong emphasis on expos-
sible magic squares and discovered gen- and the interests of the problem ing students to the authentic ways of
eral principles (e.g., an algebraic form solvers. thinking of a culture and its conceptual
for describing the squares). It also led The third phase of instruction viewpoint, as much as to its subject
to some further generalizable mathe- gradually introduces students to the matter.
matical strategies that are less common- standard algorithm, now that such an Figure 3 shows how, in the terms of
ly seen in classroom practice, such as algorithm has a meaning and a purpose cognitive apprenticeship, we can repre-

38 EDUCA TIONAL RESEARCHER


sent the progress of the students from
embedded activity to general principles FIGURE 2
of the culture. In this sequence, appren- Story Problems for Teaching Multiplication
ticeship and coaching in a domain be-
gin by providing modeling in situ and T: Can anyone give me a story that could go with this
scaffolding for students to get started multiplication. . . 12 x 47
in an authentic activity. As the students S1: There were 12 jars, and each had 4 butterflies in it.
gain more self-confidence and control, T: And if I did this multiplication and found the answer,
what would I know about those jars and butterflies?
ihev move into a more autonomous
S1: You'd know you had that many butterflies altogether.
phase of collaborative learning, where
T: Okay, here are the jars. [Draws a picture to represent
they begin to participate consciously in the jars of butterflies-see diagram.] The stars in them
the culture. The social network within will stand for butterflies. Now, it will be easier for
the culture helps them develop its lan- us to count how many butterflies there are altogether,
guage and the belief systems and pro- if we think of the jars in groups. And as usual, the
motes the process of enculturation. Col- mathematician's favorite number for thinking about
groups is?
laboration also leads to articulation of
S2: 10
strategies, which can then be discussed
T: Each of these 10 jars has 4 butterflies in it. [Drazus a
and reflected on. This, in turn, fosters loop around 10 jars.]. . .
generalizing, grounded in the students'
situated understanding. From here, T: Suppose I erase my circle and go back to looking at the 12 jars again altogether. Is there
students can use their fledgling concep- any other way I could group them to make it easier for us to count all the butterflies?
tual knowledge in activity, seeing that S6: You could do 6 and 6.
activity in a new light, which in turn T: Now, how many d o I have in this group?
leads to the further development of the S7: 24
conceptual knowledge. T: How did you figure that out?
In language learning, for instance, the S7: 8 and 8 and 8. [He puts the 6 jars together into 3 pairs, intuitively finding a grouping that made
original frail understanding of a word the figuring easier for him.]
is developed and extended through
subsequent use and social negotiation, T: That's 3 x 8. It's also 6 x 4. Now, how
many are in this group?
though each use is obviously situated.
S6: 24. It's the same. They both have 6 jars.
Miller and Gildea (1978) describe two
T: And now how many are there altogether?
stages of this process. The first, in
S8: 24 and 24 is 48.
which people learn the word and assign
T: Do we get the same number of butterflies
it a semantic category (e.g., the word as before? Why?
olive is first assigned to-the general S8: Yeah, because we have the same number
category of color words), is quickly of jars and they still have 4 butterflies in
done. The second, in which distinctions each.
within this semantic category (e.g., be-
Note. From Lampert, 1986.
tween olive and other colors) are ex-
plored as the word occurs again and
again, is a far more gradual process,
which "may never be completely fin- work at and membership in their trade. zciorlds (see Burton, Brown, & Fischer,
ished" (p. 95). This second phase of Through this process, apprentices enter 1984) can be replaced by increasing
word learning corresponds to the de- the culture of practice. So the term ap- complex enculturating environments.
velopment through activity of all con- prenticeship helps to emphasize the cen- Cognitive emphasizes that apprentice-
ceptual knowledge. The threadbare trality of activity in learning and knowl- ship techniques actually reach well be-
concepts that initially develop out of ac- edge and highlights the inherently yond the physical skills usually asso-
tivity are gradually given texture as they context-dependent, situated, and en- ciated with apprenticeship to the kinds
are deployed in different situations. culturating nature of learning. And ap- of cognitive skills more normally asso-
prenticeship also suggests the paradigm ciated with conventional schooling.
Apprenticeship and Cognition of situated modeling, coaching, and This extension is not as incompatible
The development of concepts out of fading (Collins, Brown, & Newman, in with traditional apprenticeship as it
and through continuing authentic ac- press), whereby teachers or coaches may at first seem. The physical skills
tivity is the approach of cognitive ap- promote learning, first by making ex- usually associated with apprenticeship
prenticeship-a term closely allied to plicit their tacit knowledge or by model- embody important cognitive skills, if
our image of knowledge as a tool. Cog- ing their strategies for students in our argument for the inseparability of
nitive apprenticeship supports learning authentic activity. Then, teachers and knowing and doing is correct. Certain-
in a domain by enabling students to ac- colleagues support students' attempts ly many professions with generally
quire, develop, and use cognitive tools at doing the task. And finally they em- acknowledged cognitive content, such
in authentic domain activity. Similarly, power the students to continue inde- as law, medicine, architecture, and bus-
craft apprenticeship enables apprentices pendently. The progressive process of iness, have nonetheless traditionally
to acquire and develop the tools and learning and enculturation perhaps been learned through apprenticeship.
skills of their craft through authentic argues that lncreasingly Complex Micro- Moreover, advanced graduate stu-

