See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/265198323

4 A Sociocultural Perspective on Opportunity to

Learn

Article · January 2008

DOI: 10.1017/CBO9780511802157.004

CITATIONS READS

48 591

All content following this page was uploaded by James Paul Gee on 28 May 2015.

The user has requested enhancement of the downloaded file.

P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

4 A Sociocultural Perspective on Opportunity to Learn

James Paul Gee

introduction
The field of psychometrics has been predominant in work on testing and
assessment. For the most part, psychometrics has been strongly influenced
by traditional psychological assumptions about the mind and learning. Work
that takes a sociocultural perspective has played a much smaller role and has
heretofore made little contact with psychometrics. This chapter discusses
contributions such a sociocultural perspective has to make to issues of assess-
ment and testing, with a focus on an expanded notion of opportunity to learn
(OTL). Ensuring that all learners have had equal OTL is both an ethical pre-
requisite for fair assessment and a solid basis on which to think about educa-
tional reforms that will ensure that all children can succeed at school. It is also
a point at which mutually informing discussion can occur between people
working in psychometrics and those working on sociocultural approaches to
learning.
This section begins with a consideration of the traditional perspective
in psychology, one that views knowledge and learning through the lens of
mental representations in individuals’ heads (Clancey 1997). Even in this
traditional view, many of the types of issues that sociocultural perspectives
emphasize also arise, although in a more backgrounded way. I will then turn
to a more direct consideration of sociocultural perspectives, starting with
the relationship between learners and their learning environments. In sub-
sequent sections, I will spell out this relationship in terms of the connections
between learning and learners’ experiences in the world; how knowledge is
distributed across people and their tools; the central importance of people’s
participation in shared talk and social practices; and the nature of the special
varieties of language used in talk and participation when people learn in con-
tent areas in school – areas like math, science, social studies, and literature.
76
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 77

Finally, I will take up the nature of the relationship between culture and
participation in school practices.

the traditional view

A traditional way to view knowledge (e.g., Fodor 1975; Newell and Simon
1972; Pylyshyn 1984) is in terms of mental representations stored in the head
(“mind/brain”). These representations are the way information from the
world is stored and organized in the mind/brain and how it is processed or
manipulated. Such a perspective leads to a focus on questions about how
information gets into the head, how exactly it is organized in the head, and
how it leaves the head when people need to use it. Indeed, these questions
have played a central role in much educational research.
This traditional perspective leads rather naturally to a way of looking at
the notion of OTL. Learners have had the same OTL if they have been exposed
to the same information (“content”). If they have been exposed to the same
content, then, according to this view, they have each had the opportunity to
store this information in their heads; that is, to “learn it.”
Even in this traditional view, complexities arise. For example, there is
the problem of “prior knowledge.” Most learning theorists agree that the
representation of something new in someone’s head varies in important
ways – important for the material that is learned – on the basis of what is
already in the person’s head (Bransford, Brown, and Cocking 2000). This is
because new information has to be integrated with prior knowledge to make
sense, and new knowledge makes sense in different ways according to how it
is integrated with such prior knowledge.
New information that cannot be tied to any prior knowledge is not learned
well or at all. New information that is well integrated with prior knowledge
is more deeply learned than new information that is only superficially inte-
grated with prior knowledge. Thus, even in a traditional view, the notion
of OTL would have to consider not just the information to which learners
have been exposed, but also what prior knowledge they have brought to the
new learning encounter, because this affects the type of learning that takes
place or even if any learning occurs at all. Even according to a traditional
view, therefore, there is an unavoidable historical dimension to learning and
to questions about OTL. Looking at just the here and now will not work,
even if traditional assumptions about knowledge and learning are taken into
account.
A second complexity is what we might call the “power of representa-
tion problem.” Some ways of representing information are better for some
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

78 James Paul Gee

purposes than for others; some forms of representation are more efficient or
effective than others. A list is one way to represent information, but a princi-
ple from which each member of the list can be deduced and from which new
members of the list can be generated is more powerful for many purposes.
For example, a list of English words ending in “-ness” (e.g., “goodness,”
“happiness,” “sadness”) is less efficient, for many purposes, than a general-
ization like: “adjective + ness = an abstract noun” coupled with a “blocking
principle” in terms of which the generalization does not apply if a non-ness
word already exists for the same meaning. Thus, there is no “tallness” because
“height” already exists.
However, a learner cannot form a given representation unless he or she
has the requisite “mental” representational resources. For example, a learner
(or computer) innocent of linguistic ways to represent grammatical gener-
alizations involving morphology would have to settle for a list of “-ness”
words. Thus, even in the traditional view, learning and OTL cannot just be
a matter of the information to which one was exposed. For true and equal
OTL, learners must all have the capacity to form the required representa-
tions at the required degree of “power.” Two learners exposed to English
“-ness” words, one of whom has the grammatical representational resources
to represent morphological generalizations and the other of whom does not,
have not had the same opportunity to learn the same thing at the same level,
even if they have been exposed to the same data. Of course, grammatical
representational resources are, in large part, biologically endowed or at least
learned in a relatively uniform way across individuals (Chomsky 1986), but
no such thing is true in learning literacy, science, or other “content” areas
in school, including learning grammar at a conscious level as a “content”
area.
The power-of-representation problem shows that, even in the traditional
view, we must consider which tools – in this case, tools in the sense of repre-
sentational resources – the learner has brought to the learning encounter or
picked up there along with information (data, content). This is, of course, a
different sort of “prior knowledge” issue, in which the required prior knowl-
edge is knowledge of powerful representational schemes. Such schemes are
often tied to the disciplinary nature of what is being learned. For example,
an academic domain like biology or linguistics has its own special repre-
sentational resources (such as the one we just reviewed) in terms of which
information is more effectively or powerfully dealt with in that domain. Thus,
learning these domains – and talking about OTL in these domains – is not a
matter of mere exposure to information, but also exposure to and practice
with the requisite representational means of these domains.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 79

This last point has obvious implications for testing and assessment. A test
made up of a list of facts cannot tell us whether someone has used a list-like
representation or a more principled representation to answer the questions
on the test. It also cannot tell us whether the student has learned the represen-
tational schemes inherent in and partly definitive of the academic domain (in
terms of which such facts are generated). It may also be said that a test of such
facts puts more burden on those who have simply had to memorize the facts
than it does on those who have had the opportunity to master the sorts of
representational resources that allow these facts to be generated or inferred.
On the other hand, a test that requires knowledge of such representational
schemes is not fair if some students have only been exposed to the sorts of
facts such schemes generate but not given the opportunity to develop those
representational schemes themselves.
Once we grant the points above regarding representational resources, then,
even in the traditional view, we have to concede a nonmental aspect to knowl-
edge. In any academic domain, the representational resources it uses – the
effective ones that learning the domain entails – are not just mental enti-
ties stored in experts’ heads. These representational devices are also written
down – inscribed – in public ways in terms of words, symbols, graphs, and so
forth, on paper, and in machines and various tools the discipline uses. They
are also available in the behavior and talk of other people who are experts in
the domain. Surely an inability to understand and be able to use these pub-
lic representational resources does not auger well – even in the traditional
view – for the nature and quality of a learner’s mental representations in the
domain. Even learning the mental representations of a domain, according
to the traditional view, must have something to do with learning to use and
interact with the public ones.
A third complexity that arises, even within the traditional view, is one that
we can see clearly if we consider an analogy with language acquisition. Lan-
guage acquisition theorists have long pointed out that there is an important
difference between “input” and “intake” (Corder 1967; Ellis 1997; Gass 1997).
Input is data from the language to be learned to which the learner is exposed.
If these data are not processed (not paid attention to and used) by the learner,
they obviously have no effect. Intake is input that has been processed in ways
that can lead to learning about the language to be acquired.
With any learning, there is an input/intake problem. Even if learners have
been exposed to the same information (data, content) – thus, to the same
input –it has not necessarily been intake for all of them. There are any num-
ber of reasons why input may not actually be intake for various learners. Of
course, the “prior knowledge” problem discussed earlier can be one reason
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

