representations to consensually coordinated actions: Towards an
intrinsically social psychology
Paper presented the NCPG on the 8th conference of the International
Society for Theoretical Psychology (ISTP), April 25-28, 2000, Sydney
Theo Verheggen &
In the last few years
of the twentieth century, there have been a number of attempts to
"reclaim" (Greenwood, 1994) or "restore" (Farr, 1998) the social
dimension within social psychology. Apparently, and despite its
explicit label, the discipline has been too much of an individualist
science, as Carl Graumann argued (1986; see also Smith, 1997); at
least since Floyd H. Allport's book Social psychology (1924), which
marks a remarkable but influential choice within the field. From
then on, the behavior of individual persons would dominate social
psychology's (research) agenda. As a consequence, the social became
radically individualized. The properties, actions and cognitions
of single individuals were identified as the sole source from which
we can -- even must -- gain understanding of all forms of human
behavior. Nowadays, within theory and research, the awareness that
there is hardly any act, emotion, belief, cognition, or other mental
state that is not socially affected or culturally modeled, has gained
sufficient ground. When thinking about human behavior, the social
dimension of that conduct needs to be accounted for from the outset.
Not surprisingly, then, several attempts have been made to design
an "intrinsically social psychology" in which the inherently social
nature of people's behavioral repertoire is the object of study.
In a gross outline, it is possible to identify two directions by
which social scientist pursue the intrinsically social nature of
psychological processes. On the one hand, certain cognitions are
believed to hold the key to the social dimension of behavior. On
the other hand, the social is searched for in a shared palette of
ideas, models, scripts, representations, and so forth.(1)
as cognitions of the social
According to Graumann
(1986), psychological social psychology (2)
searches for the social dimension of human behavior in the mental
states, cognitions and emotions of individual people "insofar as
they have been affected by stimulation from other individuals" (p.
100). In the cognitive age, this has been translated in the tendency
to search for "the social" in the cognitions an individual person
has of others, of the group he or she belongs to, of the "outgroup";
ór it is sought for in the cognitions several others have of a certain
person. The social is discounted as a function of the reactions
or stimuli related to other people. As Graumann rightfully argues,
such a psychology studies intra-personal rather than interpersonal
processes. As a sub-discipline of psychology, then, social psychology
is not a social science. It remains the study of socially sterile
persons as the true unit of analysis. Consequently, the social production
of these cognitions is hardly addressed at all. "There is virtually
no consideration of the possibility that cognition itself has social
dimensions (...) and it is simply assumed that the same individual
psychological explanations that apply to our cognition of non-social
objects will apply to the social domain", as Greenwood (1994, p.
95) put it.
as an observer-dependent category
John Greenwood (1994)
focuses attention on the difference between aggregate (or derivatively
social) and intrinsically social phenomena. His point is that we
often confuse "the social" with a feature people may happen to have
in common. For instance, as he would argue (p. 87), all the people
present in Central Park at 4 p.m. on Sunday afternoon do not for
that matter -- of sharing a place and time -- also constitute a
social group. For a group to be truly (intrinsically) social, it
is necessary that the members of the group conform to a set of arrangements,
conventions and agreements to which they are parties. It is along
these arrangements, conventions and agreements that people can be
observed to engage in patterns of interactions by which they can
be distinguished as a group from people that are not parties to
these interaction patterns. The important point for our discussion
is that social scientist themselves often "construct" social groups
by categorizing people on the basis of a feature they are observed
to posses, as in the example. It is thus that scholars produce and
compare such aggregate groups (e.g. woman, Catholics, students,
Danes, and people wearing blue caps) while assuming that the salient
variables that they in fact have introduced themselves (a) actually
represent an existing dimension or force that makes a difference
in everyday (social) life and (b) that these variables/features
can explain behavioral differences between the members of the "groups".
But again, the group is only an aggregate here. It is the constructive
act of, and a category made by the scientist/observer. As such,
these observer-dependent categories have been often mistaken for
the beacons of "the social".
representation approach of Wolfgang Wagner
Quite a different approach
in the "restoration" of the social dimension comes from social representation
theory (SRT). To be sure, there are many variants of SRT, but we
single out Wofgang Wagner's approach (1996; 1998)because he poses
and tackles the type of questions that we pursue also: Wagner looks
for the production principles of social cognitions/representations.
What, then, is a social representation about?
begin with, Wagner observes -- in line with Serge Moscovici -- that
the term "social" must not be understood as a property of objects.
