DEVELOPMENTAL REVIEW 2, 371-384 (1982)

Exogenous, Endogenous, and

Dialectical Constructivism
DAVIII MOSHMAN

Three constructivist paradigms are distinguished. Exogenous constructivism

(rooted in a mechanistic metaphor) emphasizes the reconstruction of structures
preformed in the environment. Endogenous constructivism (rooted in an organis-
mic metaphor) emphasizes the coordination of previous organismic structures.
Dialectical constructivism (rooted in a contextualistic metaphor) emphasizes the
construction of new structures out of organism/environment interaction. It is
suggested that more general metatheories integrating exogenous, endogenous, and
dialectical aspects of the construction of knowledge can and should be formulated.
Such formulations would not attempt an impossible synthesis of the root
metaphors, but rather integrate them in a coherent metatheory by specifying the
boundary conditions in which each root metaphor best applies. An example of
such a metatheory, based primarily on Piagetian ideas, is presented.

Consider the following quotation:

Intellectual development may be conceived as the building of increasingly complex
and interacting structures. . The structures of capability so developed can inter-
act with each other in patterns of great complexity, and thus generate an ever-
increasing intellectual competence. Each structure may also build upon itself
through self-initiated thinking activity.

This may be considered a constructivist viewpoint in its emphasis on the

active construction of new knowledge and intellectual competencies. The
quote, however, comes not, as one might initially guess, from Piaget, but
rather from a 1968 article by the learning theorist Robert Gag@ entitled
“Contributions of Learning to Human Development.” Consider it again
but with some missing sections tilled in:
Intellectual development may be conceived as the building of increasingly complex
and interacting structures of learned capabilities. The entities which are learned
build upon each other in a cumulative fashion, and transfer of learning occurs
among them. The structures of capability so developed can interact with each other
in patterns of great complexity, and thus generate an ever-increasing intellectual
competence. Each structure may also build upon itself through self-initiated think-
ing activity. There is no magic key to this structure-it is simply developed piece by
piece. The magic is in learning and memory and transfer (Gagne, 1968, p. 190).

Presented in an earlier version at the meeting of the Jean Piaget Society, Philadelphia,
May 1980. Requests for reprints should be sent to David Moshman, Educational Psychology
and Social Foundations, University of Nebraska-Lincoln, Lincoln, Nebr. 68588-0440.
371
0273-2297/82/040371-14$02.00/O
Cupynght (0 I982 by Academic Press, Inc.
All rights of reproduction in any form reserved.
372 DAVID MOSHMAN

Though there is no longer any temptation to attribute the statement to

Piaget, its constructivist flavor remains obvious: It posits not a passive
copying or direct internalization but rather an active construction of cog-
nitive novelties that vigorously interact both with each other and with the
environment. Equally constructivistic statements from other learning
theorists are not difficult to find. For example:
Let us look first at comprehension as the construction of knowledge. This first
theme reflects a change from viewing comprehension as the storage of knowledge
to viewing it as the construction of one’s knowledge (Glaser, 1979, p. 7).

Comprehension involves the construction of meanings and inferences that may

differ from the original message (Bransford, 1979, p. 158).

In social learning theory, people play an active role in creating information-

generating experiences as well as in processing and transforming informative
stimuli (Bandura, 1978, p. 356).

Social learning theory as proposed by Bandura is constructivistic in that previ-

ously acquired cognitive rules are considered during responding in reciprocal con-
junction with environmental sources of information (Zimmerman, 1981, p. 41).

The result of interaction that takes place between the new material to be learned
and the existing cognitive structure is an assimilation of old and new meanings to
form a more highly differentiated cognitive structure (Ausubel, Novak, & Hane-
Sian, 1978, p. 67-68, italics in original).

