Content uploaded by Raf Vanderstraeten
Author content
All content in this area was uploaded by Raf Vanderstraeten on Mar 31, 2018
Content may be subject to copyright.
Raf Vanderstraeten
Autopoiesis and socialization: on Luhmann’s
reconceptualization of communication and
socialization
ABSTRACT
In 1984, Niklas Luhmann published Soziale Systeme in which he applies the idea
of autopoiesis (= self-production) to social systems. Abstracted from its biological
connotations, the concept of autopoiesis leads to a sharp distinction between
different kinds of autopoietic organization, i.e. between life, consciousness and
communication. According to Luhmann, the relationship between social systems
and human beings cannot be adequately analysed except by taking into account
that they are environments for one another. If this theoretical background is
accepted, the concepts and theor y of socialization need to be revised. Luhmann
takes issues with classical notions such as internalization, inculcation, or ‘social-
ization to the grounds of consensus’ (Talcott Parsons). After a historical overview
of social systems research and general systems theor y, it is indicated how com-
munications trigger further communications and realize the autopoiesis of social
systems. In the second part of the article, the distinction between social systems
and psychic systems is used to discuss issues crucial to socialization theor y. Both
a revision of the concept of socialization, and lines for an empirical research pro-
gramme are proposed in accordance with Luhmann’s theory of social systems.
KEYWORDS: Niklas Luhmann; autopoiesis; social systems; communication;
socialization; education
1. INTRODUCTION
With the publication in 1984 of Soziale Systeme, Niklas Luhmann provided
us with what he himself later called the ‘introductor y chapter’ of a general
theory of modern society. The book – which consists of 675 pages – is an
attempt ‘to reformulate the theory of social systems via the current state of
the art in general systems theory’ (1995: 11). Its central aim is the appli-
cation of the idea of autopoiesis (= self-production) to social systems. Soziale
Systeme wants to indicate the autonomy of social systems with regard to the
production and reproduction of their elemental units. Luhmann argues
that social reality continually organizes its own self-renewal. To clarify the
British Journal of Sociology Vol. No. 51 Issue No. 3 (September 2000) pp. 581 –598
© 2000 London School of Economics and Political Science ISSN 0007 1315 print /1468-4446 online
Published by Routledge Journals, Taylor & Francis Ltd on behalf of the LSE
argument, a large number of concepts are introduced and determined in
reference to one another. In his following books, this general theory of
social systems has been speci ed and applied to par ticular kinds of social
systems – notably to society understood as a comprehensive social system,
and to some of its functional subsystems (economy, science, law, art,
religion and politics). In all these ‘chapters’ of his social theor y, Luhmann
developed the argument – and far more radical than has ever been done
before – that social systems are realities sui generis.
A corollary of this application of the idea of autopoiesis is the sharp
distinction between different kinds of autopoietic organization, and in
particular that between social and psychic systems. While the distinction
between human beings and social systems is rmly established within socio-
logical theor y, Luhmann clearly focused on the division and separation
between these types of systems, and not on their interdependence (e.g. in
terms of roles). Consciousness (thoughts) and communication are, accord-
ing to Luhmann, different modes of autopoietic reproduction. They do
not participate in each other’s self-renewal. In an often repeated, provoca-
tive for mulation: thoughts can only be produced by other thoughts, com-
munication can only be produced by communication. A clear consequence
of this perspective is the virtual absence of the concept of socialization in
his theoretical framework. As prominent as this concept has been in soci-
ology, and especially in systems theoretical sociology (e.g. in the work of
Talcott Parsons), as marginal has it become in Luhmann’s writings. Only a
handful of pages of Soziale Systeme and of the subsequent monographs are
devoted to a discussion of socialization.
This article presents a systems theoretical approach to socialization, that
elaborates on Luhmann’s brief remarks on the subject. It aims to indicate
and underline the improbability of successful socialization. After a brief
historical overview of ‘paradigm shifts’ in social systems research and
general systems theory, some of Luhmann’s core concepts will be pre-
sented. The reconstruction focuses on his de nition of communication.
Next, this perspective is used to highlight a number of issues crucial to
socialization theory. The concept of socialization, the possibility of inten-
tional socialization, and the socializatory effects of educational settings will
be discussed. These applications seek to display the fruitfulness of the
overall theoretical framework, and in particular its sharp distinction
between social and psychic systems, with regard to the interpretation of
empirical phenomena.
2. THE AUTOPOIESIS OF SOCIAL SYSTEMS
2.1. System Dynamics
In the rst decades after the Second World War, systems theoretical
sociology was almost identi able with Talcott Parsons’ structural
582 Raf Vanderstraeten
functionalism. Structural functionalism started from the assumption that
social systems (and societies in particular) require, to maintain themselves,
a nor mative system that is internally coherent and broadly shared by its
members. The stability of this normative system of order – which Parsons
called a structural imperative – ‘explained’ different social processes. It was
also used to de ne the function of socialization. ‘The maintenance of a
normative order requires that it be implemented in a variety of respects:
there must be ver y considerable – even if often quite incomplete – com-
pliance with the behavioral expectations established by the values and
norms. The most basic condition of such compliance is the internalization
of a society’s values and norms by its members, for such socialization
underlies the consensual basis of a societal community’ (Parsons 1966: 14).