JANUARY-FEBRUARY 1989 39
'
McCloskey, Caramazza, & Green, 1980;
FIGURE 3 White, 1983) that students have many mis-
conceptions about qualitative phenomena in
Students' Progress from Embedded Activity to Generality physics. Teachers rarely have the opportuni-

-
apprenticeship collaboration reflection ty to hear enough of zuhat students think
to recognize when the information that is of-
WORLD1
ACTIVITY- GENERALITY fered back by students is only a sutface retell-
ing for school purposes (the handing back
[coaching [multiple
practice articu,ation of an uncomprehended tool, as zcie described
it at the beginning) that may mask deep mis-
conceptions about the physical zuorld and
problem solving strategies. Groups however,
dents in the humanities, the social lective wisdom of the community. can be efficient in drazuing out, confronting
sciences, and the physical sciences ac- The role of narratives and conversa- and discussing both misconceptions and in-
quire their extremely refined research tions is perhaps more complex than effective strategies.
skills through the apprenticeships they might first appear. An intriguing role Providing collaborative work skills.
serve with senior researchers. It is then in learning is played by "legitimate Students zuho are taught individually rather
that they, like all apprentices, must peripheral participation," where peo- than collaboratively can fail to develop skills
recognize and resolve the ill-defined ple who are not taking part directly in needed for collaborative zuork. In the col-
problems that issue out of authentic ac- a particular activity learn a great deal laborative conditions of the zuorkplace, knozci-
tivity, in contrast to the well-defined ex- from their legitimate position on the ing how to learn and zclork collaboratively
ercises that are typically given to them periphery (Lave & Wenger, in prepara- is increasingly important. If people are go-
in text books and on exams throughout tion). It is a mistake to think that im- ing to learn and zuork in conjunction with
their earlier schooling. It is at this stage, portant discourse in learning is always others, they must be given the situated op-
in short, that students no longer behave direct and declarative. This peripheral portunity to develop those skills.
as students, but as practitioners, and participation is particularly important In looking at Schoenfeld's and
develop their conceptual understanding for people entering the culture. They Lampert's teaching, in noting what we
through social interaction and collabor- need to observe how practitioners at believe are important features of their
ation in the culture of the domain, not various levels behave and talk to get a methods, and in stressing social interac-
of the school. sense of how expertise is manifest in tion and collaborative learning, we are
In essence, cognitive apprenticeship conversation and other activities. trying to show how teaching through
attempts to promote learning within the a form of apprenticeship can accom-
nexus of activity, tool, and culture that Cognitive apprenticeship and collabor- modate the new view of knowledge
we have described. Learning, both out- ative learning. If, as we propose, learn- and learning we have been outlining.
side and inside school, advances ing is a process of enculturating that is The increasing role of the teacher as a
through collaborative social interaction supported in part through social inter- master to apprentices, and the teachers'
and the social construction of knowl- action and the circulation of narrative, use of authentic domain activity as a
edge. Resnick has pointed out (1988) groups of practitioners are particularly major part of teaching will perhaps,
that throughout most of their lives peo- important, for it is only within groups once and for all, dismiss George Ber-
ple learn and work collaboratively, not that social interaction and conversation nard Shaw's scurrilous criticism of
individually, as they are asked to do in can take place. Salient features of group teachers, "He who can, does. He who
many schools. Lampert's and Schoen- learning include: cannot, teaches." His comment may
feld's work, Scardamalia, Bereiter, and Collective problem solving. Groups are then be replaced with Alexander Pope's
Steinbach's teaching of writing (1984), not just a convenient way to accumulate the hopeful "Let such teach others who
and Palincsar and Brown's (1984)work individual knowledge of their members. themselves excell."
with reciprocal teaching of reading all T h q g~ive rise synergistically to insights and
employ some form of social interaction, solutions that would not come about zclithout Conclusion-Toward an Epistemology
social construction of knowledge, and them (Schoenfeld, in preparation). of Situated Cognition
collaboration. Displaying multiple roles. Successfil Much research investigating situated
Within a culture, ideas are exchanged execution of most individual tasks requires features of cognition remains to be
and modified and belief systems de- students to understand the many different done. It is, however, already possible
veloped and appropriated through con- roles needed for carrying out any cognitive to begin serious reappraisal of the
versation and narratives, so these must task. Getting one per sot^ to be able to play assumptions about learning that under-
be promoted, not inhibited. Though all the roles entailed by authentic activity lie current classroom practice (see, for
they are often anathema to traditional and to reflect productively upon his or her example Resnick, 1988; Shanker, 1988).
schooling, they are an essential compo- petformance is one of the monumental tasks One of the particularly difficult chal-
nent of social internction and, thus, of of education. The group, however, permits lenges for research, (which exceptional
learning. They provide access to much different roles to be displayed and engenders teachers may solve independently) is
of the distributed knowledge and reflective tzarratives and discussions about determining what should be made ex-
elaborate support of the social matrix the aptness of those roles. plicit in teaching and what should be
(Orr, 1987). So learning environments Confronting ineffective strategies and left implicit. A common strategy in try-
must allow narratives to circulate and misconceptions. We know from an exten- ing to overcome difficult pedagogic
"war stories" to be added to the col- sive literature (disessa, 1982, 1983, 1986; problems is to make as much as possi-