80 James Paul Gee

why, for some learners, certain sorts of input are not intake. Another impor-
tant variable discussed in the second-language-acquisition literature (see Gee
2008 for a more general discussion) that can cause input not to be intake is
that a learner resists using input for social, cultural, or emotional reasons –
the learner resists learning because of some perceived threat or insult to his
or her individual, social, or cultural sense of self.
In the second-language-acquisition literature, this matter is sometimes
viewed this way: Each learner has an “affective filter” (Dulay, Burt, and
Krashen 1982; Krashen 1983). When perceived threat is low, the filter is low
and input is allowed to get in the head; that is, to become intake. When per-
ceived threat is high, the affective filter is raised and input does not get in
or is not properly processed; that is, it does not become intake. Obviously,
if the learning situation itself is what is causing the affective filter to rise for
some students, we have clear implications for OTL, because these students
are not really exposed to the same information as are other students, as the
information is now not intake for them.
The input/intake problem tells us, even in the traditional view, that we
must go beyond purely cognitive considerations in thinking about learning
and the notion of OTL. Learners whose “affective filters” have been raised
have not had the same OTL as those learners whose filters have not been
raised, despite the fact that they were exposed to the same information, the
same input. For the former, the input is not intake – thus, it does not lead to
learning. Because the affective filter is tied to social, cultural, and emotional
considerations – to learners’ views of themselves and their identities in rela-
tion to what is to be learned – such sociocultural and effective considerations
arise even in traditional views of knowledge and learning.
Our discussion so far has sought to make clear that issues of the histori-
cal trajectory of one’s learning, of the different public resources available to
learners, and the nature of learners’ social and cultural identities (and their
interpretation of the learning situation in terms of those identities), issues
often associated with sociocultural approaches to learning, arise even in the
traditional framework that stresses mental representations. Even in the tra-
ditional approach, the notion of OTL would have to broaden well beyond
mere exposure to the same information, data, or content.

the situated/sociocultural view

We have seen that in the traditional view, knowledge is viewed in terms of
mental representations stored in the head (“mind/brain”). We also have seen
that complexities arise in this view, although they tend to be backgrounded
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 81

in favor of studies of cognitive processing. We will turn now to an alternative

to the traditional view. This alternative is based on a variety of related, but
different, approaches to knowledge and learning stemming from research on
situated cognition and sociocultural approaches to language and learning.
This alternative emphasizes some of the complexities we discussed with the
traditional view and other issues closely related to them.
There is no one accepted situated/sociocultural view, but rather a vari-
ety of different perspectives developed in work using different disciplinary
lenses from areas like the learning sciences (e.g., Brown 1994), cognitive sci-
ence (Clark 1997), sociolinguistics (e.g., Gee 1992, 1996, 2004), and cognitive
anthropology (Lave and Wenger 1991) – just as there is, in reality, no single
accepted formulation of the traditional view. However, to simplify matters,
we will develop a perspective on situated/sociocultural work that seeks to
capture some of the leading themes in the body of disparate work relevant
to a situated/sociocultural viewpoint.
A situated/sociocultural viewpoint looks at knowledge and learning not
primarily in terms of representations in the head, although there is no need
to deny that such representations exist and play an important role. Rather, it
looks at knowledge and learning in terms of a relationship between an individ-
ual with both a mind and a body and an environment in which the individual
thinks, feels, acts, and interacts. Both the body and the environment tend to
be backgrounded in traditional views of knowledge and learning.
Any environment in which an individual finds him or herself is filled with
affordances. The term “affordance” (coined by Gibson 1977, 1979; see also
Norman 1988) is used to describe the perceived action possibilities posed by
objects or features in the environment. The affordances of an individual’s
environment are what the individual can perceive as feasible to, in, on, with,
or about the objects or features in that environment. Of course, an affordance
does not exist for an individual who cannot perceive its presence. Even when
an affordance is recognized, however, a human actor must also have the
capacity to transform the affordance into an actual and effective action.
Effectivities are the set of capacities for action that the individual has for
transforming affordances into action. An effectivity means that a person can
take advantage of what is offered by the objects or features in the environment.
Focusing on affordance/effectivity pairs places the focus not on the indi-
vidual or the environment but on a pairing of the two, a relationship between
them. A hammer affords certain actions (e.g., hammering) better than others
if one perceives the hammer in properly functional terms. Yet this affordance
cannot be transformed into action unless the individual has the capacity to
take up the hammer in the right way to transform its affordances into action.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

82 James Paul Gee

A door that swings out has a set of affordances for movement of a certain
sort that can be effected by individuals who have the capacity to push on the
door in the way it best affords action.
If we apply this terminology to knowledge and learning in school, we
can say that learning involves developing effectivity toward the affordances
in specific sorts of environments – for example, in environments in which
students are seeking to learn aspects of science, mathematics, social studies,
literary criticism, and so forth. Learners must come to be able to perceive
fruitful affordances and transform these action possibilities into appropriate
actions in thought, word, and deed. In this perspective, studying learning
is a matter of studying the relationship between learners and their environ-
ments. We have to ask what affordances are available in the environments
of particular learners and what effectivities they have or are developing for
transforming these affordances into action. We cannot ask only about what
is in the learner’s head.
Of course, other people (experts and peers) are one special category of
“objects” in learners’ environments. Different people with different sorts of
knowledge and skills afford different learners quite distinctive possibilities
of action through talk and shared practices, provided that learners can effect
the transformation of these resources into fruitful action and interaction.
This perspective has direct implications for how we view OTL. According
to this perspective, learners have not had the same OTL just because they
have been exposed to the same information or content. The learning and
assessment environment must afford them similar capacities of action. A
learner for whom certain objects, people, or features of the environment are
not affordances, either because the learner cannot perceive their possibilities
for action or cannot effect that action, is not being exposed to the same
environment as is a learner for whom these objects, people, or features are
true affordances open to the learner’s developed or developing effectivity.
Notice, too, that there are issues of affordance/effectivity pairings for both
learning environments and assessment environments, because both are places
where learners have to act.
The affordance/effectivity distinction is only one way to say that thinking
and learning are “situated”, that is, that we can only understand and define
them relevant to the relationships between individuals and specific environ-
ments or classes of environments. Yet environment is a complex term in this
context. For human beings, the material world and our bodies are part of our
environment; human-made tools and artifacts are part of our environment;
and other people and their actions and talk are part of our environment.
These three categories – each of which interacts and overlaps with the
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 83

others – define three related emphases in work on situated/sociocultural

approaches to knowledge and learning: embodiment, distributed cognition,
and social practices.

embodiment
The traditional view of knowledge and learning that we discussed earlier is
often connected with a closely related viewpoint that the meaning of a word
is some general concept in the head that can be spelled out in something like
a definition. For example, the word “bachelor” might be represented by a
complex concept in the head that the following definition would capture: “a
male who is not married.”
However, today there are accounts of language and thinking that are dif-
ferent. Consider, for instance, these two quotes from some recent work in
cognitive psychology:

. . . comprehension is grounded in perceptual simulations that prepare agents for

situated action. (Barsalou 1999a, 77)
. . . to a particular person, the meaning of an object, event, or sentence is what
that person can do with the object, event, or sentence. (Glenberg 1997, 3)

These two quotes are from work that is part of a “family” of related view-
points which, for want of a better name, we might call the “situated cognition”
family, which means that these viewpoints all hold that thinking is connected
to and changes across actual situations and is not always or usually a pro-
cess of applying abstract generalizations, definitions, or rules (e.g., Barsalou
1999a, 1999b; Brown, Collins, and Dugid 1989; Clancey 1997; Clark 1997,
2003; Engeström, Miettinen, and Punamaki 1999; Gee 1992; Glenberg 1997;
Glenberg and Robertson 1999; Hutchins 1995; Latour 1999; Lave 1996; Lave
and Wenger 1991; Wertsch 1998; Wenger 1998). Although there are differences
among the different members of the family, they share the viewpoint that
language and thinking are tied to people’s experiences of situated action in the
material and social world. Furthermore, these experiences are stored in the
mind/brain not in terms of language (“propositions”) but in something like
dynamic images tied to perceptions both of the world and of our own bod-
ies, internal states, and feelings. Increasing evidence suggests that perceptual
simulation is indeed central to comprehension (Barsalou 1999a, 74).
Let us use a metaphor to make clear what this viewpoint means, a metaphor
drawn from the realm of video games (Gee 2003, 2004; for similar perspec-
tives not built on a video game metaphor see Holland and Quinn 1987; see
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

84 James Paul Gee

Strauss and Quinn 1997 for a perspective from anthropological psychology;

see Barsalou 1999a, b; Glenberg 1997; Glenberg and Robertson 1999; Church-
land 1995; Churchland and Sejnowski 1992; Clark 1989, 1997, 2003 for cogni-
tivist perspectives). Video games like Deus Ex, Half-Life, Age of Mythology,
Rise of Nations, or Neverwinter Nights involve a visual and auditory world in
which a player manipulates a virtual character. Such games often come with
editors or other sorts of software with which the player can make changes
to the game world or even build a new game world. The player can make a
new landscape, a new set of buildings, or new characters. The player can set
up the world so that certain types of actions are allowed or disallowed. The
player is building a new world by using, but modifying, the original visual
images (really, the code for them) that came with the game.
One simple example of this is the way players can build new skateboard
parks in a game like Tony Hawk’s Pro Skater. Players must place ramps, trees,
grass, poles, and other things in space so that they or other players can
manipulate their virtual characters to skateboard in the park in a fun and
challenging way. In the act, the player can create problems that other players
must solve in order to skate the park successfully.
Imagine the mind works in a similar way. We have experiences in the world,
including things we have experienced in dialogue with others. Let us use as
an example experiences of weddings. These are our raw materials, like the
game with which the gamer starts. Based on these experiences, we can build
a simulated model of a wedding. We can move around as a character in the
model as ourselves, imaging our role in the wedding, or we can “play” other
characters at the wedding (e.g., the minister), imaging what it is like to be that
person. The model we build is not “neutral”; rather, the model is meant to
take a perspective on weddings. It foregrounds certain aspects of weddings
that we view as important or salient. It backgrounds other elements that we
think are less important or less salient and leaves some things out altogether.
However, we do not build just one wedding-model simulation and store
it away once and for all in our minds. Rather, we build different simulations
on the spot for different specific contexts we are in. In a given situation or
conversation involving weddings, we build a model simulation that fits that
context and helps us to make sense of it. Our models are specially built to help
us make sense of the specific situations we are in, conversations we are having,
or texts we are reading. In one case, we might build a model that emphasizes
weddings as fun, blissful, and full of potential for a long and happy future. In
another case, we might build a model that emphasizes weddings as complex,
stressful, and full of potential for problematic futures.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 85