In this Parsonian regard, both the family and the school class per form an
instrumental role for society at large; their function is to transmit this
normative structure to the future generations.1
As is well-known, Talcott Parsons’ static conception of structures has
been endlessly criticized, especially in the 1960s and 1970s. Criticism has
focused on Parsons’ commitment to problems of ‘pattern maintenance’,
and par ticularly the maintenance of a system of norms and value orien-
tations (e.g. Gouldner 1970). Sociological theory was said to offer an ‘over-
socialized view of man’ (Wrong 1961; Parsons 1962). Moreover, Parsons
was blamed for exaggerating the importance of widely accepted normative
commitments, and thus for defending the status quo in society. This criti-
cism in a way seemed to overturn social systems theor y in general. Systems
theory was quite generally reproached for conformance with regard to
existing societal structures of norms and value orientations (e.g. Keren
1979). There were, however, at the same time efforts to develop a more
dynamic systems theoretical view of society, and of social str uctures. Still
best-known is Walter Buckley’s description of societies as complex adaptive
systems. According to Buckley (1967, 1968), complex adaptive systems are
characterized by feedback loops, and are able to maintain themselves by
means of structural changes. ‘Persistence or continuity of an adaptive
system may require, as a necessar y condition, change in its structure’
(Buckley 1968: 493).
Attempts to develop a model of dynamic structures brought, on the
other hand, a fundamental problem within the foundations of modern
systems theor y to the fore. General systems theor y – of which Ludwig von
Bertalanffy was the leading author – intended to be a super theory, un-
covering universal principles applying to systems in general: ‘its subject
matter is the formulation and derivation of those principles which are valid
for systems in general’ (von Bertalanffy 1988: 32). Its favourite model was
one of open systems. Characteristic of open systems is that they are not
simply surrounded by an environment – as is the case for a closed system,
such as a clock – but that they actively exchange matter with their en-
vironment. Their boundar y is per meable. Living organisms for example
have to maintain themselves through a continuous in ow and out ow of
Autopoiesis and socialization 583
components (food, excrement, etc.). Without this metabolism, they would
soon die. This environmental contact is at the same time the basic con-
dition and the basic problem of their existence. A system can never control
its environment; the environment is a continuous threat to open systems.
Systems have to continually adapt to their chaotic and unpredictable
environment if they want to maintain themselves.
From the perspective of this general systems theory, the behaviour of
complex, open systems has to be understood as the result of the interaction
between a system and its environment. The perspective accords, in a way,
the primacy to the environment. As a consequence, structural changes
have to be understood as functional, adaptive reactions to the environ-
ment. Attempts to develop a dynamic view of structures within the margins
of this open systems model, however, did nd themselves almost inevitably
confronted with a fundamental problem. Where do open systems get the
‘matter’ which is needed to develop new structures or alter existing ones.
If it depends on a system’s structures which external elements can get an
informational value within the system, how can a system provide itself with
the novel information which it needs to innovate its structures? The
recording of external elements for the purpose of internal structuration
already presumes the structures which have to be constructed (Schmid and
Haferkamp 1987: 8–10). This leads to a circular contention: dynamic
systems are only able to adapt to their environment, if they are already
adapted to their environment (and their environment to them).
This circularity has fascinated cyberneticians, such as Heinz von Foer-
ster, Gordon Pask, or W. Ross Ashby. One can say that they transformed
this problem for systems theor y into a point of departure for the develop-
ment of a fundamentally dynamic view of reality. System dynamics was no
longer attributed to external causes, but to internal efforts. Their work
initially focused on processes of self-reference on the structural level, i.e.
on self-organization, on structures which structure themselves (e.g. von
Foerster 1960). But this focus was later extended. The concept of
autopoiesis, as de ned by Humberto Maturana and Francesco Varela
(1980), situates self-reference on ever y level; it pertains to all units or ele-
mental components of a system. Autopoietic systems reproduce all the
components out of which they arise by means of a network of these com-
ponents themselves and in this way distinguish themselves from their
environment. For Maturana and Varela, the concept of autopoiesis
enabled a characterization of living organisms (in contrast with non-living
organisms). In this regard, Zeleny (1981: 4–5) gives a clarifying example.
The cell . . . is a complex production system, producing and synthesizing
macromolecules of proteins, lipids, and enzymes, among others; it con-
sists of about 105macromolecules on the average. The entire macro-
molecular population of a given cell is renewed about 104times during
its lifetime. Throughout this staggering turnover of matter, the cell
maintains its distinctiveness, cohesiveness, and relative autonomy. It
584 Raf Vanderstraeten
produces myriads of components, yet it does not produce only some-
thing else – it produces itself. A cell maintains its identity and distinctive-
ness during its lifespan. The maintenance of unity and wholeness, while
the components themselves are being continuously or periodically dis-
assembled and rebuilt, created and decimated, produced and con-
sumed, is called ‘autopoiesis’.