Even if we recognize the ubiquity of constructivism, however, it would

be a serious error to conclude that modern cognitive learning theories
such as those of Gagne, Bandura, Ausubel, and contemporary
information-processing theorists are fundamentally Piagetian or, on the
other hand, that Piagetian theory is simply a particular sort of learning
theory (cf. Reese & Overton, 1970; Overton & Reese, 1981). Instead, it
seems that we must be prepared to distinguish radically different ap-
proaches to understanding the construction of knowledge or, to put it
another way, a variety of constructivist paradigms. The aim of this paper
is to distinguish three such constructivist paradigms, which shall be dis-
cussed below under the labels (a) exogenous constructivism, (b) endoge-
nous constructivism, and (c) dialectical constructivism. It should be noted
in advance that the three paradigms to be discussed are idealizations and
that, for purposes of this paper, I will be ignoring significant differences of
approach within each.
EXOGENOUS CONSTRUCTIVISM
Exogenous constructivism, rooted in the mechanistic metaphor of
knower as machine (Pepper, 1942), is reflected in such cognitive concep-
tions of learning as contemporary social learning theory (Bandura, 1977)
 CONSTRUCTIVIST PARADIGMS 373

and information-processing theories (see Bransford, 1979). Though such

theories are far more constructivistic than their behaviorist predecessors,
they, to varying extents, maintain the empiricist assumption that knowl-
edge is derived from one’s environment and, in that sense, can be said to
be learned. The construction of knowledge is thus fundamentally a recon-
struction of structures (empirical relationships, presented information,
observed behavior patterns, etc.) preformed in the external reality.
Though the abstraction of knowledge from that environment is assumed
to involve an active organism, empirical (environmental, exogenous, ex-
ternal) guidance of this constructive activity (by physical reality, pre-
sented information, social models, etc.) remains the principal factor in
directing the course of learning. Thus, despite the necessity of the or-
ganism’s activity, and despite the fact that the effective environment is
partially a function of that activity, the structure of the environment upon
which the organism acts exerts a crucial determining influence upon the
structure of the knowledge constructed. In Piagetian terminology, learn-
ing is seen as fundamentally an accommodation of the organism’s prior
structures to those imposed by its current environment. Though it is in
some sense the organism (or its prior knowledge) that does the accom-
modating, it is the environment that directs the accommodation by pro-
viding the structures to which the organism must accommodate. Knowl-
edge, within this paradigm, is conceptualized mechanistically as consist-
ing of intricate, flexible, and open frames, scripts, networks, schemas,
hierarchies, or production systems, composed of discrete content-laden
elements, functioning in real time, and possessing no systematic prop-
erties over and above those of the elements of which they are composed.
The development of such structures tends toward increasing adaptation to
particulars of the environment and thus follows a path that is in practice
unpredictable, though in principle a strict function of the environmental
history. Structures of knowledge are adequate or “true” to the extent that
they accurately copy (or at least adapt or accommodate to) the external
structures that they ideally represent.
ENDOGENOUS CONSTRUCTIVISM
Endogenous constructivism, rooted in the organismic metaphor of
knower as biological organism (Pepper, 1942), is strikingly illustrated
(though perhaps caricatured) in Piaget’s (e.g., 1971a) familiar anecdote of
his mathematician friend who recalled constructing the concept of number
at the age of about five as a result of solitary manipulation of pebbles in his
garden. The point Piaget emphasizes is that the concept of conservation of
number was not learned either from the social environment (no one else
was present) or from the physical properties of the pebbles (they could
just as well have been marbles, sticks, etc.) but rather was internally
374 DAVID MOSHMAN

(endogenously) constructed through the boy’s active intercoordination of

his cognitive actions of counting, ordering, etc. This is development
rather than learning, not in the sense of an emergence or maturation of
preformed structures of knowledge (as in a simple nativism) but in the
sense of an internal construction of new knowledge from old. Such en-
dogenous construction differs sharply from learning (at least in the em-
piricist sense) in that there is no empirical abstraction of information from
the environment but rather a reflective abstraction of new structures via
intercoordination of, and/or metacognitive reflection on, current struc-
tures (Bickhard, 1978; Piaget, 1977a, b). The emphasis is thus not on
accommodation of present structures to the environment but rather on the
reciprocal assimilation of current structures. The environment may pro-
vide occasion for the functioning of cognitive structures and may induce
disequilibrium, but provides little or no informational input relevant to the
new structures to be constructed. The locus of activity in the construction
of new knowledge is thus seen as the organism rather than its environ-
ment. The increasingly organized internal structures of knowledge are
ontologically at a more abstract level than the real-time mental processes
emphasized in exogenous constructivism and are more divorced from
specific content. Moreover, in keeping with the organismic root
metaphor, endogenously constructed structures meet stringent criteria of
wholeness, transformation, and self-regulation that endow them with
properties (such as logical necessity) that are properties of the structure,
not of any part in isolation (Piaget, 1970). The whole thus determines and
explains the functioning of the parts more than vice versa. Since each
structure is constructed endogenously from earlier structures, rather than
from information provided by the environment, the structures follow one
another in predictable, invariant sequences, though without being caus-
ally linked in the sense of a direct antecedent-consequent relation. Ade-
quacy or “truth” of such structures is less a matter of match or accom-
modation to given realities than of internal coherence. Endogenously con-
structed structures do not, even in principle, represent anything in the
environment; their adaptation to that environment is a by-product of their
intrinsic organization.
DIALECTICAL CONSTRUCTIVISM
Dialectical theories, rooted in the contextualist metaphor of cognitive
change as a concrete historical process (Pepper, 1942), have been in-
creasingly influential in American psychology, and particularly in Ameri-
can developmental psychology, during the past decade. Dialectical con-
siderations play an important role, for example, in Riegel’s (1979) con-
ception of the complex and changing reciprocity in the relations between
the developing individual and the simultaneously changing world. Dialec-
 CONSTRUCTIVIST PARADIGMS 375