We build our model simulations to help us make sense of things and

prepare for action in the world. We can act in the model and test which
consequences follow before we act in the real world. We can role-play other
people in the model and try to see what motivates their actions or could
follow from those actions before we respond to them in the real world. In
fact, human beings tend to want to understand objects and words in terms
of their “affordances” for actions. Take something as simple as a glass:

The meaning of the glass to you, at [a] particular moment, is in terms of the
actions available. The meaning of the glass changes when different constraints
on action are combined. For example, in a noisy room, the glass may become a
mechanism for capturing attention (by tapping it with a spoon), rather than a
mechanism for quenching thirst (Glenberg 1997, 41).

Faced with the word “glass” in a text or a glass in a specific situation, the
word or object takes on a specific meaning or significance based not just on
the model simulation we build, but also on the actions with the glass that
we see as salient in the model. In one case, we build a model simulation in
which the glass is “for drinking”; in another it is “for ringing like a bell to get
attention”; in another it is a precious heirloom in a museum that is “not for
touching.” Our models stress affordances for action so that they can prepare
us to act or not act in given ways in the real world.
We think and prepare for action with and through our model simula-
tions. They are what we use to give meaning to our experiences in the world,
and they prepare us for action in the world. They help us give meaning to
words and sentences, yet they are not language. Furthermore, because they
are representations of experience (including feelings, attitudes, embodied
positions, and various sorts of foregroundings and backgroundings of atten-
tion), they are not just “information” or “facts.” Rather, they are value-laden,
perspective-taking “games in the mind.”
Of course, talking about simulations in the mind is a metaphor that, like
all metaphors, is incorrect if pushed too far (see Barsalou 1999b for how a
similar metaphor can be cashed out and corrected by a consideration of a
more neurally realistic framework for “perception in the mind”). It should
be pointed out, though, for those who find an analogy to games trivializing,
that simulations are often used at the cutting edge of the sciences of complex
systems to form and test hypotheses, test predictions, and generate analyses.
Simulations involving multiple “players” are also widely used for learning in
the military and in workplaces where people must learn to coordinate their
skills with others.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

86 James Paul Gee

Thus, meaning is not about general definitions in the head. It is about

building specific models for specific contexts. Even words that seem to have
very clear definitions, like the word “bachelor” that we used as an example
at the beginning of this section, do not really have such clear definitions.
Meaning is not about definitions but about simulations of experience. For
example, what model simulation(s) would you bring to a situation in which
someone said of a woman, “She’s the bachelor of the group?” You might
build a simulation in which the woman was attractive, at or a little older than
marriageable age, perhaps a bit drawn to the single life and afraid of marriage,
but open to the possibilities. You would see yourself as acting in various ways
toward the woman and see her responding in various ways. The fact that
the woman was not an “unmarried man” would not stop you from giving
meaning to this utterance. Someone else, having had different experiences
than you, would form a different sort of simulation. Perhaps the differences
between your simulation and the other person’s would be big, perhaps small.
They are small if you and that person have had similar experiences in life and
larger if you have not.
If we admit the importance of the ability to simulate experiences in order
to comprehend oral and written language, we can see the importance of
supplying all children in school with the range of necessary experiences with
which they can build good and useful simulations for understanding subjects
like science. Nearly everyone will have experiences of weddings and bachelors
sufficient for building simulations with which to think and prepare for action.
Not all learners have adequate experiences with concepts like reflection and
refraction, atoms and molecules, or force and motion that will allow them
to build simulations that can serve for thinking and meaning in science.
This is clearly an important issue regarding OTL. If some children have
had experiences through which they can build and manipulate appropriate
simulations in a domain and others have only interacted with oral and written
words, the latter have only general and verbal understandings; they cannot
assign the richer and more useful meanings to words and texts that the former
can. They have not had the same opportunity to learn the material in as deep
and meaningful a way.
One issue that arises when we think of meaning as situated in actual expe-
riences people have had is generality. Of course generality is important, but
in a situated viewpoint it is often (and sometimes best) attained, at least
initially, bottom up by comparing and contrasting various specific experi-
ences that can then serve as materials for building simulations that apply
more generally to a domain. Let’s consider a specific example. The science
educator diSessa (2000) has successfully taught the algebra behind Galileo’s
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 87

principles of motion (principles related to Newton’s laws) to children in sixth

grade and beyond using a specific computer programming language called
Boxer.
The students type into the computer a set of discrete steps in the program-
ming language. For example, the first command in a little program meant
to represent uniform motion might tell the computer to set the speed of a
moving object at one meter per second. The second step might tell the com-
puter to move the object, and a third step might tell the computer to repeat
the second step over and over. Once the program starts running, the student
will see a graphical object move one meter each second repeatedly, a form of
uniform motion.
Now the student can elaborate the model in various ways. For example,
the student might add a fourth step that tells the computer to add a value a to
the speed of the moving object after each movement the object has taken (let
us say, for convenience, that a adds one more meter per second at each step).
Now, after the first movement on the screen (when the object has moved
at the speed of one meter per second), the computer will set the speed of
the object at two meters per second (adding one meter), and, then, on the
next movement, the object will move at the speed of two meters per second.
After this, the computer will add another meter per second to the speed,
and on the next movement the object will move at the speed of three meters
per second. This will repeat forever, unless the student has added a step that
tells the computer when to stop repeating the movements. This process is
obviously modeling the concept of acceleration. Of course, you can set a to
be a negative number instead of a positive one and watch what happens to
the moving object over time instead.
The student can keep elaborating the program and watch what happens at
every stage. In this process, the student, with the guidance of a good teacher,
can discover a good deal about Galileo’s principles of motion through his or
her actions in writing the program, watching what happens, and changing
the program. The student is seeing, in an embodied way, tied to action, how a
representational system that is less abstract than algebra or calculus (namely,
the computer programming language, which is actually composed of a set of
boxes) “cashes out” in terms of motion in a virtual world on the computer
screen.
An algebraic representation of Galileo’s principles is more general – basi-
cally, a set of numbers and variables that does not directly tie to actions or
movements as material things. As diSessa points out, algebra doesn’t dis-
tinguish effectively “among motion (d = rt), converting meters to inches
(i = 39.37 × m), defining coordinates of a straight line (y = mx), or a host
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

88 James Paul Gee

of other conceptually varied situations” (diSessa 2000, 32–33). They all just
look alike. He goes on to point out that “[d]istinguishing these contexts is
critical in learning, although it is probably nearly irrelevant in fluid, routine
work for experts,” (diSessa 2000, 33) who, of course, have already had many
embodied experiences using algebra for a variety of different purposes of
their own.
Once learners have experienced the meanings of Galileo’s principles about
motion in a situated and embodied way, they have understood one of the
situated meanings for the algebraic equations that capture these principles
at a more abstract level. Now these equations are beginning to take on a
real meaning in terms of embodied understandings. As learners see algebra
spelled out in additional specific material situations, they will come to master
it in an active and critical way, not just as a set of symbols to be repeated in
a passive and rote manner on tests. As diSessa puts it:

Programming turns analysis into experience and allows a connection between

analytic forms and their experiential implications that algebra and even calculus
can’t touch. (diSessa 2000, 34)

Abstract systems originally got their meanings through such embodied expe-
riences for those who really understand them. Abstraction (at least in many
important cases) rises gradually out of the ground of situated meaning and
practice and returns there from time to time, or it is meaningless to most
human beings.