Living systems do not live thanks to the import of life from their environ-
ment. They have to produce their own ‘being alive’. This does not mean
that they are windowless ‘monads’ (Leibniz). The environment resonates
in the system by means of the elements which the system itself produces.
Von Bertalanffy’s distinction between open and closed system is tran-
scended. Autopoietic systems are at the same time open and closed systems.
Or, to put it more precisely: they can be open, because they are closed (cf.
Morin 1977: 197–203). Their dynamics resides in their autopoietic closure.
This theor y of autopoietic systems is no general systems theory. It only
pertains to living, organic systems; the concept of autopoiesis was invented
to de ne life. The extension of this concept to other elds and other types
of systems has, none the less, been quickly taken into consideration. One
of the authors who was taking the lead in this regard was Niklas Luhmann
(cf. Geyer 1992; Bailey 1997). Luhmann (1984, 1995) started from the idea
that ‘autopoiesis’, if de ned more abstractly, characterizes a wide range of
systems. He distinguished between organic, psychic and social systems.
Each of these types realizes its autopoiesis in an autonomous mode of
reproduction. Abstracted from biological connotations, Luhmann’s
concept leads to ‘a sharp distinction between meaning and life as different
kinds of autopoietic organization; and meaning-using systems again have to
be distinguished according to whether they use consciousness or com-
munication as modes of meaning-based reproduction’ (Luhmann 1986:
173). But how do social systems realize their autopoiesis? And how does
participation in social systems affect the functioning of psychic systems, i.e.
how is socialization possible?
2.2. Communication
According to Luhmann, social systems use communication as their par-
ticular mode of autopoietic reproduction. The autopoiesis of social systems
emerges from ‘communication’s triggering further communication’
(1995: 218). But this does not yet answer the chief question regarding
social autopoiesis. How does communication realize the self-reproduction of
social systems? How does Luhmann conceive of units of communication?
Customarily, communication is described by means of a metaphor of
‘transmission’. Communication is the transmission of messages or infor-
mation from one place (the sender) to another place (the receiver) by
means of a medium. It includes processes of encoding by the sender in
order to get the message through the medium, and processes of decoding
Autopoiesis and socialization 585
by the receiver in order to get at the meaning of the encoded message.
Against the use of this metaphor of transmission, Luhmann raises several
objections. He prefers to avoid the metaphor because ‘it would burden us
with problematic preliminary decisions’. Firstly: ‘The metaphor of trans-
mission is unusable because it implies too much ontology. It suggests that
the sender gives up something that the receiver then acquires. This is
already incorrect because the sender does not give up anything in the
sense of losing it. The entire metaphor of possessing, having, giving, and
receiving, the entire ‘thing metaphoric’ is unsuitable for understanding
communication’ (1995: 139). Secondly, the metaphor locates what is essen-
tial in communication in the act of transmission, in the utterance (cf.
Shanon 1989). It focuses attention on the sender (and reduces communi-
cation to a question of skills, of communicative competence). But com-
munication emerges only to the extent that the sender’s utterance is
picked up, and processed by a receiver! Thirdly, the transmission
metaphor prematurely assumes an identity of what is transmitted. It might
be the case that the information transmitted is the same for the sender and
the receiver. But the content of the information cannot guarantee this
identity. Infor mation might mean something very different for the sender
and the receiver. The identity of an information is constituted in the com-
munication process.
These objections do lead Luhmann to a reorganization of terminology.
He focuses on communication as an occurrence or event, which emerges
from the processing of selections. According to Luhmann, a unit of com-
munication consists of the co-ordination or synthesis of three different
selections. These selections are: information, utterance and understanding
(‘Verstehen’).2Communication is an emergent, three-part unity.
For Luhmann, information is – in line with today’s standard de nition (cf.
Shannon and Weaver 1949) – a selection from a repertoire of possibilities.
It is the selection that is actualized in the communication. Without this
selectivity of information, no communication would emerge, however
minimal the news value of the exchanges (e.g., if communication is only
engaged in to pass the time and avoid periods of silence). ‘Communication
grasps something out of the actual referential horizon that it itself consti-
tutes and leaves other things aside’ (Luhmann 1995: 140). A communi-
cative act, however, does not make a selection in the same way in which one
grabs one thing rather than another off the rack. Pieces of information
don’t exist ‘out there’, waiting to be picked up by the system. Such an
interpretation would bring us back to the ‘thing metaphoric’ of the sender-
receiver model. Communication is not just a two-part matter of sending
and receiving messages; the selection of information is one of its crucial
components. ‘What is uttered is not only selected, but also already a selec-
tion – that is why it is uttered’ (1995: 140).
The second selection concerns the choice of a behaviour, an utterance,
that does express the information. Information should be provided in a
form which the utterer and the addressee are able to understand.