Instead, the adjective "becomes a relational attribute characterizing
the relationship between a person and an object, event or phenomenon
which constitutes his or her group's world" (p. 301). So it is the
relationship between the person and the object that constitutes
the world, more precisely, that constitutes the world for a person
or an observer. An object is, in the words of Wagner (p. 306), any
material, imaginary or symbolic entity that people name, assign
properties and values to, talk about, in short: that they relate
to. The key point is that all these actions we direct towards objects
are social actions, since naming, assigning value, and so forth
imply discourse, elaboration, orientation and coordination of our
actions towards these "objects". By this, we make them social, according
to Wagner. He continues claiming that this assigning process has
to be carried out in a coordinated or tuned fashion with respect
to other members of the community. There has to be some sort of
consensus among these members in order to communicate, understand
what they are doing and talking about, et cetera. We agree with
Wagner (1998, p. 307) that "concerted interaction is the cornerstone
of the social construction of the world". Let us tip our hand early:
The important question is how interaction becomes concerted ~ within
SRT, this remains a question, however.
social, world-constituting relationship we identified is constituted
precisely in the concerted manner in which we elaborate on the undomesticated
"somethings" (cf. Searle's "brute facts", 1995). According to Wagner,
this practice "may be bodily or verbal or both, and [it is] the
expression of and inseparable from the representation" (p. 307).
It is "discursive elaboration of a meaning system" on the one hand,
and "acting as if the object had those characteristics which it
is thought to possess" on the other (ibid.). Wagner's next move
is to identify the representation with the social object it seems
to represent. "A representation and its object are coexistent as
a consequence of people's concerted discourse and conduct and (...)
this discourse and conduct realizes the object in the social world"
(p. 314), and "(...) the world of domesticated objects is the local
universe of representations" (p. 308).
we thus shortly summarize Wagner's argument: (1) in the concerted
manner in which we elaborate on a "something", we create the social
object; (2) that creative, coordinated praxis is the social representation
itself; (3) the something is elaborated upon by means of a verbal,
discursive and/or bodily act.
So far, so good...
According to Wagner, it is through discursive processes that people
come to produce and share a similar set of representations
-- apparently by means of some sort of internalization and externalization
processes that SRT refers to but does not further specify -- such
that people in groups come to "calibrate their minds" (p. 304).
Wagner claims that the social representation exists on the intersection
of personal experience and the "collectively shared experience of
culturally similar others" (p. 301; our emphasis). An object becomes
"a social object within the group's system of common sense and in
the course of interactions in which actors sharing a representation
engage" (p. 307; our emphasis). Likewise, Rob Farr (1998, p. 279)
contends that what lends a social representation its "inherently
social nature" is the fact that it is shared by a number of people.
More precisely, he gives what he calls a minimal definition of a
social representation: "A representation is social if it is, or
has been, in two or more minds" (p. 291). A social representation
is thus an idea, image, belief and so forth that is somehow shared
by to a number of people.
may indeed appear for an observer that people belong to a group
because they share a distinctive feature. This assumption underlies
many social psychological theories and most experimental research.
See, for instance, theories of cultural models (D'Andrade, 1995),
scripts (Fischer, 1991), scenario's, cultural schema's (Strauss
& Quinn, 1997). To give an example, people are held to belong to
a group because they share the same or at least very similar representations.
Or, people are believed to be members of the same culture because
they seem to adhere to the same cultural models or scenario's. It
is then assumed that "the social" is somehow locked-in in this shared
nature of such models, scripts, or representations. Unfortunately,
it not at all clear what term "shared" means in different arguments.
It sometimes refers to an idea that exists in a number of minds,
but this is problematic: how can we be sure that I have exactly
the same idea(s) as my colleague has about this contents paper,
for instance? In other cases, "shared idea" may indicate that a
number of people have very similar -- instead of identical -- notions.
This, however, is merely asking the same question in a different
manner. Confusion only increases when different connotations are
applied within one and the same argument. For instance when people
not only are believed to have experiences and representations in
common, but also to "collectively share" them (as in one of the
just given quotes by Wagner). Is this a pleonasm, then? Or is it
precisely not that? In either case, "shared" does not explain much:
if we all wear "the same" uniform, do we then "share" a collective
uniform? Certainly not!, as Voestermans (1998, personal communication)
would say. Many decades ago, the fuzzyness of the term "shared"
already led Gordon Allport to claim that we should radically avoid
it when describing ideas, norms, and the like (Greenwood, p. 97).