distributed knowledge
In the study of knowledge and learning, a situated/sociocultural perspective
takes as its unit of analysis not the person alone, but “person plus mediating
device” (Brown, Collins, and Dugid 1989; Wertsch 1998). A mediating device
is any object, tool, or technology that a person can use to enhance perfor-
mance beyond what could be done without the object, tool, or technology. It
obviously makes little sense to ask how high a pole vaulter can jump without a
pole. Furthermore, poles made of different material enable different types of
jumps (Wertsch 1998). What learners can understand and accomplish with
diSessa’s Boxer program as a mediating device is obviously different than
what they can do without it.
When people use mediating devices, knowledge is distributed, some of
it existing in their heads, some of it existing in the ways in which they can
coordinate themselves (as bodies and in terms of social practices) with the
tools they are using, and some of it existing in the tools themselves. Other
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 89

people are also “tools” for learners when and if the learners can interact
with them so as to gain and produce mutual knowledge. One problem for
the traditional view of knowledge and learning – the view that focuses on
mental representations – is that almost all human thought and interaction is
mediated by objects, tools, technologies of various sorts, or other people.
In fact, it is clear that a mental representation itself is a mediating device.
A learner who has internalized geometry as a form of mental representation
can understand and use the laws of the pendulum better and more deeply
than one who has not (in fact, using geometry is how Galileo discovered
these laws). However, once we concede this fact, it is clear, too, that public
representations – like geometry on paper or diSessa’s Boxer program – are
just as important as mental representations in serving as mediating devices. If
we are interested in learning and OTL, we must ask which mediating devices
are available, how they are made public, and how they come to be used.
People are smarter when they use smart tools. Better yet, people are smarter
when they work in smart environments; that is, environments that contain,
integrate, and network a variety of tools, technologies, and other people,
all of which store usable knowledge. When we ask where knowledge resides
in such smart environments, the answer is that it is distributed across the
insides of individuals’ heads, their bodies, their tools and technologies; other
people; and the ways in which all of these are integrated and linked together
in a network. This perspective is common now in businesses and work-
places, less so in schools (Gee, Hull, and Lankshear 1996; Hagel and Brown
2005).
People are always parts of environments, whether they are particularly
smart ones or not. They always think and act as part of larger systems that
contain more than their own heads do. This perspective has been well cap-
tured by work in activity theory. The Russian psychologist Vygotsky (1978)
argued that human beings do not react directly to or interact directly with
the environment. Rather, human reactions and interactions are mediated by
signs (language and other symbol systems) and tools. Vygotsky went on to
argue that people learn how to use these mediating devices primarily through
social interaction. Through participation in common activities with already
adept others, people internalize the workings of their culture, their language,
and various symbols, artifacts, norms, values, and ways of acting and inter-
acting. The furniture of the human mind first exits publicly in the world of
social interaction and participation.
For activity theorists, the proper unit of analysis in studying activity, cer-
tainly including learning, is an activity system; that is, a group (“community,”
though without any connotation of people personally having to feel close to
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

90 James Paul Gee

Instruments

Subject Object Outcome

Rules Community Division of labor

Figure 4.1. Conceptual model of an activity system (Engeström 1987, 78).

each other) of actors who have a common object or goal of activity (Cole
and Engeström 1993; Engeström 1987). An activity system as a unit of analysis
connects individual, sociocultural, and institutional levels of analysis. The
study of activity ceases to be just the psychology of an individual, focusing
instead on the interaction of individuals and systems of artifacts in institu-
tional settings that develop across time.
Figure 4.1 (from Engeström 1987) models the integrated elements of an
activity system. The whole system has certain intended and unintended “out-
comes.” The outer triangle contains the integration of “instruments” (various
tools and technologies), “rules” (norms of use), and “division of labor” (the
differential expertise of different actors in the system). Various other relation-
ships in the model capture the diverse ways in which “subjects” (actors), the
“object” (goal) of the activity system, and the “community” (various types
of actors in the system) interrelate with each other and with the instruments,
rules, and division of labor.
To see the model at work, consider the example of a doctor working at a
clinic (example taken from Center for Activity Theory and Developmental
Work Research 2003). The object (goal) of the doctor’s work is the health
problems of his or her patients. The outcomes include both intended ones
like improvements in health and unintended ones like patients getting lost
in the midst of overcrowding in the clinic. The instruments include tools like
x-rays, laboratory tests, and medical records as well as medical knowledge that
is partly internalized and partly stored in books and tools. The community
consists of the various actors who constitute the staff of the clinic and its
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 91

patients. The division of labor determines the tasks and powers of the doctors,
nurses, aides, patients, and other actors in the system. Finally, various rules
and norms regulate how, when, and where various actions and interactions
take place, as well as the use of time, how outcomes are measured and assessed,
and the criteria for rewards.
The same activity system will look different if we take the point of view
of another subject (actor) in the system; for example, a nurse. Both the
doctor and the nurse share the same overall object (goal), the health care
of the patients, but they do not necessarily construe it in the same way.
Different actors, because of their different histories and different positions
in the division of labor, may very well construe the object and the other
components of the activity system in different – sometimes quite different –
ways. Yet they still must coordinate their different interpretations of the object
(goal) and the activity system as a whole to ensure that the system operates,
however well or poorly. This coordination requires continual overt and tacit
negotiation, carried out in word and deed, among the various actors in the
system.
An activity system does not exist by itself; it interacts within a network
of other activity systems. For example, our clinic may receive various rules
and instruments from management, another activity system, and in turn,
the clinic produces outcomes for other activity systems, such as insurance
companies.
If we take an activity-system view of students in a classroom, we cannot
ask only about the individual student. We have to ask what sort of activity
system the student is in, what his or her role is in it, what the system looks
like from his or her perspective, what it looks like from the perspective of
other actors (e.g., the teacher, other students) in the system, and what other
systems interact with the one the student is in. From an opportunity-to-
learn perspective, we must consider more than the information to which the
learner has been exposed. All of the other elements in the system need to
count as well, including the ways in which all of these elements mediate the
learner’s knowledge and performance.

participation
In situated/sociocultural work, activity systems have often been analyzed in
terms of the notion of a “community of practice” (Lave and Wenger 1991;
Wenger 1998). In communities of practice, people share a set of practices,
often carried out collaboratively, related to carrying out a common endeavor.
Newcomers pick up both overt and tacit knowledge through a process of
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

92 James Paul Gee

guided and scaffolded participation in the community of practice, a process

that has been compared to apprenticeship.
The term “community of practice” – which is now common in the business
literature as a way to reform work (Gee, Hull, and Lankshear 1996; Wenger
1998; Wenger, McDermott, and Snyder 2002) – has been used to cover a variety
of different social configurations in which shared practices and participation
are central (for a critique of the notion of “community of practice,” see Gee
2004 and Barton and Tusting 2005). Not all of these fit the sometimes “warm”
connotations of the word “community.” Nonetheless, the central ideas are
that people learn new practices through participation with others, that they
are networked with others and with various tools and technologies in ways
that allow them to accomplish more than they could by themselves, and that
knowledge is stored as much in the network and the practices of the group
as it is in any one person’s head.
Communities of practice – in workplaces and in some educational instan-
tiations – are beginning to take on a distinctive shape. Some salient features
are listed below. These features have commonly been mentioned in literature
seeking to reform modern workplaces to make them smarter, but they have
all been applied to reform in schools as well and are found in one form or
another in classrooms that place a premium on students generating knowl-
edge through participation in authentic practices in areas like science and
mathemetics (DuFour and Eaker 1998; Fink and Resnick 2001; Fullan and
Hargreaves 1991):

(1) Members of the community of practice are affiliated with each other
primarily through a common endeavor and shared practices and only
secondarily through ties rooted in shared culture, race, class, gender,
or ability. These latter ties – as well as other forms of diversity – are
not seen as dividers but are leveraged as differential resources for the
whole group in carrying out its common endeavors and practices.
(2) The common endeavor is organized around a whole process (involving
multiple but integrated functions), not single, discrete, or decontextu-
alized tasks carried out outside of or without knowledge of the wider
contexts that give them meaning.
(3) Members of the community of practice must all share extensive
knowledge. By “extensive knowledge,” I mean that members must
be involved with many or all stages of the endeavor; able to carry
out multiple, partly overlapping, functions; and able to reflect on the
endeavor as a whole system, not just their part in it. This shared exten-
sive knowledge also involves shared norms, values, and ways of acting
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 93

and interacting that allow the community of practice to carry out its
endeavors.
(4) Members of the community of practice also each have intensive knowl-
edge; that is, specialized and deep knowledge that goes beyond the
group’s shared extensive knowledge, which they have built up and can
supply to others who do not share it when they need aspects of it for
their own work.
(5) Much of the knowledge in the community of practice is tacit (embod-
ied in members’ mental, social, and physical coordinations with other
members, and with various tools and technologies) and distributed
(spread across various members, their shared sociotechnical practices,
and their tools and technologies) and dispersed (available offsite from
a variety of different sources).
(6) The role of leaders is to design communities of practice, continually
resource them, and help members turn their tacit knowledge into
explicit knowledge to be used to further develop the community of
practice, while realizing that much knowledge will always remain tacit
and situated in practice.