586 Raf Vanderstraeten
Communication does require an adequate standardization of the utterance
(e.g. linguistic forms). To be sure, this utterance can occur intentionally or
unintentionally. It is also possible without language, e.g. through ‘reveal-
ing’ looks, through dress or out t, through absence, etc. But the utterance
must always be interpretable as selection, and not just appear as a sign of
something else. ‘In this sense, rushing about can be observed as a sign of
urgency, just like dark clouds as a sign of rain. But it can also be interpreted
as a demonstration of urgency’ (1995: 151). The difference between both
interpretations underlines at the same time the importance of the last
selection of Luhmann’s concept of communication. Understanding implies
a change in the state of the receiver. ‘The communication determines a
state of the receiver that would not exist without the communication’
(1995: 148). But not every change in the receiver’s state is equivalent to
understanding. What is decisive is the fact that the third selection can base
itself on a distinction, namely the distinction between infor mation and its
utterance. Understanding therefore implies more than mere observation;
it only takes place if the receiver construes the information from the utter-
ance. Communication depends on the processing of this difference.
Communication is a three-part unity. It does not come about if the
addressee does not x her own state on the basis of uttered information. It
does not come about without understanding. In contrast to prevailing
descriptions of communication, which emphasize the utterance of infor-
mation (the act of transmission), Luhmann underlines the importance of
the third selection, of ‘Verstehen’. Seen from this perspective, one could
say that communication is made possible from behind. Understanding
(and understanding will almost inevitably contain some misunderstand-
ing) concludes the communicative act. An understanding, however, needs
to manifest itself. The receiver needs to show her understanding, by
addressing herself to the information component (e.g., question what is
said) or to the utterance (e.g., question the way something is said). A com-
munication necessitates a new communication. Each communication ‘is an
element only as an element of a process, however minimal or ephemeral
that process may be’ (1995: 144). Communications conclude preceding
communications and enable connecting ones. The elements of social
systems are recursively produced and reproduced by a network of com-
munications and cannot exist outside of such a network. They organize
their own renewal autopoietically.
2.3. System and Environment
Social systems are, as we have seen, emergent realities that use communi-
cation to process meaning. They consist of communications, not ‘human
beings’ (the term indicates both the psychic and the organic systems of
human beings). Human beings are part of the social environment.3This
point has important consequences.
Autopoiesis and socialization 587
If one views human beings as part of the environment of society (instead
of as part of society itself), this changes the premises of all the traditional
questions . . . It does not mean that the human being is estimated as less
important than traditionally. Anyone who thinks so (and such an under-
standing either explicitly or implicitly underlies all polemics against this
proposal) has not understood the paradigm change in systems theor y.
(Luhmann 1995: 212)
The distinction between social system and environment offers the possi-
bility of conceiving human beings in a way that is both more complex and
less restricting than if they had to be interpreted as parts of the social
order. Because they are part of the environment of the societal system,
human beings are conceded greater freedom (greater complexity) than
social roles, norms and structures would allow. This raises the following
questions: How do human beings participate in communication, notwith-
standing the autopoietic closure of psychic systems? How does partici-
pation in communication contribute to psychic system formation?
For Luhmann, meaning – as the medium which psychic and social
systems use – causes and enables a fundamental instability of these systems.
The elements, out of which they exist, have the character of events, i.e. they
vanish immediately after their appearance. Their elements are continually
replaced by other elements (different thoughts, different communi-
cations). Psychic and social systems are radically temporalized systems. This
characteristic allows for a high degree of congruence between social
systems and psychic systems, who tr y to communicate. Communications
can be at the same time conscious events; thoughts can be communicated.
This does not imply that the sociality of communication becomes entirely
present in one’s mind, nor that everything that goes on in one’s mind
could be entirely communicated. Social systems, however, can properly
assume that psychic systems experience and know what has been com-
municated. Also, the activities of psychic systems will be more or less deter-
mined by the ongoing communication.
This congruence of psychic and social systems is compatible with the
idea of autopoietic closure. As Luhmann (1995: 215) indicated:
To be sure, interpenetrating systems converge in individual elements –
that is, they use the same ones – but they give each of them a different
selectivity and connectivity, different pasts and futures . . . The elements
signify different things in the participating systems, although they are
identical as elements: they select among different possibilities and lead
to different consequences .
Social systems and psychic systems thus remain separated. Their con-
gruence is only temporar y and vanishes time and time again. For psychic
systems, the so-called turntaking of active and passive participation in com-
munication almost inevitably re-establishes the difference between psychic
588 Raf Vanderstraeten
and social systems. The mind might for example wander away, think of
something incommunicable, interrupt or pause, while the burden of com-
municating passes to somebody else. Also, communication can be rejected.
Psychic systems do not have to accept what is communicated, or how it is
communicated.
3. CONFORMANCE/DEVIANCE
3.1. Socialization
The preceding remarks already hint at issues of socialization, and its
systematized and planned version: education. The course pursued in the
preceding sections should, however, also be able to illuminate Luhmann’s
reticent and sceptical position with regard to the traditional systems
theoretical conceptualizations of these issues. What can be said about
socialization, if the autopoietic closure of social and psychic systems is
taken into account?