tics is also important in Vygotsky’s (1935) conception of the zone of

proximal development, in which learning is constrained by development
but simultaneously serves as its cutting edge, actively directing the course
that development takes. Neither exogenous learning nor endogenous de-
velopment is, in this view, predominant over the other: The two exist in a
relation of reciprocal constraint and facilitation.
For the dialectician, the source of all knowledge lies in the continuing
interactions between organism and environment, neither of which can
simply impose itself on the other. New knowledge is a constructed syn-
thesis which resolves the inevitable contradictions arising during the
course of such interactions. In Piagetian terms, such a synthesis may be
seen as an equilibration of assimilatory and accommodatory poles in the
dynamic interaction of the inseparable knower and known. Most current
dialecticians, however, in keeping with their contextualist root metaphor,
view the resulting knowledge not as an equilibrated, abstract, content-free
structure but as a concrete item in a population of interrelated concepts,
always context-laden and highly subject to further idiosyncratic con-
tradictions (e.g., Riegel, 1979). For the most part, then, development does
not follow a universal, predictable sequence but rather is a partially un-
predictable outcome of interacting changes at several levels-biological,
psychological, sociological, and physical (e.g., Lerner, Skinner, & Sorell,
1980; Riegel, 1979). New constructions are never true in any absolute
sense but are more adequate than their predecessors in that they synthe-
size a previous thesis and antithesis in such a way as to incorporate and
transcend both.
TOWARD A CONSTRUCTMST INTEGRATION
A summary of the three constructivist paradigms is provided in Table 1.
Though the differences suggested by the present analysis are real and
important, they do not necessarily rule out the possibility of at least a
partial integration. In fact, some such integration may be necessary for
further progress in our understanding of cognition and its development.
Conceivably, such an integration might come from someone approaching
the field from a perspective outside any of the current constructivist
paradigms delineated above. Alternatively, it might come from someone
already trained in and partially committed to a current paradigm who,
rather than rejecting the alternatives and attempting to show the superi-
ority of his or her preferred paradigm, instead attempts to expand that
paradigm to encompass the legitimate concerns and real accomplishments
of those other perspectives. Thus, for example, much dialectical theoriz-
ing is quite explicit about its intent to encompass the exogenous and
endogenous perspectives (e.g., Riegel, 1979). There have also been nota-
ble efforts to expand exogenous constructivism in the direction of endog-
376 DAVID MOSHMAN

TABLE 1
THREE CONSTRUCTIVIST PARADIGMS

Exogenous Endogenous Dialectical

constructivism constructivism constructivism
Root metaphor Mechanism Organicism Contextualism
Current Social learning Piaget (as Vygotsky
examples theory usually Riegel
Information- interpreted)
processing
theories
Major Learning Development Dynamic
emphasis interactionism
Source of External Internal Interaction
knowledge structures coordinations (subjective
(environment) (previous experience)
knowledge)
Mechanism of Empirical Reflective Dialectical
construction abstraction abstraction synthesis
Locus of Environment Organism Strong
activity (accommodation) (reciprocal interaction
assimilation) (equilibration)
Characterization Elementaristic Holistic Populations of
of knowledge structures structures concrete
(no systemic (systemic concepts
properties) properties)
Predictability Function of Invariant Unpredictable
of change environment sequences syntheses
Criterion of Adaptation to Internal Continuing
truth reality organization transcendence
of contradiction