In such communities of practice, people are committed through their

immersion in practice, because it is the practice itself that gives them their
identity. Diverse individual skills and cultures are recruited as resources for
the community, not as identities that transcend the community of practice
itself.
It is clear that an activity-system perspective, with its links to the notion of
a community of practice, treats people, including learners in school, as actors
and not just as passive recipients of information (Greeno and Gresalfi, this
volume). What people do in an activity system or a community of practice is
not entirely a result of what is going on inside their heads but is contingent
on interactions among all the elements of the system in which they are acting
and interacting. As Greeno (1997, 8) has stated, “just presenting hypotheses
about the knowledge someone has acquired, considered as structures in the
person’s mind, is unacceptably incomplete, because it does not specify how
other systems in the environment contribute to the interaction.”
It is also clear from our discussion of activity systems and communities of
practice that the notion of participation in social interaction is foundational.
This is so for several reasons. First, according to this perspective, following
Vygotsky’s (1978) ideas about the Zone of Proximal Development, learning
starts when learners are first able to accomplish with others through partic-
ipation in interaction what they cannot yet accomplish on their own. Such
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

94 James Paul Gee

skills are said to be in the learner’s Zone of Proximal Development, and

these are the skills that will soon become individual accomplishments. By
the time they become individual accomplishments, learners will have inter-
nalized these skills in terms of the schemes they have seen publicly at work
in their social interactions with others using various sign systems and tools.
These skills, in this sense, retain a social element.
Second, we have seen earlier that the meanings of words and signs, if
they are to be truly useful, must be situated in experiences that learners
have had that they can simulate in their minds and from which they can
eventually build simulations that are more generally applicable. Yet learners
can only come to see how words and signs fit particular patterns of experience
if they see these words used in specific situations in ways that make clear how
they apply. Thus, models of language in use in specific situations from masters
and more expert peers is crucial for learning how to situate the meanings of
words and signs in specific ways – otherwise, learners have only general verbal
definitions as meanings that are hard to apply in specific situations.
Third, participation in social practices not only makes meanings public,
but it also allows us to ask how we can position learners in the interaction
so as to allow them to be active generators of knowledge (Greeno 1997, 1998;
Greeno and Gresalfi, this volume). We saw, at the outset of this chapter, that
even with the traditional “in the head” approach to knowledge and learning,
there is the problem that some representations are much more powerful
than others. When learners must generate knowledge as part and parcel of
social practice, they must use more powerful representational systems in their
heads and on paper – the sorts of representations that yield new results and
do not just store already-provided information. However, to produce such
knowledge, learners must be given both the authority for and the resources
with which to build knowledge (tools and interactions with masters and
more expert peers). In turn, placing participation in talk and other social
practices at the center of learning allows us to investigate the affordances and
constraints of different forms of participation, a crucial question that hardly
arises in the traditional view. To the extent that different forms of talk and
social interaction lead to different affordances and constraints for different
learners, we confront, once again, a key concern for thinking about equitable
opportunities to learn.
Fourth, one of the most important tools for learning in the content areas
like mathematics and science is mastery of the specialist varieties of language
these areas use, varieties that, as a family, are often referred to as (forms of)
“academic language” (Schleppegrell 2004). Specialist varieties of language
are crucial tools (mediating devices) for creating meanings in the academic
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 95

content areas. In this sense, learning in the content areas is a form of language
development, and like all forms of language development, it is for the most
part dependent on specific forms of social interaction with masters (people
who know the form of language) and peers (Halliday and Matthiessen 1999).
In fact, we know a good deal about how varieties of language are acquired,
knowledge that can be applied to learning in the content areas in a situated/
sociocultural perspective (Gee 2004; Schleppegrell and Colombi 2002). Let
us turn, then, to the issue of language and learning.

academic registers
Although people tend to think of a language like English as one entity, actually
it’s not one but many entities (Gee 2004, 2005). There are many different
varieties of English. Some of these are different dialects spoken in different
regions of the country or by different sociocultural groups. Some are different
varieties of language used by different occupations or for different specific
purposes; for example, the languages of carpenters, lawyers, or video game
players.
Every human being, early in life, acquires a vernacular variety of his or
her native language. This form is used for face-to-face conversation and for
“everyday” purposes. Different groups of people speak different dialects of
the vernacular, connected to their family and community. Thus, a person’s
vernacular dialect is closely connected to his or her initial sense of self and
belonging in life.
After the acquisition of their vernacular variety has begun, people often
also go on to acquire various nonvernacular specialist varieties of language
used for special purposes and activities. For example, they may acquire a
way of talking (and writing) about fundamentalist Christian theology, video
games, or bird watching. Specialist varieties of language are different – some-
times in small ways, sometimes in large ways – from people’s vernacular vari-
ety of language. Linguists often refer to these specialist varieties of language,
tied to specific tasks and identities, as “registers” (Halliday and Martin 1993).
One category of specialist varieties of language is what we can call academic
varieties of language; that is, the varieties of language connected to learning
and using information from academic or school-based content areas (Gee
2002; Halliday and Matthiessen 1999; Schleppegrell 2004; Schleppegrell and
Colombi 2002) The varieties of language used in (different branches) of
biology, physics, law, or literary criticism fall into this category.
Some texts are, of course, written in vernacular varieties of language; for
example, some letters, e-mail, and children’s books. The vast majority of texts
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

96 James Paul Gee

in the modern world, though, are not written in the vernacular but in some
specialist variety of language. People who learn to read the vernacular often
have great trouble reading texts written in specialist varieties of language. Of
course, there are some texts written in specialist varieties of language (e.g.,
nuclear physics) that many very good readers cannot read.
Specialist varieties of language, whether academic or not, often have both
spoken forms and written ones, and these may themselves differ from each
other. For example, a physicist or computer scientist can write in the language
of physics or computer science and speak a version of it, too (e.g., in a lecture).
It is obvious that once we talk about learning to read and speak specialist
varieties of language, it is hard to separate learning to read and speak this way
from learning the sorts of content or information that the specialist language
is typically used to convey. That content is accessible through the specialist
variety of language and, in turn, that content is what gives meaning to that
form of language. The two – content and language – are married (Halliday
and Matthiessen 1999).
Of course, one key area where specialist varieties of language differ from
vernacular ones is vocabulary. Yet they also often differ in syntax and dis-
course features as well (“syntax” means the internal structure of sentences;
“discourse” in this context means how sentences are related to each other
across a text and what sorts of things can or cannot be said in a particular
type of text). For example, suppose someone is studying the development
of hornworms (cute green caterpillars with yellow horns). Contrast the ver-
nacular sentence “Hornworms sure vary a lot in how well they grow” with
the (academic) specialist sentence “Hornworm growth exhibits a significant
amount of variation.”
The specialist version differs in vocabulary (e.g., “exhibits”), but it also
differs in syntactic structure. Verbs naming dynamic processes in the ver-
nacular version (e.g., “vary,” “grow”) show up as nouns naming abstract
things in the specialist version (“variation,” “growth”). The vernacular sen-
tence makes the hornworms (cute little caterpillars) the subject/topic of the
sentence, but the specialist sentence makes hornworm growth (a measurable
trait for hornworms) the subject/topic. A verb–adverb pair in the vernacular
version (“vary a lot”) turns into a verb plus a complex noun phrase in the
specialist version (“exhibits a significant amount of variation”).
Although we do not have space to pursue the matter fully here, specialist
varieties of language also differ from vernacular varieties at the discourse
level. We can see this even with our two sentences. Note that the specialist
version does not allow an emotional word like “sure” that occurs in the
vernacular version. We would not usually write or say, “Hornworm growth
sure exhibits a significant amount of variation.” There is nothing wrong with
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 97

this sentence syntactically. We just don’t normally speak or write this way in
this variety of language. It doesn’t “go with” the other things we say or write
in this variety. At the cross-sentential level, specialist languages use many
devices to connect, contrast, and integrate sentences across stretches of text
that are not used as frequently, or exactly in the same way, in vernacular
varieties of language (like the phrase “at the cross-sentential level” at the
beginning of this sentence).
Specialist languages draw, of course, on grammatical resources that exist
also in vernacular varieties of language. For example, any vernacular variety
of English can make a noun (like “growth”) from a verb (like “grow”). Yet to
know the specialist language, you have to know that this is done regularly in
such a variety; you have to know why (its function in the specialist language);
and you have to know how and why doing this goes together with a host of
other related processes (for example, using a subject like “hornworm growth”
rather than “hornworms” or avoiding emotive words like “sure”). Any variety
of a language uses certain patterns of resources, and to know the language,
you have to be able to recognize and use these patterns (Halliday 1973, 1985a,
1985b; Halliday and Martin 1993; Halliday and Matthiessen 1999). This is
much like recognizing that the pattern of clothing “sun hat, swimsuit, and
thongs” means someone is going to the beach.
Earlier we stressed the close connections between meaning and experi-
ence. Yet our experiences of talk, dialogue, and social interaction with other
people are a large part of what teaches us how words and other signs apply
to reality. Let us consider for a moment how people learn the meaningful
functional features of their everyday language. Note, however, that we are not
talking about the acquisition of “core grammar” (the basic design features
that all languages share at least parameters for), an innate competence for
human beings (Chomsky 1986). We are talking about how people learn which
discourse and pragmatic functions words and syntactic structures can carry
out within the social groups of which they are members. Consider, in this
regard, the following quote from Michael Tomasello (1999):