For Luhmann, issues of socialization fall largely beyond the scope of a
theory of social systems. Socialization, as he uses the concept, refers to
changes that take place in society’s environment . Socialization is the process,
steered by communication, that forms the psychic system and the bodily
behaviour of human beings (1995: 241). Luhmann’s concept impinges on
several system references, and it is deliberately open-ended. It takes issue
with classical notions such as ‘internalization’, ‘inculcation’ or ‘socializa-
tion to the grounds of consensus’ (Parsons 1966: 14). It does not refer to
the ful lment of societal functions, but overlays positively and negatively
valued effects, and comprises conforming and deviant, healthy and patho-
logical behaviour. As Luhmann writes: ‘Socialization in this sense is no
occurrence structured by the standards of success (which at worst could
fail). A theory that binds the concept of socialization to the creation of
adaptive behaviour that conforms to expectation cannot explain the emer-
gence of opposite behavioral patterns, and it is helpless before discoveries
such as, for example, that adaptation can have neurotic consequences’
(1995: 241). For anyone familiar with the tradition of socialization theory,
Luhmann’s sweeping de nition certainly entails surprising implications. It
highlights the major weaknesses of this tradition.
To indicate the exact purport of Luhmann’s concept of socialization,
two clari cations are necessar y. First of all, socialization can only be self-
socialization. For human beings, participation in communication cannot
result in the transmission of knowledge, nor in the internalization of the
norms and value orientations of a social group. The meaning of norms,
rules, habits, etc. which are ‘transmitted’ does not remain the same. In the
different participating systems, these elements have different meanings.
They signify different things, while they select among different possibilities
and lead to different consequences. In this sense, socialization presupposes
Autopoiesis and socialization 589
deviant reproduction! It is always self-socialization. The basic process is the
self-referential reproduction of the system that brings about socialization in
itself. As Luhmann stresses, this kind of interpretation does not obscure the
decisive role of the environment in the process of socialization. Autopoietic
closure is, as we have seen, the condition for environmental openness.
‘Moreover, it makes little sense to ask whether the system or the environ-
ment is more important in determining the result of socialization, because
precisely this difference makes socialization possible’ (1995: 242).
Secondly, socialization inevitably reiterates the option between con-
forming and deviant behaviour. A norm or rule cannot be an item for
socialization in itself. It can only be presented together with its alternative
(how would one otherwise account for the possibility of non-confor-
mance?). Socialization creates so-called bifurcations; it enforces a choice
between conforming to expectations or behaving differently, between
adaptation or resistance. As indicated before, the concept of socialization
cannot be aligned with the internalization of collective norms – which
would allow, at the most, to discriminate between successful and unsuc-
cessful socialization. Socialization brings about a continuous processing of
options, i.e. of conformance vs. deviance, of commitment vs. non-commit-
ment, of attraction vs. aversion. The possibility to reject the instruction or
information which a communication contains marks the distinction
between social system and psychic system. Only on the basis of this distinc-
tion is an assessment of conforming vis-à-vis deviant behaviour possible. A
social system might then seek to in uence what appears as contingent indi-
vidual behaviour; its reactions can depend upon whether this behaviour
does or does not live up to the norm (cf. Luhmann 1981: 161–3).
These bifurcations have far-reaching consequences for the socialization
process. But socialization is also determined by which value becomes the
dominating experience. In our late-modern or post-modern society, a
preference for deviance might be assumed (Luhmann 1987: 175–6). Struc-
tural individualization gives way to increasing deviance, simply because
non-conformance (originality) offers the best opportunities to display
one’s individual uniqueness. There have developed many patterns of posi-
tive deviance in our society, like its dominant spirit of performance and
competition, its emphasis on exceeding normal expectations (more,
better, higher, faster, further, etc.), but also its legitimization of sub-
cultures (e.g. youth cultures), its transient life and fashion styles, which
celebrate deviance from what’s usual, etc. Channelling the possibility of
deviance into these forms, and thus intermingling them with conformity
with modern norms and values, may turn out relatively well. How this
happens, and with what consequences for the social system and for the
psychic systems that are involved, needs yet to be studied (e.g. by means of
longitudinal social-psychological research). It is, on the other hand, also
possible that patterns of negative deviance will become more penetrating.
The opportunities which our contemporar y society generate might en-
danger its own structural characteristics (cf. Vanderstraeten 1999a).
590 Raf Vanderstraeten
3.2. Intentional socialization
For Talcott Parsons, socialization ful lled a fairly unambiguous role within
society. Due to socialization, individuals bear the stamp of their social
environment. Their inner structure is determined by the norms and value
orientations of the society in which they live. The model of open systems,
of which Talcott Parsons (and many other social scientists) made use,
indeed leads to an ‘oversocialized view’ of (wo-)man. The possibilities
which human beings have to travel a certain distance, to use their indi-
vidual degrees of freedom, cannot be adequately taken into account. If
attention is given to the operational closure or autopoiesis of psychic
systems and social systems, it is no longer possible to describe socialization
in terms of the transfer of a meaning pattern from one system to the other.