40 EDUCATIONAL RESEARCHER
ble explicit. Thus, we have ended UD deeply indebted to her groundbreaking work. ing research into a theory of computation
with ;holly inappropriate methods df *The dictionary definitions that the students and semantics built on notions o f situafed-
used in writing these sentences are as follows:
teaching. Whatever the domain, expli- Correlate-be related one to the other; ness, embeddedness, and embodiedness; of
cation often lifts implicit and possibly meticulous-very careful; stimulate-stir up. Susan Stucky's important nezo idealization
even nonconceptual constraints They were given these definitions with little of mind in tenns of "radical" efficienc~y
(Cussins, 1988) out of the embedding or no contextual help, so it would be unfair rather than rationalif!/;and also of the zoork
world and tries to make them explicit to regard the students as foolish for using the on indexicality of Philip Agre and David
words as they did.
or conceptual. These now take a place the linguistics literature, the term deisis Chapman.
in our ontology and become something is often used instead of indexicality. See, for This research ~ L J U Ssupported in part by the
more to learn about rather than simply example, J. Fillmore, Santa Cruz Lectures. Personnel and Training Research Programs,
something useful in learning. But in- ?This image is, of course, not original. For Psychological Sciences Division, Office of
the way it is developed here, we are particular-
dexical representations gain their effi- ly indebted to Richard Burton, who explored Naval Research under Contract NO.
ciency by leaving much of the context it during a symposium on education organized N00014-C-85-0026. Contract Authority
underrepresented or implicit. Future by the Secretary of Education of Kentucky and ldentification Number, N R 667-540.
work into situated cognition, from to D. N. Perkins' book Ki~oiuledgeas Design An extended version of this article zoill ap-
which educational practices will benefit, (1986). pear as IRL report No. 88-0008 (available
5The JPF must, of course, have access to a
must, among other things, try to frame culture and become what Lave and Wenger from the lnsfifutefor Research on Learning)
a convincing account of the relationship (in preparation) call a "legitimate peripheral and as BBN Research Report 6886 (available
between explicit knowledge and im- participant." And, of course, an apprentice from Bolt Beranek G. Nezuman Inc.).
plicit understanding. usually has to do a great deal of work. We are
not trying to suggest that anything magical oc-
We have described here only a frag- curs in the process of enculturation. (Medical Agre, P. (1985). R o ~ i t i ~ l e sMIT . A1 Memo.
ment of an agenda for a fully developed interns testify to how hard it can be.) But the Barwise, K. J., &Perry, J . (1983). Sitlmtioiis and
theory of situated cognition. There re- process, we stress, is not qualitatively different attitudes. Cambridge, MA: MIT Press.
mains major theoretical work to shift from what people do all the time in adopting Burton, R., Brown, J . S., & Fischer, G . (1984).
the behavior and belief systems of their peers. Skiing as a model of instruction. In B.
the traditional focus of education. For 6To get some sense of how foreign this is to Rogoff & J. Lave (Eds.), Ezler;/da!/ ct~gtlitioi~:
centuries, the epistemology that has school tasks, it might be useful to imagine the Its dez~elop~rletit in social context. (pp. 139-150)
guided educational practice has concen- impropriety of a student's being given this Cambridge, MA: Haward University Press.
trated primarily on conceptual repre- problem and asked "Does the dieter have a Collins, A,, Brown, J . S., & Newman, S . E.
sentation and made its relation to ob- measuring cup, cutting board, and knife at (in press). Cognitive apprenticeship:
hand?" Though word problems are meant to Teaching the craft of reading, writing and
jects in the world problematic by as- ground theory in activity, the things that struc- mathematics. In L. B. Resnick (Ed.), Kilorc,-
suming that, cogrutively, representation ture activity are denied to the problem solvers. I I I ~Iearnii~,y,
~, aild instr~ictio~l: Essays iii i1oilor
is prior to all else. A theory of situated Textbooks ask students to solve supposedly of Robert Glaser. Hillsdale, NJ: Erlbaum.
cognition suggests that activity and per- "real-life" questions about people who do Cussins, A. (1988). Tile coi~i~ectiorlist corlstrirc-
very unreal things, such as driving at constant tiorl i f cotlcepts. (SSL Research Report). Palo
ception are importantly and epistemol- speeds in straight lines or filling leaking Alto, CA: Xerox Palo Alto Research Center.
ogically prior-at a nonconceptual troughs with leaking buckets. Students are deKleer, J . , & Brown, J . S . (1984). A qualita-
level-to conceptualization and that it usually not allowed to indulge in real-life tive physics based on confluences. Artificial
is on them that more attention needs to speculation. Their everyday inventiveness is lr~telligeflcc]oiirilal, 24, 1-3.
be focused. An epistemology that be- constrained by prescribing and proscribing disessa, A. (1982). Unlearning Aristotelian
ways in which the solution must be found. The physics: A study of knowledge-based
gins with activity and perception, ubiquitous Mr. Smith might, after all, wisely learning. Ci~gnitioeScience, 6, 37-75.
which are first and foremost embedded repair the hole in his bucket or fill the trough diSessa, A. (1983). Phenomenology and the
in the world, may simply bypass the with a hose. Sitting down and calculating how evolut~onof intuition. In D. Gentner and
classical problem of reference-of many journeys it will take with a leaking A. Stevens (Eds.), Mental inodeis. (pp.
bucket is probably the very last thing he would 15-33) Hillsdale, NJ: Erlbaum.
mediating conceptual representations. do. (See also Lave, 1988c.) diSessa, A. (1986). Knowledge In pieces. In G.
In conclusion, the unheralded impor- Forman & P. Pufall (Eds.), Co~lstrlict~s'isin
tance of activity and enculturation to 111 the coir~plrterage. Hillsdale, NJ: Erlbaum.