. . . the perspectival nature of linguistic symbols, and the use of linguistic symbols
in discourse interaction in which different perspectives are explicitly contrasted
and shared, provide the raw material out of which the children of all cultures
construct the flexible and multi-perspectival – perhaps even dialogical – cognitive
representations that give human cognition much of its awesome and unique
power. (p. 163)

Let’s briefly explore what this means. From the point of view of the the-
ory Tomasello is developing, the words and grammar of a human language
exist to allow people to take and communicate alternative perspectives on
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

98 James Paul Gee

experience (see also Hanks 1996). That is, words and grammar exist to give
people alternative ways to view one and the same state of affairs. Language is
not about conveying neutral information; rather, it is about communicating
perspectives on experience and action in the world, often in contrast to alter-
native and competing perspectives: “We may then say that linguistic symbols
are social conventions for inducing others to construe, or take a perspective
on, some experiential situation” (Tomasello 1999, 118). This is not to say, by
the way, that some perspectives are not better or worse than others, only
that language allows its users to state and debate different perspectives or
interpretations of the world about them.
This is not surprising, because we have argued already that people give
meaning to language by running simulations of our previous experiences. We
see that language is already built to convey perspectives on experience, not
to offer neutral viewpoints detached from how people actually see things.
Human language is built to support human thinking, both of which are
perspectival.
Let’s give some examples of what it means to say that words and grammar
are not primarily about giving and getting information, but rather about
giving and getting different perspectives on experience. You open Microsoft’s
Web site: Are products you can download from the site without paying a price
for them “free,” or are they being “exchanged” for prior Microsoft purchases
(e.g., Windows)? Saying “the download was free because I already owned
Windows” is a different perspective on the same sort of experience than “the
download was paid for when I bought Windows.” If I use the grammatical
construction “Microsoft’s new operating system is loaded with bugs,” I take
a perspective in which Microsoft is less agentive and responsible than if I
use the grammatical construction, “Microsoft has loaded its new operating
system with bugs.”
These are all examples from daily life. However, such perspective taking is
equally important for the specialist varieties of language used in the content
areas in school – a type of language development that occurs after children’s
early socialization into their native vernacular dialects. Earlier we discussed
diSessa’s programming language for capturing some specific applications of
Galileo’s laws of motion. These laws, and diSessa’s specific symbolic instanti-
ation of them, take a perspective on the material world that is quite different
from the perspectives we tend to use everyday language for to take on that
same world. Furthermore, other symbolic forms take yet a different perspec-
tive on much the same phenomena (e.g., geometrical expression of the laws
of motion). Learners don’t really understand any of these symbolic expres-
sions unless they see what perspective they are designed to take on reality;
that is, how they imply the world is and what the symbolic forms allow us
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 99

to do to/with the world. The best way to see this is to participate in social
interactions and activities in which these symbolic forms are used in ways
that make clear what they mean and how they apply.
How do children learn how words and grammar line up to express partic-
ular perspectives on experience? Here, interactive, intersubjective dialogue
with more advanced peers and masters appears to be crucial. In such dia-
logue, children come to see, from time to time, that others have taken a
different perspective on what is being talked about than they have. At a cer-
tain developmental level, children have the capacity to distance themselves
from their own perspectives and (internally) simulate the perspectives the
other person is taking, thereby coming to see how words and grammar come
to express those perspectives (in contrast to the way in which different words
and grammatical constructions express competing perspectives).
Later, in other interactions or when thinking, the child can rerun such
simulations and imitate the perspective-taking the more advanced peer or
adult has done by using certain sorts of words and grammar. Through such
simulations and imitative learning, children learn to use the symbolic means
that other persons have used to share attention with them:

In imitatively learning a linguistic symbol from other persons in this way, I

internalize not only their communicative intention (their intention to get me to
share their attention) but also the specific perspective they have taken. (Tomasello
1999, 128)

Tomasello also points out (1999, 129–30) that children come to use objects in
the world as symbols at the same time (or with just a bit of a time lag) as they
come to use linguistic symbols as perspective-taking devices on the world.
Furthermore, they learn to use objects as symbols (to assign them different
meanings encoding specific perspectives in different contexts) in the same
way they learn to use linguistic symbols. In both cases, the child simulates
in his or her head and later imitates in his or her words and deeds the per-
spectives his or her interlocutor must be taking on a given situation by using
certain words and certain forms of grammar or by treating certain objects
in certain ways. Thus, meaning for words, grammar, and objects comes out
of intersubjective dialogue and interaction: “ . . . human symbols [are] inher-
ently social, intersubjective, and perspectival” (Tomasello 1999, 131).
The same dialogic, socially interactive process of language acquisition
that shapes children’s early understanding of the meaningful functions of
their everyday language applies to their learning later specialist varieties of
language that are crucial for school success. Learners need to participate in
social interactions and activities in which they can make good guesses about
what perspectives on reality the language and other symbol systems they see
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

100 James Paul Gee

in use are being used to take. They need to be able to simulate these in their
minds and try them out in interactions, hopefully in contexts that do not
punish them for initially unsuccessful or partially flawed attempts.
If students fail to know the languages of the content areas, no really deep
learning can occur, although they memorize and recite facts they don’t fully
understand and cannot themselves use in proactive ways. If children do
not start early on the acquisition of academic forms of language, they are
swamped by the later grades and high school and beyond, where language
demands in the content areas become intense and complex. Again, we face
a crucial opportunity-to-learn issue: Have all children in a given learning
environment had equal opportunity to learn the specialist forms of language
vital for thought and action in the domain they are seeking to learn?

culture
A sociocultural approach places a premium on learner’s experiences, social
participation, use of mediating devices (tools and technologies), and posi-
tion within various activity systems or communities of practice. The word
“culture” has taken on a wide variety of different meanings in different dis-
ciplines. Nonetheless, it is clear that, as part and parcel of our early social-
ization in life, we each learn ways of being in the world, of acting and inter-
acting, thinking and valuing, and using language, objects, and tools that
crucially shape our early sense of self. A situated/sociocultural perspective
amounts to an argument that students learn new academic “cultures” at
school (new ways of acting, interacting, valuing, and using language, objects,
and tools) and, as in the case of acquiring any new culture, the acquisition of
these new cultures interacts formidably with the learners’ initial cultures (in
Gee 1996, 2005, I use the term “Discourse” – with a capital “D” – instead of
“culture”).
So far we have discussed the kinds of specialized experiences, tools, forms
of participation, and varieties of oral and written language that are found
in school and elsewhere in public settings like academics, workplaces, and
institutions for which schools are meant to prepare people. Our early social-
ization in life gives us what we might call our “vernacular” culture; that is,
the ways of being, doing, acting, interacting, and using language, objects, and
tools that we associate with being an “everyday” (“nonspecialized”) person
belonging to specific social groups. Each of us has a culturally different way
of being an “everyday” person, and it is this identity that we bring to school
when we start the process of learning the specialized ways associated with
formal schooling and academic content areas.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 101

As we saw earlier, linguists have long made a distinction between a per-

son’s “vernacular” dialect (which is different for different geographical and
social groups) and the specialized varieties of language (“registers”) like those
used in the content areas in school and in academic disciplines (Labov 1972;
Milroy and Milroy 1985). We can broaden this distinction to consider not
only people’s vernacular dialects but also their vernacular cultural ways as
a whole (Barton 1994). Just as one’s vernacular dialect is the core basis on
which new varieties of language are learned, so, too, one’s vernacular culture
is the core basis on which new cultures at school are learned.
We argued earlier that we humans give meanings to things and plan for
action by building perspectival simulations in our minds based on the sorts
of experiences we have had. Yet each of us has had different experiences – and
learned different perspectives on those experiences – in our early encultura-
tion. These home- and community-based ways of building meaning interact
with and form the initial base for the child’s new experiences at school and
in academic content areas, the experiences with which the child will learn to
build school-based models of the world.
The same is true in the other areas we have discussed. Children come to
school with culturally specific, favored methods of participating in interac-
tions and activities and using language and mediating devices (objects, tools,
and technologies) of various sorts (Lee, this volume). These methods also
interact with the new, school-based methods the child must learn.
The specialized forms of language and interaction that the child finds at
school can resonate with and bridge to the child’s vernacular cultural ways
and thereby enhance learning. Alternately, they can lack such resonance or fail
to create such bridges, thereby raising the child’s affective filter and ensuring
that school-based input is not actually intake. A great deal of the literature in
the sociocultural area is devoted to this topic (e.g., Delpit 1995; Heath 1983;
Scollon and Scollon 1981; Taylor 1983; Taylor and Dorsey-Gaines 1987). It has
been shown, for instance, that the home-based practices of many middle-class
families involve actions, interactions, and ways with words that resonate with
practices that occur at school (e.g., Delpit 1995; Gee 1996; Heath 1983; Ochs
et al. 1992). For example, practices in which children engage in early pretend
readings of books or give scaffolded reports at dinnertime to their parents
about some activity they have done that day have been shown to facilitate
early success with early school literacy practices. Allowing children to develop
“islands of expertise” with regard to subjects like trains or dinosaurs, for
which they tend to give and hear school-related explanatory language, is
another practice that appears to facilitate early school success (Crowley and
Jacobs 2002).
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