enous and dialectical considerations. This can be seen, for example, in the
movement of social learning theory toward dialectical/contextualist ideas
of an active agent in reciprocal interaction with an environment partially
of its own making (e.g., Bandura, 1977, 1978; Zimmerman, 1981) and in a
similar trend toward contextualism in recent information-processing for-
mulations (e.g., Jenkins, 1974).
My own background and predilections lead me to take Piaget’s theory
(see, especially, Furth, 1981; Moshman, 1979; Piaget, 1971b, 1977a,b; von
Glasersfeld, 1979), which is generally viewed as primarily in the endoge-
nous tradition, as the starting point for a potential integration. The at-
tempted integration is, however, only a partial one, for a very important
 CONSTRUCTIVIST PARADIGMS 377

reason. As noted earlier, the exogenous, endogenous, and dialectical

paradigms differ not only in their respective emphases on an external,
internal, or interactive source of knowledge but, deriving directly from
this, in the root metaphor seen as most helpful in understanding the nature
of that knowledge. The following integration of the three paradigms
should not be construed as an attempt to synthesize the root metaphors of
mechanism, organicism, and contextualism, a task which Reese & Over-
ton (1970), among others, convincingly argue to be impossible. Rather, it
is proposed (a) that exogenous learning and endogenous development are
usefully viewed as special cases of a more general dialecticism which are
approximated in certain definable circumstances (i.e., when external or
internal factors, respectively, predominate), and (b) that whether exoge-
nous, endogenous, or dialectical considerations predominate in turn de-
termines which root metaphor is most useful. In this sense, the view
presented below is an intrgration of the three root metaphors but not a
synthesis. It is an integration in that all three play a role in a single
metatheoretical view, but it is not a synthesis in that they remain discrete.
In other words, the present formulation sets the boundary conditions that
determine when each root metaphor best applies, thus integrating them in
a single framework but without attempting an impossible synthesis.
A PIAGETIAN INTEGRATION
Though Piaget has always stressed the endogenous aspects of con-
structivism over the exogenous, his theory appears potentially able to
incorporate both aspects and is, moreover, strongly dialectical (e.g.,
Boden, 1980; Youniss, 1978), though without the antipathy to postulation
of abstract, formal structures shared by most dialecticians. There has,
however, been considerable difficulty in reconciling the endogenous (e.g..
reflective abstraction) and dialectical (e.g., interactionism) aspects of
Piaget’s work, a situation which has led to considerable confusion among
his interpreters and followers. Kuhn (1978), for example, describes the
ambivalence in Kohlbergian moral development research between an en-
dogenous equilibration emphasis and an exogenous emphasis on imitation
of (or abstraction from) models one stage above the developing individual.
She suggests that the problem is due to the fact that, despite the wide-
spread view that the social environment must somehow do more than just
yield disequilibrium and thus set off independent (endogenous) equilibra-
tion processes, it is difficult to account for the more specific informational
influence of particular social environments without returning to some sort
of (exogenous) social learning theory. An interesting recent dialogue on
predeterministic (endogenous) and probabilistic (dialectical) interpreta-
tions of epigenesis (Kitchener, 1978, 1980; Lerner, 1978, 1980) revolves
around similar issues. Part of the problem may lie in the continuing diffi-
378 DAVID MOSHMAN