learning suggests that much common Editor's Note: In an efforf to encourage in- Engestrom, Y . (1987). Learning by expatiding.
Helsinki: Orienta-Konsultit Oy.
educational practice is the victim of an formed discussion and debate on the themes Fillmore, J . (1974). Sarlta Crlrz lectiires or7 deixis.
inadequate epistemology. A new of this article, the nezo ER zoill publish a set Bloomington, IN: Indiana University
epistemology might hold the key to a oj commentaries in the May 1989 issue. Linguistics Club.
dramatic improvement in learning and Geertz, C. (1983). Local knoic~ledge.New York:
a completely new perspective on edu- Basic Books.
Hutchins, E. (in press). Learning to navigate.
cation. Acknozoledgement: Many of the ideas in this In S. Chalkin and J . Lave (Eds.), Sitirated
paper emerged from group discussions at the Learning.
institute for Research on Learning. We are Lampert, M. (1986). Knowing, doing, and
'All work in this area is to a greater or lesser especially grateful to James Greeno, Jean teaching multiplication. Cognitiuti a ~ l d111-
degree, built upon research of activity theorists struction, 3, 305-342.
such as Vygotsky, Leontiev, and others. For Lave, Susan Nezoman, Roy Pea, and John Lave, J . (1977). Tailor-made experiments and
examples of recent work, see for instance, Rheinfrank, zoho read earlier drafts and evaluating the intellectual consequences of
Rogoff and Lave, 1984; Scribner, 1984; Hut- commented on them zoifh great care. We are apprenticeship training. T h e Q~inrterly
chins, in press; Engestrom, 1987; Lave and also grateful to Richard Burton, William Neiosletter of the 111st1tlitefir Cuinparati~ie
Wenger, in preparation; and in particular Hurnan D e z ~ e l o ~ ~ m e n1,t , 1-3.
Lave, 1977, 1988a, 1988b, 1988c, in prepara- Clancey, and Alan Schoenfeld for helpfill Lave, J . (1988a). T h e c~iltureof acquisitio11 and
tion. Anyone familiar with Jean Lave's work and insightful contributions. More general- the practice 11f ur~derstarldifig.(IRL report
on learning, apprenticeship, and everyday ly, zile zi1ould like to acknozoledge the in- 88-00087). Palo Alto, CA: Institute for Re-
cognition will realize at once that we are fluence of Brian Canfzi1ell Smith's pioneer- search on Learning.