The interaction between a human being and its social environment might
or might not provoke particular structural changes in the ‘inner sphere’ of
the individual; a human being might or might not adapt to particular
aspects of its environment. There is always the possibility of choosing a
different path. An interesting question is whether this possibility of resist-
ance disappears or increases when education comes into play.
Education, according to Luhmann, is an activity of social systems special-
izing in ‘people processing’. Of course, it affects the psychic systems which
are actively or passively involved. But education itself is – in contrast to
socialization – a communicative activity, and displays the characteristics of
social systems. Its distinctive unity (or autopoiesis) can only be understood,
when this de nition is put to use. Consequently, Luhmann has on repeated
occasions paid attention to education, and especially to the structural
characteristics of the educational system and pedagogy’s re ection of the
particularities and problems of this social structure (notably Luhmann and
Schorr 1988). Here, I will not discuss this work (cf. Vanderstraeten 1995),
but continue the course pursued in the preceding sections and focus on
education as intentional socialization.
While socialization is limited by/to the stimuli of the socializing context,
education strives for a particular, ‘unusual’ output. Luhmann writes: ‘Edu-
cation is (and here it differs from socialization) action that is intentional-
ized and attributable to intentions’ (1995: 244). It aims to attain something
that cannot be left to chance socializing events, something that presup-
poses co-ordinating a plurality of efforts. The modes of behaviour that one
would like to achieve are de ned; the situation from which one begins is
evaluated (grade level, ability, previous learning experiences); the peda-
gogical means to achieve what could not occur by itself are chosen. The
current large-scale organization of learning situations, school classes, and
school systems is only the application of this principle. For societies, social-
ization suf ces as long as social mobility and internal complexity are low.
But once a relatively high degree of complexity is reached, they cannot
seem to avoid going beyond mere socialization and mere ad hoc education.
‘Only thus can they reproduce knowledge and capabilities acquired in long
Autopoiesis and socialization 591
sequences of co-ordinated individual steps. Only this enables processes of
specialization and the distribution of roles on the basis of specialization’
(1995: 206). For the concept of education, however, it makes no difference
whether educational communications have or have not differentiated
themselves into an autonomous functional subsystem of society (cf.
Vanderstraeten 1999b).
Socialization comes about by living in a social context. It presupposes the
possibility of reading the behaviour of others as selected information, e.g.
about dangers or social expectations. As indicated before, the meaning of
this communication can be rejected if the addressee nds the information
unsatisfactor y or unacceptable. Education cannot eliminate this possibility
of resistance. It cannot be conceived of as the rational form of socialization,
as successfully effective action.4On the contrary, intentional communi-
cation with educational goals will double the motives for rejection. The
addressee now also has the opportunity to reject the communication
because it intends her education, if she refuses the role of someone who
needs to be educated. In other words, intentional communication enables
the addressee to oppose both the information component and the utterance
(Luhmann 1987: 178–80). This situation illustrates a more general con-
clusion which Luhmann has often stressed (e.g. 1982), namely the evolu-
tionary improbability of the functional differentiation of particular modes
of operation (economic, political, artistic, religious, educational, etc.).
One can further underpin this conclusion with a social experiment: try to
educate someone who does not expect to be educated, for example a
young boy or girl you meet by chance on the street. It is indeed very likely
that your efforts will not be appreciated and will fail.
In all types of ‘people processing’ systems (e.g. spiritual, medical, legal,
or therapeutic), the targeted changes mostly cannot be realized without
the collaboration of the clients. This characterizes the condition of pro-
fessional work in general, and explains the high degree of professional
autonomy and the importance of face-to-face interactions in the course of
the treatment (cf. Abbott 1988). On the other hand, clients often enrol in
these ‘people processing’ systems because of biographical crises, and thus
do not always need to be urged on to collaborate. Education nds itself in
an exceptional position in this regard. Because children have to go to
school, this source of motivation or willingness to collaborate mostly fails.
One may conclude that education is in need of social and cultural sup-
ports, which neutralize the possibility of rejection. In schools for example,
the educational intention is mostly attributed to the institution and not to
the momentar y decisions of the teacher. Curricula specify what needs to be
acquired. But the dominance of organizational and institutional arrange-
ments can only partially unburden the teacher. Pupils are continually
engaged in reading the behaviour of their teacher. Observing whether one
is being observed or is temporar y out of the teacher’s sight, hiding behind
another one’s back, etc. are frequent operations in a classroom.
Some historical obser vations might underpin this argument. The
592 Raf Vanderstraeten
morphogenesis of our modern educational system, which can be situated
at the end of the eighteenth century, encompasses a number of inter-
related changes: the so-called discovery of the child, the universalization of
classroom education accompanied by the professionalization of the teacher,
the development of new curricular principles (Allgemeinbildung). It is no
coincidence that these developments occurred in the same period, and
mutually reinforced each other. This triadic structure was indispensable
for the morphogenesis of the modern system of education. The curricu-
lum stabilizes the unstable dual relationship between pupils and teacher
(cf. Weick, 1969: 38). It is also interesting to note that, at the end of the
eighteenth century, the very possibility of education in the classroom
became an important topic in the literature. One doubted whether a
teacher could exercise the authority expected of an educator, because
there were no blood-ties between the teacher and his pupils (Vander-
straeten 1995: 113–9).