culty of integrating structural (stage) and functional (process and transi-

tion) aspects of development, a central and longstanding Piagetian con-
cern (cf. Broughton, 1981).
It has recently been suggested by a number of theorists (e.g., Youniss,
1978) that a greater emphasis on the dialectical aspects of Piaget’s theory
would help resolve this continuing dilemma. The following outline is in-
tended to briefly suggest that, with all due regard for the very real dangers
of uncritical eclecticism (Reese & Overton, 1970), there exists in Piaget’s
dialecticism the potential for a coherent and powerful integration of the
three constructivist paradigms differentiated above.
The dialectical aspects of Piaget’s theory stress that knowledge derives
from one’s subjective experience, which is a joint function of one’s previ-
ous knowledge (emphasized by endogenous constructivism) and one’s
objective environment (emphasized by exogenous constructivism). The
waning of egocentrism during the course of development consists pre-
cisely of the progressive differentiation by the knowing subject of its own
contribution to experience from that of the known object. To the extent
that the subject’s contribution (assimilation) is predominant in a given
epistemic transaction, experience is logico-mathematical and new knowl-
edge is constructed by the relatively endogenous (developmental) process
of reflective abstraction. The result would be an invariant sequence of
logico-mathematical structures or stages (cf. Moshman & Timmons,
1982). The most enlightening root metaphor in this case would be organi-
cism. Alternatively, to the extent that the object’s contribution (accom-
modation of subject to environment) is predominant, the experience is
empirical and the new knowledge is constructed by the relatively exoge-
nous (learning) processes of empirical abstraction. Mechanism would now
be a more useful root metaphor. Though Piaget has been much less in-
terested in exogenous processes and has not studied them in anything like
the detail that information-processing and social learning theories have,
his theory clearly requires the existence of such processes (e.g., Piaget,
1977a, b) and thus, in this context, can profit from a mechanistic perspec-
tive.
To take an example, consider a child engaged in classifying members of
the animal kingdom. To the extent that her own classificatory activities
are predominant, her experience is primarily logico-mathematical and will
lead, through successive reflective abstractions over the course of many
years, from (sensorimotor) actions of grasping, grouping, separating, etc.,
to (preoperational) actions of mentally arranging, dividing, combining,
counting, etc., to highly structured and thus reversible (concrete opera-
tional) mental actions of class inclusion, class multiplication, etc., and
finally to (formal operational) classifications of classifications and for-
malizations that transcend empirical content. Such endogenous construc-
 CONSTRUCTIVIST PARADIGMS 379

tions have traditionally been explained in organismic terms and it may be

that the organismic root metaphor is most useful here. To the extent that
the empirical characteristics of the animals classified are predominant,
experience is primarily empirical and will lead, through successive em-
pirical abstractions, also over the course of many years, to increasingly
sophisticated content knowledge about the animal kingdom itself. In cases
of primarily exogenous construction such as this, we may continue to find
the root metaphor of mechanism particularly apt.
Given the profound interpenetration of subject and object, however-
that is, given that neither assimilation nor accommodation is ever
pure-the distinction between empirical and reflective abstraction is not a
sharp dichotomy but rather highlights the endpoints of a continuum. Clas-
sification actions, for example, are never entirely independent of the ani-
mals being classified (assimilation requires accommodation) nor can those
animals be classified without this involving some (mental if not physical)
action upon them (accommodation requires assimilation). Thus, equili-
bration at all levels of development invariably involves aspects of both
exogenous learning and endogenous development, though one or the
other may predominate at a particular time. To the extent that learning,
development, or neither of these is predominant over the other, the
equilibration will be primarily exogenous, endogenous, or dialectical (re-
spectively) and the most useful root metaphor will be mechanism, organi-
cism, or contextualism. Piagetian theory thus includes a dialectical con-
structivism which, when allotted a more central role, has the potential to
integrate both the endogenous and exogenous aspects of construction by
stressing the activity of both organism and environment in the construc-
tion of higher-order syntheses, that is, in the equilibration of assimilatory
and accommodatory processes.
To put the matter in Piaget’s own words, cognitive development un-
avoidably involves:
the dialectical idea of a constant becoming. The history of science as much as the
study of individual development shows that this interaction. while remaining indis-
sociable, passes from an undifferentiated phase to one of coordination. Starting
from a state of centration on a self uncognizant of itself and in which the subjective
and objective are inextricably intermingled, the progressive decentration of the
subject leads to a twofold movement, of externalization, tending to physical objec-
tivity, and internalization tending to logico-mathematical coherence (Piaget, 197lc,
p. 11.5).