JANUARY-FEBRUARY 1989 41
Lave, J. (1988b). Cogiiitioii in l~ractrce.Boston, 257, (3), 94-99. Perkins, D. N. (1986). Kiiozc~ledge as desigii.
MA: Cambridge. Nunberg, G. (1978). The pragrnafics of reference. Hillsdale, NJ: Erlbaum.
Lave, J . (1988~).Worii probleitis: A tiricrocosrll o f Bloomington, IN: Indiana University Lin- Perry, J . (1979). The problem of the essential
thec~ries 11j Icarr~ing. Paper presented at guistics Club. indexical. Nous, 13, 3-21.
AERA annual conference, New Orleans, Orr, J . (1987). Talking about machines. (SSL Re- Resnick, L. (1988).Learning in school and out.
LA. port). Palo Alto, CA: Xerox Palo Alto Re- Educatioiial Researcher, 16(9), 13-20.
Lave, J . (in preparation). Tailored leari~iiig: search Center. Rogoff, B., &Lave, J. (Eds.). (1984). Eueyday
Educatrolr aiid eoeyday practice ai?~oiigc rafts- I'alincsar, A. S. (1986).Metacognitive strategy cogtiitio)~:Its developmeiit in social context.
riieil rri West Africn. instruction. E ~ c e p t i o n a l Childreti, 53, Cambridge, MA: Haward University Press.
Lave, J . , & Wenger, E. (in preparation). 118-124. Rubin, A. (1980). A theoretical taxonomy of
Situated leari~ir~g:Legitiriiate yeripherni Palincsar, A. S., & Brown, A. L. (1984). the differences between oral and written
participation. Reciprocal teaching of comprehension- language. In R. J. Spiro, B. C. Bruce, &
McCloskey, M . , Caramazza, A,, & Green, B. fostering and monitoring activities. Cogiii- W. F. Brewer (Eds.), Theoretical issues in
(1980). Curvilinear motion in the absence tioti and Iiistructiori, 1, 117-175. reading comprcheiisioii (pp. 411-438). Hills-
of external forces: Naive beliefs about the Pea, R. D. (1988). Distribiited iiiteiligence in dale, NJ: Erlbaum.
motion of objects. Science, 210, 1139-1141. leariiiiig and reasonitig processes. Paper pre- Scardamalia, M., Bereiter, C., & Steinbach, R.
Miller, G . A,, & Gildea, P. M. (1987). How sented at the meeting of the Cognitive (1984). Teachability of reflective processes
children learn words. Screi~tificAinericarl, Science Society, Montreal. in written composition. Cognitiz~eScience,
8, 173-190.
Schoenfeld, A. H. (1985).Mathernotical problein
solz~iiig.Orlando, FL: Academic Press.
Schoenfeld, A. H. (in press). On mathematics
as sense-making: An informal attack on the
Educational Testing Service unfortunate divorce of formal and informal
1989-90 Fellowship Programs mathematics. In D. N. Perkins, J. Segal, &
J. Voss (Eds.), liiformal reasoningalid educa-
William H. Angoff, Director tioii. Hillsdale, NJ: Erlbaum.
Schoenfeld, A. H. (in preparation). Ideas it1 the
Educational Testing Service invites applications for the ETS Postdoctoral Fellowship air. (IRL report 88-0011).Palo Alto, CA: In-
Program and the National Assessment of Educational Progress (NAEP)Visiting Scholar stitute for Research on Learning.
Program. Scribner, S. (1984). Studying working intelli-
Programs gence. In B. Rogoff & J. Lave (Eds.), Ez~ery-
day cogiiitioii: Its development in social context
ETS Postdoctoral Fellowship Program (pp. 9-40). Cambridge, MA: Harvard Uni-
Up to four participants will conduct research for 1 year (September 1, 1989, through
versity Press.
July 3 1 , 1990) at ETS in Princeton, NJ, in association with ETS senior staff in one
Shanker, A. (1988). Exploring the missing con-
of the following areas: psychometrics, cognitive psychology, educational psychology,
nection. Nezcl York Times, E7.
statistics, higher education, technology, occupationallvocational testing, minority
Suchman, L. (1987). Plans and situated actions.