In sum, education cannot guarantee that it will be successful. Moreover,
concretizations of pedagogical behaviour are laden with difference. They
indicate lines of success and thereby establish the possibility of failure.
Then, the question is how pupils do react, when they are constantly con-
fronted with this option, and when they are constantly expected to
conform to expectations.5It makes sense to assume once more that they
will look for some kind of ‘opting-out’ strategy. They can react with total
rejection, but also with unexpectedly good performance, with non-
chalance vis-à-vis evaluation criteria, with humour and irony, with the
celebration of a deviant school or youth subculture, with slang language,
with deviant assessments of qualities and personal merits, etc. This ‘opting-
out’ strategy has, of course, an impact on the educational process and the
teacher. Burn-out feelings of teachers for example are not only related to
low wages and lack of career prospects, but can also be attributed to ‘criti-
cal’ attitudes of pupils and to classroom management problems.
3.3. Secondary socialization
As we have seen, communication succeeds when three selections (infor-
mation, utterance, understanding) form a unity to which further com-
munication can connect. This unity that is communication can never be
entirely reduced to the meaning of an intended and attributable action,
not even if the action itself wishes to be communication or contains com-
municative aspects. In social systems, intentional action cannot control the
difference(s) it makes. It also communicates its own intention, and this is
experienced and evaluated within the system. It makes it possible to react
to this intention as such, and to seek and nd ‘other possibilities’. Inten-
tional action is, so to speak, caught in the coils of self-reference.
When a pedagogically stylized act communicates itself its intention, the
person who is expected to be educated acquires the freedom to travel some
distance – e.g., to pursue the intention out of mere opportunism or to
Autopoiesis and socialization 593
avoid it as much as possible. In systematized educational settings – where
an elaborate apparatus of goals, tests, interventions, etc. is put to use to
attain a certain output – these unintended and mostly unforeseen effects
will be multiplied, and have particular rami cations. ‘They [pedagogical
means] transfor m equality into inequality. They motivate and discourage.
They link experiences of success to experiences of success and experiences
of failure to experiences of failure. They promote attitudes that make it
possible to handle educational problems in special ways via educators,
teachers, schools, and grade levels’ (1995: 207). One can describe these
effects as secondary socialization – ‘secondary’ not in the sense of follow-
ing a primary socialization (e.g. Berger and Luckmann 1966: 120–35), but
as the consequence of the particular social settings that are used to
educate, to socialize intentionally. Some of them, especially those that
ensue from classroom education, are of course currently pretty well-known.
As the so-called hidden curriculum, they have been (and continue to be)
the object of extensive research. The previous remarks on socialization and
education, however, enable a reformulation of the explanator y goal of this
research perspective.
The concept hidden curriculum is built by means of the traditional
sociological distinction between manifest and latent functions/structures
(Merton 1963: 19–84).6It de nes a contrast between the expressed or
manifest purposes of the ofcial curriculum, and the latent functions of
the system, which are ful lled alongside the of cial curriculum. Accord-
ingly, the hidden curriculum is promulgated by the way schools are organ-
ized and operated as much as by explicit teaching methods and content,
and may be far from the expressed motives of teachers and curriculum
planners. It has in fact provoked a fairly pessimistic account of the possi-
bilities of a teacher. No matter what she thinks and does, or how hard she
tries, apparently it is the hidden curriculum which determines what is
really learned at school. The concept presupposes – and here it clearly
reveals its structural-functionalist origins – a high degree of structural
determination of the educational process. To what degree these long-
lasting effects do back up the dominant structures and value orientations
of modern society (in more familiar words: to what degree the hidden
curriculum prepares students, in affective as well as cognitive ways, for their
adult lives as workers and citizens in a liberal capitalist society), is on the
other hand an issue which has never been settled.
Starting from an ‘autopoietic systems’ model (instead of an ‘open
systems’ one), the premise of the structural determination of education in
the school class presents only half the picture. With Luhmann, one can
argue that autonomous functional systems develop a susceptibility to both
structural and operational forms of determination at the expense of an
external, environmental determination of the system. This is exactly what
their autopoietic closure implies. With regard to the educational system
and the school class, the operational perspective points to the typicalities,
peculiarities and consequences of the ongoing operations in the
594 Raf Vanderstraeten
classroom. Occasional deviant or conformist behaviour, certain unex-
pected events, surprising reactions, etc. might deeply in uence the habitus
or self-concept which each student develops. What is attributed to the
hidden curriculum is an important part of the effects of an autonomous
educational system. At the same time, however, it should be acknowledged
that chance is to an important extent at work in education, while minimal,
occasional events may induce major structural changes in psychic systems.
In research on secondar y socialization, one needs to look ver y generally
for the intended and unintended consequences of an autonomous edu-
cational system that aims to rationalize socialization.