Thus for both sciences and individuals knowledge moves in both

exogeneous and endogenous directions. What needs to be added is that
the exogenous and endogenous tendencies are not insulated from each
other: Each of these directions of cognitive construction is possible only
to the extent that it exists in dialectical relationship with the other.
380 DAVID MOSHMAN

The nature and interrelations of the exogenous, endogenous, and

dialectical aspects of cognitive construction are highlighted and clarified
in many of the recent developments in Piagetian theory. One example of
this is Piaget’s (1977~) elaboration of the concept of phenocopy, involving
the endogenous reconstruction of exogenous knowledge. This concept is
closely related to the Vygotskian notion of a zone of proximal develop-
ment, discussed earlier. It includes the idea, traditional in endogenous
constructivism, that development considerably constrains learning. It in-
tegrates this view, however, with the opposing view, traditional in exoge-
nous constructivism, that learning constrains development by determin-
ing the path in which it proceeds. The result is a dialectical view of
endogenous development and exogenous learning each playing an impor-
tant role but existing not in isolation but in continuing dialectical interac-
tion.
Another recent development in Piagetian theory that extends the pres-
ent analysis is his extensive work on the nature of contradiction (Piaget,
1980). It becomes clear that the fundamental contradictions, for Piaget,
are not of a formal nature, that is, errors in some formal system and thus
eliminable mistakes. Rather they are conflicts in the dialectical sense of an
emergent antithesis which can only be transcended by a new and higher
synthesis.
Finally, Piaget’s (1977a) new model of equilibration is more explicit
than ever before in differentiating and interrelating exogenous and endog-
enous aspects of cognitive change through its careful attention to the
objective and subjective poles of various sorts of interaction and to the
role of observables and coordinations in each of these types of interac-
tion. Excellent presentations and discussions of the revised equilibration
theory are available elsewhere (Furth, 1981; Gallagher & Reid, 1981;
Moessinger, 1978).
AN EXAMPLE: MORAL DEVELOPMENT
It may be useful to recapitulate and further explicate some of the central
points made in this paper in the context of a specific example. Consider
the domain of moral development. An endogenous constructivist, focus-
ing on internal variables and the individual construction of knowledge,
would be expected to emphasize the construction of increasingly sophisti-
cated modes of reasoning about moral issues (e.g., Kohlbergian stages)
through internal (endogenous) coordination of and reflection on earlier
stages. Such theorists have generally found the organismic root metaphor
(including assumptions of wholeness, self-regulation, and so forth) most
illuminating. An exogenous constructivist, on the other hand, focusing on
external variables and the social construction of knowledge, would in-
stead emphasize the internalization of moral rules and norms through
 CONSTRUCTIVIST PARADIGMS 381

observation of parents, other adults, and/or peers. The mechanistic root

metaphor (including assumptions of reductionism, predictability, and so
on) has generally been favored by such theorists as most helpful.
Unfortunately, the organismic worldview of the endogenous construc-
tivist not only helps interpret the developmental phenomena that are of
central interest to such a theorist but tends to ignore or trivialize the
phenomena of moral learning that most interest the exogenous theorist.
And the reverse is of course true for the latter individual. Moreover, at
this point a dialectical constructivist might enter the theoretical arena and,
noting the partial blindness of both the above approaches, argue instead
that moral development is a continuing dialectical interaction between a
changing individual and a changing social environment, and thus is best
understood from the perspective of the contextualist worldview.
The present approach suggests that, although the dialectical approach is
in a sense the most fundamental, it cannot simply replace the other two.
Rather, it provides a general perspective of which exogenous and endoge-
nous constructivism are special cases. Thus, to the extent that we focus
on the rational coordination of perspectives and the resulting under-
standing of the logic of moral reciprocity within an individual constructing
a consistent moral framework, endogenous considerations will be pre-
dominant and the organismic worldview is likely to be most useful. To the
extent that we focus on the internalization and generalization of moral
rules and norms from external authorities, our focus is exogenous and we
may find a mechanistic root metaphor of greater utility. And, to the extent
that we look at moral development in its most general sense, focusing on
continuing interaction between internal and external factors, a dialectical
perspective rooted in the contextualist worldview may be unavoidable.
(For detailed presentation of a similar approach in the domain of deduc-
tive reasoning, see Moshman & Timmons, 1982).
Although this approach somewhat integrates the exogenous, endoge-
nous, and dialectical orientations by showing how they interrelate, it does
not even attempt to synthesize the underlying mechanistic, organismic,
and contextualist worldviews. Thus, although it is an integrative view
with respect to sources of knowledge (external, internal, or interactive)
and relevant variables, it remains eclectic, perhaps confusingly so, at
the deeper level of root metaphors. If some of the individual’s knowledge
is constructed mechanistically, some organismically, and some contextu-
ally, how do these very different sorts of knowledge communicate within
the individual?
The best answer, I think, is that the individual does not necessarily
huvc these radically different types of knowledge but rather that we have
chosen to use these several metaphors in order to understand the indi-
vidual’s construction of knowledge. It would be much nicer, of course, if
382 DAVID MOSHMAN