issues, testing issues, or policy studies. Stipend: $24,000.Some relocation expenses
New York: Cambridge University Press.
will be included, a s will a small allowance for an accompanying family.
White, B. (1983). Sources of difficulty in un-
Goals of t h e Program: To provide research opportunities for recent awardees of
derstanding Newtonian dynamics. Cogni-
the doctorate and to increase the number of women and minority professionals work-
ing in the areas specified above. tive Scieiice, 7, 41-65.
Whitehead, A. N. (1929). The aims ofeducation.
Who Should Apply: The program is open to any individual who holds a doctorate
Cambridge: Cambridge University Press.
in a relevant discipline and provides evidence of prior research.
National Assessment of Educational Progress (NAEP) Visiting Scholar
Program
One or two participants, using the NAEP data base, will conduct their own studies
(normally from September 1, 1989, through July 3 1 , 1990) at ETS in Princeton, NJ, Contiizued from page 10
with access to senior NAEP and ETS research staff. Studies should pertain to educa- (Eds.), lob training for youth. Columbus,
tional policy or measurement issues associated with black, Hispanic, or other minori- OH: The Ohio State University, The Na-
ty students. Stipend: set in relation to compensation at home institution. Some reloca- tional Center for Research in Vocational
tion expenses will be included, a s will a small allowance for an accompanying family. Education.
Goals of the Program: To provide research opportunities for recent awardees of National Federation of Independent Business.
the doctorate and to increase the number of women and minority professionals work. (1987). [Survey of a stratified random sam-
ing in the areas specified. ple.] Unpublished raw data.
Who Should Apply: The program is open to any individual who holds a doctorate A. C. Nielsen Company. (1987).Unpublished
in a relevant discipline and provides evidence of prior research. raw data.
How to Apply (There is no special application form.) Organization of Economic Cooperation and
Development. (1986). Living conditions in
Applicants should submit: OECD countries: A compendium of social
A resume of educational and job history, honors, awards, etc. indicators. Social Policy Studies No. 3. Paris,
A detailed description of research interests and experience, plus a description of France: Author.
the nature of the research the applicant would be interested in pursuing during Sizer, T. R. (1984). Horace's comproniise: The
the fellowship year (letter of about three pages). dilemma of the Anierican high school. Boston,
Names, addresses, and telephone numbers of three individuals who are willing to MA: Houghton Mifflin.
provide recommendations for the candidate. Slavin, R. (1983). When does cooperative
Transcripts-undergraduate and graduate. learning increase student achievement?
An applicant should specify the program for which he or she is applying. Psychological Bulletin, 99, 429-445.
Stevenson, H., Lee, S., & Stigler, J. W. (1986,
Applications for 1 9 8 9 - 9 0 must be received by ETS o n or before February). Mathematics achievement of
February 1 , 1989. All applicants will b e notified by April 3 0 , 1989. Chinese, Japanese & American children.
Contact: Direct required materials and inquiries to: Margaret B. Lamb, ETS, Mail Science, 231, 693-699.
Stop 30.8, Princeton, NJ 08541-0001; telephone (609) 734-1124. Taubman, P., & Wales, T. (1975). Education
as an investment and a screening device.
In F. T. Juster (Ed.), Education, iiicome, and
human behaz~ior.New York: McGraw-Hill.