But even if we would have a clear notion of the term, we, as analysts,
must still avoid the pitfall of ourselves constructing the sharedness
of the phenomena we observe. We cannot explain intrinsically social
behavior by merely saying that the interaction partners have the
same representations, for instance, because this is an account in
terms of observer-dependent categories. We then confuse the social
with what people have in common.
sum up our argument so far: Neither cognitions about social phenomena
nor aggregate features can count as social markers since they have
no intrinsically social value. If we want to make a statement about
inherently social psychological phenomena, then, we must not search
for the social in a shared property (independent variable) but instead
direct our attention toward the consensually coordinated interactions
people engage in. The challenge is to understand how these interactions
are produced and how they take shape. Wagner's ideas about social
representations as processes of concerted interaction and as world
constituting "enactions" appear a promising approach. However, he
still adheres to a problematic notion of social representations
as shared representations. Where to find an alternative then?
consensual coordination instead of sharedness
Let us start by repeating
SRT's claim that all the actions we direct towards objects are social
actions since they imply coordination of our behavior towards other
members of our community: what we recognize as, call, think of,
do with, and communicate about an object has somehow become concerted
with respect to others.
enactive paradigm (see Varela, Thompson & Rosch, 1991; Baerveldt
& Verheggen, 1999a/b) also focuses on the mutual coordination of
actions. One of its starting points is the observation that experiencing
agents fundamentally have no direct access to the experiences of
others. This inaccessibility does not exclude these actors from
having social interactions. Quite on the contrary, it is because
we cannot immediately know the feelings and cognitions of our fellow
men that we need to mutually adjust our actions with respect to
others. Put differently: it is because of their experiential closure
that people, as well as other experiencing systems, have to communicate
in order to operate in world populated by other experiencers.(3)
Enactivism states that the interactions we have with others -- other
experiencing systems that is -- are almost always of the form of
second, or higher, order coordination. This means that experiencers
can again (consensually) interact with the product of their concerted
interactions. To given a simple example: a history of coordinated
interactions with respect to a "fair" division of resources among
a couple of people, may lead to a formalization of these patterns
of interactions. The competitors could make a division rule that
they all support. In effect, they can also relate their actions
to this second or maybe even higher order product of their interactions:
they can obey or ignore or change the established patterns of conduct
(more formally, the rule). The important point is that a history
of interactions has gained a particular meaning for the agents,
with which they can interact in effect. This notion comes of course
very close to Wagner's ideas on the social constitution of "objects".
We would like to add that any form of second or higher order coordination
could constitute objects; the latter need not be made explicit in
language. Non-linguistic rituals, dances or melodies too can communicate
"this means", "this is play", "this is as if", and so forth without
linguistically representing such a message.
when the coordinating behavior of two or more experiencing agents
obtains a recursive character, an observer may come to the conclusion
that they have a "shared" reality. Maturana and Varela call such
a seemingly shared reality a "consensual domain" (Maturana & Varela,
1980; Varela, 1979). The word "shared" is misleading however, as
we stated above, because what seems to be shared actually belongs
to the descriptive domain of an observer -- notice that the observer
and the actor/agent may be one and the same organism. A consensual
domain is in fact a "co-operative domain of interactions" (Whitaker,
1997). Therefore, when we speak about culture as a socially shared
reality, we implicitly or explicitly refer to the consensual coordination
of individual actions that constitutes this "shared" reality.
closure implies the impossibility of shared experience. Moreover,
it implies that the experiences of agents differs by definition.
Consensual coordination of actions is therefore a much more accurate
description of the process underlying meaningful world construction
(as is Wagner's calibration or orchestration). And that world is
inherently an enacted one. "Coordinating", here, is about arranging
differences. It should be the object of analysis to understand how
this consensual coordination takes place in real life interactions.
in the case of such consensually coordinated patterns of behavior,
we can justly speak of intrinsically social interactions. In these
interactions, the identity -- as experientially closed systems --
of all the agents that are parties to the interaction must be constitutive
for that interaction. Consider the following example: while driving
your car you hit a biker; by accident and only mildly, fortunately
nobody gets injured, but still, you hit him. In our view, that interaction
cannot be labeled intrinsically social. The experiential autonomy
of the cyclist is not constitutive for the interaction pattern.