102 James Paul Gee

In all of these cases and many others like them, children are not only prac-
ticing early versions of school-based practices but are doing so as part and
parcel of being socialized into their vernacular culture. These children come
to associate school and school-based ways with their home and community-
based identities, thanks to the initial overlap between home and school prac-
tices. This is a powerful form of affiliation.
There is also ample literature demonstrating that children from groups
that have tended to fare less well in school also engage in complex and sophis-
ticated language and interactional practices at home. For example, the com-
plex and often poetic verbal practices of many African American children
have been well documented (e.g., Delpit 1995; Gee 1996; Labov 1972, 1974;
Rickford and Rickford 2000; Smitherman 1977). However, too few schools
make use of early school-based practices that resonate with these vernacular
practices and build on them, thereby failing to build the initial strong sense
of affiliation with school that often occurs for other children.
The same issues that arise for children’s entry into school continue to
apply throughout the school years. However, young people bring to school
not only their vernacular language and culture but also their peer-based cul-
tures (Shuman 1986). Some of these themselves involve specialist varieties of
language and culture (“discourses”) connected not to school “content” but
to their own peer-based and community identities – for example, if they have
become adept at hip-hop or anime. The language and practices associated
with hip-hop or Yu-Gi-Oh (an anime card game, video game, and televi-
sion series) are quite complex – indeed, “specialist” and often “technical.”
However, they are not school based. Schools, therefore, face the issue of how
to bridge – and not denigrate – not just children’s home-based cultures but
their peer-based and “popular cultural” cultures as well.
Schools can make use of students’ cultural knowledge and practices and
link to their cultural senses of self or they can ignore – or worse, denigrate –
these and risk raising the learner’s affective filter. Work like that of Lee (1993,
1995, 1997, 2001, this volume) has amply demonstrated that such resonance
and links can be made and can make for real school success, even in high
school. Lee has built a curriculum that allows African American students
to use and research their own vernacular dialects (dialects well studied by
professional linguists) and vernacular verbal practices of using metaphor
and other tropes. The students also study specialized practices in domains
like rap, which have integral links to their vernacular culture. These studies,
which are academic in their own right, serve, too, as a mediating device for the
students’ later studies of standard school-based fare involving, for example,
literary critical studies of African American and other novelists.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 103

Lee has also demonstrated that classroom interaction, talk, and participa-
tion can be enacted in ways that resonate with some African American stu-
dents’ home- and community-based discourse practices (i.e., ways of making
sense through language in social interaction). At the same time, these class-
room practices involve deep learning of school-based content via the types of
situated and participatory learning we have discussed earlier in this chapter
and, together with Lee’s overall curriculum, eventually lead as well to the
acquisition of more specialized registers and school-based ways with words.
What Lee is doing is, of course, no more than what schools do often at an
unconscious level for more privileged children from other cultural groups –
those more often associated with so-called “mainstream” middle-class chil-
dren’s homes.
Surely two children have not had the same opportunity to learn if schooling
or a given assessment is built on resonances with one child’s vernacular culture
and not on the other’s. Worse yet, two children have not had the same OTL
if, however unconsciously, schooling or assessment ignores, dismisses, or
demeans the one child’s home- and community-based sense of self and ways
with words, deeds, and interactions.

conclusions: assessment
Current assessments often don’t mean what we often think they do. For
example, the well-known phenomenon of the “fourth-grade slump” – the
common situation in which children who have passed early reading tests can-
not “read to learn” by the time more complex language demands, connected
with academic content areas, are made in the fourth grade – shows that early
reading tests do not mean children are learning to read in any academically
useful way (Chall, Jacobs, and Baldwin 1990; see also the Research Round-Up
section of the Spring 2003 issue of American Educator).
If we think of learners in terms of developmental trajectories (Greeno
and Gresalfi, this volume) in the space of academic content learning and
the learning of the complex forms of academic language associated with
different content areas, teachers and policy makers alike need assessments
that tell them where learners are in their trajectories and whether they are on
course for successful progress in the future. Even certain forms of “failure”
may be indicative of progress (e.g., when young children start saying “goed”
instead of “went,” demonstrating that they are catching on to the existence of
an underlying rule system rather than just memorizing forms), and certain
forms of “success” may not really portend success, as is so well shown by the
fourth-grade slump.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

104 James Paul Gee

Assessing the development of learners rather than static stores of skills

and information that may not be meaningful in terms of the course of devel-
opment means that we must ensure that all learners are given the resources
and environments necessary for development in school. Americans believe
that schooling is a right for all children. Thus, it makes sense to argue that
children have certain rights with regard to learning and assessment; that is,
rights to conditions that if unmet mean assessment is meaningless or unfair
and full development impossible. This chapter, while overemphasizing cer-
tain sociocultural approaches to learning, has discussed a number of these
“rights,” each couched in terms of the notion of OTL.
In summary, we can state the following rights we would argue each child
has with regard to OTL. When these rights are not honored, assessments are
meaningless and unfair, unless their point is to tell us that the learners’ rights
were not honored.

1. Classrooms must offer learners not just the same “content” but also
equal affordances for action, participation, and learning.
2. Because comprehension requires the ability to simulate relevant expe-
riences in the mind, all learners must be offered the range of necessary
experiences with which they can build good and useful simulations for
understanding in the content areas (e.g., science, mathematics, social
studies, history).
3. Learning for humans is mediated by “smart tools”; that is, represen-
tations, technologies, and other people networked into knowledge
systems. Thus, learners must be offered equal access to such smart
tools.
4. Learning takes place within activity systems, systems that, in school,
should be a form of a community of practice. Thus, we must consider
more than the information to which the learner has been exposed.
All the other elements in the system need to count as well, including
access to the forms of participations and social interaction that make
one an agent and knower in the system.
5. Content learning in school requires learning new forms of language
and the identities, values, content, and characteristic activities con-
nected with these forms of language (e.g., the language of literary
criticism or of experimental biology). Every learner has the right for
these “new cultures” to be introduced in ways that respect and build on
the learner’s other cultures and indigenous knowledge, including his
or her home-based vernacular culture and peer-based and “popular
culture” cultures (“Discourses”).
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 105

There are, of course, other such rights connected to authentic OTL. Yet
caring about these rights means caring – in research, teaching, and assess-
ment – about the trajectories of learners as they develop within content areas
in school as part of communities of practice, engaged in mind, body, and
culture, and not just as repositories of skills, facts, and information.

References
Barsalou, L. W. 1999a. Language comprehension: Archival memory or preparation for
situated action. Discourse Processes 28: 61–80.
Barsalou, L. W. 1999b. Perceptual symbol systems. Behavioral and Brain Sciences 22:
57–660.
Barton, D. 1994. Literacy: An introduction to the ecology of written language. Oxford:
Blackwell.
Barton, D. and K. Tusting. 2005. Beyond communities of practice: Language power and
social context. Cambridge: Cambridge University Press.
Bransford, J., A. L. Brown, and R. R. Cocking. 2000. How people learn: Brain, mind,
experience, and school: Expanded edition. Washington, D.C.: National Academy Press.
Brown, A. L. 1994. The advancement of learning. Educational Researcher 23: 4–12.
Brown, A. L., A. Collins, and P. Dugid. 1989. Situated cognition and the culture of
learning. Educational Researcher 18: 32–42.
Center for Activity Theory and Developmental Work Research. 2003. The activity sys-
tem. http://www.edu.helsinki.fi/activity/pages/chatanddwr/activitysystem/.
Chall, J. S., V. Jacobs, and L. Baldwin. 1990. The reading crisis: Why poor children fall
behind. Cambridge: Harvard University Press.
Chomsky, N. 1986. Knowledge of language: Its nature, origin, and use. New York: Praeger.
Churchland, P. M. 1995. The engine of reason, the seat of the soul. Cambridge: MIT
Press.
Churchland, P. S. and T. J. Sejnowski. 1992. The computational brain. Cambridge:
Bradford/MIT Press.
Clancey, W. J. 1997. Situated cognition: On human knowledge and computer representa-
tions. Cambridge: Cambridge University Press.
Clark, A. 1989. Microcognition: Philosophy, cognitive science, and parallel distributed
processing. Cambridge: MIT Press.
Clark, A. 1997. Being there: Putting brain, body, and world together again. Cambridge:
MIT Press.
Clark, A. 2003. Natural-born cyborgs: Why minds and technologies are made to merge.
Oxford: Oxford University Press.
Cole, M. and Y. Engeström. 1993. A cultural-historical approach to distributed cogni-
tion. In Distributed cognition: Psychological and educational considerations, edited by
G. Salomon, 1–46. Cambridge: Cambridge University Press.
Corder, S. P. 1967. The significance of learners’ errors. International Review of Applied
Linguistics 5: 160–70.
Crowley, K. and M. Jacobs. 2002. Islands of expertise and the development of family
scientific literacy. In Learning conversations in museums, edited by G. Leinhardt,
K. Crowley, and K. Knutson, 333–56. Mahwah, N.J.: Lawrence Erlbaum.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