42 EDUCATIONAL RESEARCHER
http://www.jstor.org

LINKED CITATIONS
- Page 1 of 2 -

You have printed the following article:


Situated Cognition and the Culture of Learning
John Seely Brown; Allan Collins; Paul Duguid
Educational Researcher, Vol. 18, No. 1. (Jan. - Feb., 1989), pp. 32-42.
Stable URL:
http://links.jstor.org/sici?sici=0013-189X%28198901%2F02%2918%3A1%3C32%3ASCATCO%3E2.0.CO%3B2-2

This article references the following linked citations. If you are trying to access articles from an
off-campus location, you may be required to first logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Bibliography]

Knowing, Doing, and Teaching Multiplication


Magdalene Lampert
Cognition and Instruction, Vol. 3, No. 4. (1986), pp. 305-342.
Stable URL:
http://links.jstor.org/sici?sici=0737-0008%281986%293%3A4%3C305%3AKDATM%3E2.0.CO%3B2-6

Curvilinear Motion in the Absence of External Forces: Naive Beliefs about the Motion of
Objects
Michael McCloskey; Alfonso Caramazza; Bert Green
Science, New Series, Vol. 210, No. 4474. (Dec. 5, 1980), pp. 1139-1141.
Stable URL:
http://links.jstor.org/sici?sici=0036-8075%2819801205%293%3A210%3A4474%3C1139%3ACMITAO%3E2.0.CO%3B2-I

Reciprocal Teaching of Comprehension-Fostering and Comprehension-Monitoring Activities


Annemarie Sullivan Palincsar; Ann L. Brown
Cognition and Instruction, Vol. 1, No. 2. (Spring, 1984), pp. 117-175.
Stable URL:
http://links.jstor.org/sici?sici=0737-0008%28198421%291%3A2%3C117%3ARTOCAC%3E2.0.CO%3B2-%23

The Problem of the Essential Indexical


John Perry
Noûs, Vol. 13, No. 1. (Mar., 1979), pp. 3-21.
Stable URL:
http://links.jstor.org/sici?sici=0029-4624%28197903%2913%3A1%3C3%3ATPOTEI%3E2.0.CO%3B2-F
http://www.jstor.org

LINKED CITATIONS
- Page 2 of 2 -

The 1987 Presidential Address: Learning in School and out


Lauren B. Resnick
Educational Researcher, Vol. 16, No. 9. (Dec., 1987), pp. 13-20+54.
Stable URL:
http://links.jstor.org/sici?sici=0013-189X%28198712%2916%3A9%3C13%3AT1PALI%3E2.0.CO%3B2-X