To be sure, cycling, driving a car, even hitting a biker are social
actions since these activities are charged with cultural artifacts,
prescriptions, meanings et cetera. Nevertheless, intrinsically social
the interaction it is not. Now consider the occasion that the cyclist
just scared you by popping up out of the blue in front of your bumper,
swearing and ridiculing your driving style. You get a little frustrated
and almost by accident you hit the cyclist, only mildly, fortunately
nobody gets injured, but still... In this case, the other party
is constitutive for the social interaction since he does contribute
to the interaction as experiencer; his identity or autonomy is at
stake in the course of the event. This, we contend, is an example
of an intrinsically social interaction.
Rather than asking
for the ontological status of social or, for that manner, collective
representations (Farr, 1998), we argued that social psychologists
should try to understand the epistemological nature of these and
other purportedly social phenomena (such as scripts, models, scenario's
and so forth) in the first place. We must understand how description,
action, observation, and cognition are inherently social from the
outset. Enactivism can avoid the epistemological and conceptual
pitfalls of (shared) social re-presentationalism. It identifies
the observer-dependent nature of alleged shared social processes
and phenomena. As such, sharedness has an epistemological rather
than an ontological status. Just as Wagner's SRT enactivism focuses
on concerted or consensually coordinated interactions of experiencing
agents. Since we principally have no access to the experiences of
others, we cannot share similar -- let alone the same -- experiences,
representations, scripts, models, and the like. What people "have
in common" is not a set of ready made ideas but a history of interlocked
conduct; the experiencing agents are parties to consensual domains.
That should be the unit of investigation when designing a psychological
study of culture, or for that manner, a psychology of intrinsically
- Some authors
conceive of the social psychological phenomena with reference
to situational factors or the social context (see Cole, 1996;
see also Greenwood, 1994, p. 99). In our opinion, as far as these
theories acknowledge an agent, it brings us again back to the
conceptions of the social as discussed in this text. We therefore
do not discriminate a third orientation, here, in which `the social'
is understood as social context or the like.
- As opposed
to sociological social psychology, the discipline that is concentrated
on the social structures individuals embody (Graumann, 1986, p.
(Baerveldt & Verheggen, 1999a/b) we dealt in detail with the formal
arguments underlying the notion of experiential closure.
Allport, F.H. (1924). Social psychology. Boston: Houghton
Baerveldt, C., & Verheggen, Th. (1999a). Enactivism and the experiential
reality of culture: Rethinking the epistemological basis of cultural
psychology. Culture & Psychology, 5 (2), 183-206.
Baerveldt, C., & Verheggen, Th. (1999b). Towards a psychological
study of culture. Epistemological considerations. In W. Maiers,
B. Bayer, B. Duarte Esgalhado, R. Jorna, & E. Schraube (Eds.). Challenges
to theoretical psychology. North York, Canada: Captus, 296-303.
Cole, M. (1996). Cultural psychology: A once and future discipline.
Cambridge: The Belknap Press of Harvard University Press.
D'Andrade, R.G. (1995). The development of cognitive anthropology.
Cambridge: Cambridge University Press.
Farr, R.M (1998). From collective to social representations: Aller
et retour. Culture & Psychology, 4(3), 275-296.
Fischer, A. (1991). Emotion scripts: A study of the social and
cognitive facets of emotions. Leiden: DSWO Press.
Graumann, C.F. (1986). The individualization of the social and the
desocialization of the individual: Floyd H. Allport's contribution
to social psychology. In C.F. Graumann and S. Moscovici (Eds.),
Changing conceptions of crowd, mind and behavior (pp. 97-116).
New York: Springer.
Greenwood, J.D. (1994). Realism, identity and emotion. Reclaiming
social psychology. London: Sage.
Maturana, H., & Varela, F.J. (1980). Autopoiesis and cognition.
The realization of the living. Dordrecht: Reidel.
Searle, J.R. (1995). The construction of social reality.
New York: The Free Press.
Smith, R. (1997). The Fontana history of the human sciences.
London: Fontana Press.
Strauss, C., & Quinn, N. (1997). A cognitive theory of cultural
Varela, F.J. (1979). Principles of biological autonomy. New
York: Elsevier North Holland.
Varela, F.J., Thompson, F., & Rosch, E. (1991). The embodied
mind: Cognitive science and human experience. Cambridge: MIT
Wagner, W. (1996). Queries about social representation and construction.
Journal for the theory of social behavior, 26 (2),
Wagner, W. (1998). Social representations and beyond: Brute facts,
symbolic coping and domesticated worlds. Culture & Psychology,
4 (3), 297-329.
Whitaker, R. (1997). Self-organization, autopoiesis, and enterprises.
Retrieved June 12, 1997 from the World Wide Web: http://www.acm.org/sigois/auto/main.html