106 James Paul Gee

Delpit, L. 1995. Other people’s children: Cultural conflict in the classroom. New York: The
New Press.
diSessa, A. A. 2000. Changing minds: Computers, learning, and literacy. Cambridge: MIT
Press.
DuFour, R. and R. Eaker. 1998. Professional learning communities at work: Best practices
for enhancing student achievement. Bloomington, Ind.: Solution Tree.
Dulay, H., M. Burt and S. Krashen. 1982. Language two. Oxford: Oxford University Press.
Ellis, R. 1997. Second language acquisition. Oxford: Oxford University Press.
Engeström, Y. 1987. Learning by expanding. An activity theoretical approach to develop-
mental research. Helsinki: Orienta Konsultit.
Engeström, Y., R. Miettinen, R.-L. Punamäki, eds. 1999. Perspectives on activity theory.
Cambridge: Cambridge University Press.
Fink, E. and L. Resnick. 2001. Developing principals as instructional leaders. Phi Delta
Kappan 82: 598–606.
Fodor, J. A. 1975. The language of thought. Cambridge: Harvard University Press.
Fullan, M. and A. Hargreaves. 1991. What’s worth fighting for in your school? New York:
Teachers College Press.
Gass, S. M. 1997. Input, interaction, and the second language learner. Mahwah, N.J.:
Lawrence Erlbaum.
Gee, J. P. 1992. The social mind: Language, ideology, and social practice. New York: Bergin
and Garvey.
Gee, J. P. 1996. Social linguistics and literacies: Ideology in Discourses, 2nd ed. London:
Routledge/Taylor & Francis.
Gee, J. P. 2002. Literacies, identities, and discourses. In Developing advanced literacy
in first and second languages: Meaning with power, edited by M. Schleppegrell and
M. Cecilia Colombi, 159–75. Mahwah, N.J.: Lawrence Erlbaum.
Gee, J. P. 2003. What video games have to teach us about learning and literacy. New York:
Palgrave/Macmillan.
Gee, J. P. 2004. Situated language and learning: A critique of traditional schooling. London:
Routledge.
Gee, J. P. 2005. An introduction to discourse analysis: Theory and method, 2nd ed. London:
Routledge.
Gee, J. P. 2008. Social linguistics and literacies: Ideology in discourses, 3rd ed. London:
Routledge/Taylor and Francis.
Gee, J. P., G. Hull, and C. Lankshear. 1996. The new work order: Behind the language of
the new capitalism. Boulder: Westview.
Gibson, J. J. 1977. The theory of affordances. In Perceiving, acting, and knowing: Toward
an ecological psychology, edited by R. Shaw and J. Bransford, 67–82. Hillsdale, N.J.:
Lawrence Erlbaum.
Gibson, J. J. 1979. The ecological approach to visual perception. Boston: Houghton Mifflin.
Glenberg, A. M. 1997. What is memory for? Behavioral and Brain Sciences 20: 1–55.
Glenberg, A. M. and D. A. Robertson. 1999. Indexical understanding of instructions.
Discourse Processes 28: 1–26.
Greeno, J. 1997. On claims that answer the wrong questions. Educational Researcher 26:
5–17.
Greeno, J. 1998. The situativity of knowing, learning, and research. American Psychologist
53: 5–26.
P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

A Sociocultural Perspective on Opportunity to Learn 107

Hagel, J. and J. S. Brown. 2005. The only sustainable edge: Why business strategy depends
on productive friction and dynamic specialization. Boston: Harvard Business School
Press.
Halliday, M. A. K. 1973. Language in a social perspective: Explorations in the functions of
language. London: Edward Arnold.
Halliday, M. A. K. 1985a. Spoken and written language. Victoria: Deakin University.
Halliday, M. A. K. 1985b. An introduction to functional grammar. London/Baltimore:
Edward Arnold.
Halliday, M. A. K. and J. R. Martin. 1993. Writing science: Literacy and discursive power.
Pittsburgh: University of Pittsburgh Press.
Halliday, M. A. K. and C. M. I. M. Matthiessen. 1999. Construing experience through
meaning: A language-based approach to cognition. New York: Continuum.
Hanks, W. F. 1996. Language and communicative practices. Boulder: Westview Press.
Heath, S. B. 1983. Ways with words: Language, life, and work in communities and class-
rooms. Cambridge: Cambridge University Press.
Holland, D. and N. Quinn, eds. 1987. Cultural models in language and thought.
Cambridge: Cambridge University Press.
Hutchins, E. 1995. Cognition in the wild. Cambridge: MIT Press.
Krashen, S. D. 1983. Principles and practices in second language acquisition. Oxford:
Pergamon Press.
Labov, W. 1972. Language in the inner city: Studies in Black English vernacular.
Philadelphia: University of Pennsylvania Press.
Labov, W. 1974. The art of sounding and signifying. In Language in its social setting, edited
by W. Gage, 84–116. Washington, D.C.: Anthropological Society of Washington.
Latour, B. 1999. Pandora’s hope: Essays on the reality of science studies. Cambridge:
Harvard University Press.
Lave, J. 1996. Teaching, as learning, in practice, Mind, Culture, and Activity. 3: 149–64.
Lave, J. and E. Wenger. 1991. Situated learning: Legitimate peripheral participation. New
York: Cambridge University Press.
Lee, C. D. 1993. Signifying as a scaffold for literary interpretation: The pedagogical implica-
tions of an African American discourse genre. Urbana, Ill.: National Council of Teachers
of English.
Lee, C. D. 1995. A culturally based cognitive apprenticeship: Teaching African American
high school students skills in literary interpretation. Reading Research Quarterly 30:
608–31.
Lee, C. D. 1997. Bridging home and school literacies: Models for culturally responsive
teaching, a case for African American English. In A handbook for literacy educators:
Research on teaching the communicative and visual arts, edited by J. Flood, S. B. Heath,
and D. Lapp, 334–45. New York: Macmillan.
Lee, C. D. 2001. Is October Brown Chinese? A cultural modeling activity system for
underachieving students. American Educational Research Journal 38: 97–141.
Milroy, J. and L. Milroy. 1985. Authority in language: Investigating language prescription
and standardisation. London: Routledge.
Newell, A. and H. A. Simon. 1972. Human problem solving. New York: Prentice-Hall.
Norman, D. A. 1988. The design of everyday things. New York: Basic Books.
Ochs, E., R. Smith, D. Rudolph, and C. Taylor. 1992. Storytelling as a theory-building
activity. Discourse Processes 15: 37–72.
 P1: KXF
9780521880459c04 CUUS083/Moss 978 0 521 88045 9 January 24, 2008 16:24

108 James Paul Gee

Pylyshyn, Z. 1984. Computation and cognition. Cambridge: MIT Press.

Research round-up. Spring 2003. American Educator.
Rickford, J. R. and R. J. Rickford. 2000. Spoken soul: The story of Black English. New
York: John Wiley.
Schleppegrell, M. 2004. Language of schooling: A functional linguistics perspective.
Mahwah, N.J.: Lawrence Erlbaum.
Schleppegrell, M. and M. C. Colombi, eds. 2002. Developing advanced literacy in first
and second languages: Meaning with power, 159–75. Mahwah, N.J.: Lawrence Erlbaum.
Scollon, R. and S. W. Scollon. 1981. Narrative, literacy, and face in interethnic communi-
cation. Norwood, N.J.: Ablex.
Shuman, A. 1986. Storytelling rights: The uses of oral and written texts by urban adolescents.
Cambridge: Cambridge University Press.
Smitherman, G. 1977. Talkin and testifin: The language of Black America. Boston:
Houghton Mifflin.
Strauss, C. and N. Quinn. 1997. A cognitive theory of cultural meaning. Cambridge:
Cambridge University Press.
Taylor, D. 1983. Family literacy: Young children learning to read and write. Portsmouth,
N.H.: Heinemann.
Taylor, D. and C. Dorsey-Gaines. 1987. Growing up literate: Learning from inner city
families. Portsmouth, N.H.: Heinemann.
Tomasello, M. 1999. The cultural origins of human cognition. Cambridge: Harvard Uni-
versity Press.
Vygotsky, L. S. 1978. Mind in society: The development of higher psychological processes.
Cambridge: Harvard University Press.
Wenger, E. 1998. Communities of practice: Learning, meaning, and identity. Cambridge:
Cambridge University Press.
Wenger, E., R. McDermott, and W. M. Snyder. 2002. Cultivating communities of practice.
Cambridge: Harvard Business School Press.
Wertsch, J. V. 1998. Mind as action. Oxford: Oxford University Press.

View publication stats