we could understand the individual as a whole from the perspective of a

single root metaphor. That being presently impossible, however, I think
the sort of systematic eclecticism being advocated in this paper is our best
bet. The present approach is eclectic in acknowledging the usefulness of
several irreconcilable worldviews, but systematic in that, in contrast to
those who argue that each scientist must pick a single worldview and stick
to it, it argues that the different worldviews may be partially coordinated
by recognizing the conditions in which each best applies. Thus, in the
present example of moral development, we can distinguish cases of
learning from moral authorities (exogenous construction), internal reflec-
tion and coordination (endogenous construction), and reciprocal interac-
tion with peers (dialectical construction) and analyze these, respectively,
from mechanistic, organismic, and contextualist perspectives.
CONCLUSION
I have argued that, although most current views of cognitive learning
and development are justifiably labeled constructivist, there are three
very different senses of construction-exogenous, endogenous, and
dialectical-which need to be distinguished. Unfortunately, since these
three approaches are rooted in three irreconcilable worldviews-
mechanistic, organismic, and contextual, respectively-a satisfactory
synthesis of the three does not seem possible. As a result, most research-
ers have, consciously or not, stuck to a single perspective and, in the
process, ignored, misunderstood, and/or rejected the alternatives. I have
suggested in this paper that, even if a fully satisfactory synthesis is impos-
sible, we may nevertheless be able to achieve a useful-though only
partial-integration which shows how the three sorts of cognitive con-
struction are interrelated and, thus, indicates when each of their underly-
ing root metaphors may be most useful.
Although I have used Piagetian theory as a starting point for my own
proposed integration, I am not suggesting that only Piaget’s theory can
lead us to a useful cross-paradigmatic integration, or even that it provides
the best such integration. My primary intention, rather, has been to illus-
trate the broader point that such integrations are possible and important.
Our ultimate aim should not be to support any of the three paradigms
presented, nor to dispute any of them, but rather to integrate work from a
variety of research traditions into a coherent view that incorporates exog-
enous, endogenous, and dialectical aspects of the construction of knowl-
edge. Such integrations promise to enlighten us with respect to the do-
mains of optimal applicability of the mechanistic, organismic, and con-
textualist root metaphors that guide our theorizing at a deeper level. It is
as a catalyst to such attempts at integration that this paper is offered.
 CONSTRUCTIVIST PARADIGMS 383

REFERENCES
Ausubel, D. P., Novak, J. D., & Hanesian, H. Edrrcational psychology: A cogniti~~~ l,icJ\c,
(2nd. Ed.). New York: Holt, Rinehart, & Winston, 1978.
Bandura, A. Sociul leurning theory. Englewood Cliffs, N.J.: Prentice-Hall, 1977.
Bandura, A. The self system in reciprocal determinism. American Ps~cho/ogi.st. 1978, 33,
3444358.
Bickhard, M. H. The nature of developmental stages. Hrrmcrr~ Dewlopmc~t~t. 1978, 21,
217-233.
Boden, M. A. ./ecln Piuget. New York: Penguin, 1980.
Bransford, J. D. Hlrmnrr c~ognirion: Lc~arnirrg, und~~rsranding ~/id rrmrmhrrin::. Belmont.
Calif.: Wadsworth, 1979.
Broughton, J. M. Piaget’s structural developmental psychology. III. Function and the
problem of knowledge. Humun Dc~rrlopm~~nt, 1981, 24, 257-285.
Furth, H. G. Piug~t und kno~c~/rdg~: Tlfroretic.trlfbrtndtrtior1.r (2nd. Ed.). Chicago: Univer-
sity of Chicago Press, 1981.
Gag&, R. M. Contributions of learning to human development. Ps~cho/o~ic~u/ Rcvir~~‘.
1968, 75, 177~191.
Gallagher, J. M.. & Reid, D. The Icurniny throry of‘Pitrgct rrnd Inhrldrr. Monterey, Calif.:
Brooks/Cole, 1981.
Glaser, R. Trends and research questions in psychological research on learning and school-
ing. Edrrcutionul Rrsrurchc~r, 1979, 8, 6- 13.
Jenkins, J. J. Remember that old theory of memory’? Well, forget it! Amc,ricrrn Ps~c~/~o/ogi.~t.
1974, 29, 785-795.
Kitchener, R. F. Epigenesis: The role of biological models in developmental psychology.
Humun Derciopnr~nt, 1978, 21, 141 - 160.
Kitchener, R. F. Predetermined versus probabilistic epigenesis: A reply to Lerner. Hurnurr
De~~lopmcnt, 1980, 23, 73-76.
Kuhn. D. Mechanisms of cognitive and social development: One psychology or two?
Humun Development. 1978, 21, 92p 118.
Lerner, R. M. Nature, nurture, and dynamic interactionism. Hnmun Dr~~c/opmc,nt. 1978,
21, l-20.
Lerner, R. M. Concepts of epigenesis: Descriptive and explanatory issues: A critique of
Kitchener’s comments. Human Devc~lopmrnt, 1980. 23, 63-72.
Lerner, R. M., Skinner, E. A., & Sorell. G. T. Methodological implications of contextual/
dialectic theories of development. Humun Dr~‘c,/opn2c~nt. 1980, 23, 22.5-235.
Moessinger, P. Piaget on equilibration. Human De\,c,/opmrnt. 1978, 21, 255-267.
Moshman, D. To reu//y get ahead, get a metatheory. In D. Kuhn (Ed.), Intellwtutrl clei.c/-
opmrnt hevond childhood. San Francisco, Calif.: Jossey-Bass, 1979.
Moshman, D., & Timmons. M. The construction of logical necessity. Hunfun De,,c,/oprllc,nt,
1982, 25, 309-323.
Overton, W. F., & Reese, H. W. Conceptual prerequisites for an understanding of
stability-change and continuity-discontinuity. Ir~trrnutionul Jorrrnul ofErhtrviortr/ DC,-
l~rlopmrnt, 1981. 4, 999123.
Pepper, S. C. World h?pothese.~t A study in c\~idencc~. Berkeley: University of California
Press, 1970 (Orig. pub. 1942).
Piaget, J. Structurulism. New York, N.Y.: Harper & Row, 1970.
Piaget, J. Genetic Epistc~mology. New York, N.Y.: Norton, 1971. (a)
Piaget, J. Biology und kno,c~kdge. Chicago, Ill.: University of Chicago Press, 1971. (b)
Piaget. J. Insights and illttsions ofphi/o.sopl~y. New York, N.Y .: World Publishing. 197 1. (c)
384 DAVID MOSHMAN

Piaget, J. The development of thought: Equilibration of cognitive structure.s. New York,

N.Y.: Viking, 1977. (a)
Piaget, J., et collaborateurs. Recherches sur I’abstraction r(fl&hissante. Paris: Presses
Universitaires de France, 1977. (b)
Piaget, J. Phenocopy in biology and the psychological development of knowledge. In H. E.
Gruber & J. J. Voneche (Eds.), The essential Piaget. New York, N.Y.: Basic Books,
1977. (c)
Piaget, J. Experiments in contradiction. Chicago, 111.: University of Chicago Press, 1980.
Reese, H. W., & Overton, W. F. Models of development and theories of development. In
L. R. Goulet & P. B. Baltes (Eds.), L$e-span del~elopmentalpsychology: Research and
theory. New York, N.Y.: Academic, 1970.
Riegel, K. F. Foundations of dialectical psychology. New York, N.Y.: Academic, 1979.
von Glasersfeld, E. Radical constructivism and Piaget’s concept of knowledge. In F. B.
Murray (Ed.), The impact ofpiagetian theory on education, philosophy, psychiatry and
psychology. Baltimore, MD.: University Park Press, 1979.
Vygotsky, L. S. Interaction between learning and development. In L. S. Vygotsky, Mind in
society: The development of higher psychological processes. Cambridge, Mass.: Har-
vard University Press, 1978. (This chapter originally published in 1935.)
Youniss, J. Dialectical theory and Piaget on social knowledge. Human Development, 1978,
21, 234-247.
Zimmerman, B. J. Social learning theory and cognitive constructivism. In I. E. Sigel,
D. M. Brodzinsky, & R. M. Golinkoff (Eds.), NW, directions in Piugetiun theory and
practice. Hillsdale. N.J.: Erlbaum, 1981.
RECEIVED: November 11, 1981; REVISED: August 3, 1982