The unintended results of education in schools cannot be eliminated
with the help of a careful selection of subject matter – no matter how much
this selection focuses on usefulness in later, professional life, on ‘pure’
intellectual development, or on the students’ life world and interests. They
inevitably appear in a pedagogical context (Vanderstraeten 1997). It is
against this background that Luhmann defends his fairly critical and
pessimistic stance vis-à-vis the role of education in modern society.
The autonomy of a differentiated input /output arrangement must then
submit to correction a reality it has itself created and direct its counter-
intuitive behavior back to reality. A system that is structured too improb-
ably and that tries to identify itself entirely with the transformation of
input into output ends up having to deal with the problems resulting
from its own increase-directed reductions. (Luhmann 1995: 207)
For Luhmann (1985), pedagogy’s major task is to ensure that the price of
these unintended socializatory effects is not too high, and that the result is
not worse than omitting educational efforts altogether.
4. CONCLUSION
Looking back at the histor y of the social sciences, one can easily observe
that the origins of socialization theor y coincide with the development of
sociology as scienti c discipline. For founding fathers such as E. Durkheim,
G. Simmel or G. H. Mead, socialization theor y was at the core of socio-
logical theor y. The concept of socialization contributed to the overall
orientation of classical sociology on structural stability. Until the 1960s, the
discipline predominantly focused on the communication of social experi-
ences to the younger generations, and on the transmission of culture,
norms and value orientations. In retrospect, it should not surprise that this
concept lost its central theoretical position when processes of change
became of focal concern to the social sciences. Moreover, empirical
research questioned interpretations of socialization in terms of internal-
ization. It became evident that the young generations are not merely
passive recipients. One way or another, their active participation needed to
be taken into account. But theoretical and empirical research have
Autopoiesis and socialization 595
hitherto mostly developed alongside each other (cf. Burkitt 1991). What
Luhmann’s theor y offers is a framework within which a theoretically con-
trolled reconceptualization of socialization is possible.
An adequate theor y of socialization needs to take different system refer-
ences into account. It needs to take individual persons and social systems
into account. Socialization is not simply the inculcation of societal values
and norms, nor the realization of individual talents. How an individual
develops, how the ‘possible world’ of an individual changes, depends upon
the social systems in which she is involved (family, peer group, school class,
etc.). Social systems select from individual possibilities (and possibilities
that are not selected will probably waste away). Participation in social
systems creates, on the other hand, additional opportunities for persons.
Socialization and education are always dependent upon what social inter-
action allows. Concrete patterns of social interaction create the difference
between possibility and reality, and it is this difference that constitutes the
effects of socialization and education.
Luhmann’s theor y is not just helpful with regard to issues of socializa-
tion. In a sense, these issues even provide a special test case. They situate
themselves at the margins of Luhmann’s theor y of social systems and of
modern society. As one of the translators of Luhmann’s work noted: ‘At
present, Luhmann’s theor y of social systems is the only general social
theory that can claim to introduce a new paradigm to the eld . . . [I]f
accepted, Luhmann’s proposal will radically change the conventional ways
of doing social theor y’ (Fuchs 1988: 21). Therefore, this article has tried to
display the close connection between Luhmann’s theoretical concepts and
what is observed with regard to socialization. It has sought to indicate the
merits of the sharp distinction between social systems and human beings.
Hopefully, it has been able to provide a glimpse of the direction this new
paradigm points to.
(Date accepted: November 1999)Raf Vanderstraeten
Faculty of Social Sciences
Utrecht University
E-mail: R.Vanderstraeten@fss.uu.nl
NOTES
596 Raf Vanderstraeten
1. In his monograph on the functions
and structures of the (American) family,
for example, Parsons writes: ‘the central
focus of the process of socialization lies
in the internalization of the culture of
the society into which the child is born.
The most impor tant part of this culture
from this focal point consists in the pat-
terns of value which in another aspect con-
stitute the institutionalized patterns of the
society’ (Parsons and Bales 1956: 17; see
also Parsons 1959).
2. This distinction between compon-
ents is related to the distinction, intro-
duced by Karl Bühler, between functions
of linguistic communication. It comes also
close to John Austin’s typology of utter-
ances and acts, viz. locutionary,
illocutionar y and perlocutionary language
acts, and to John Searle’s fairly similar
classification of speech acts. But these
authors depart from an action-theoretical
approach to communication, within which
communication is synonymous with the
transmission of information (cf. Luhmann
1995: 117).
3. For Luhmann, saying that social
systems consist of (relations betw een)
human beings is no more appropriate then
saying that human beings consist of
(associations among) organic cells. Of
course, social systems utilize human
resources such as consciousness to process
communication, just as human beings
cannot have consciousness without
ongoing neuronal and physiological pro-
cesses. Autopoiesis qua life and qua con-
sciousness is a presupposition for forming
social systems, which means that social
systems can actualize their own reproduc-
tion only if they can be sure that life and
consciousness will continue. But social
systems also draw their boundaries so as to
exclude the parts of human beings they
attribute to their environment, just as
concrete human beings do not identify
themselves in terms of organismic pro-
cesses of which they may or may not even
be aware. Although, stated this way,
Luhmann’s argument seems obvious, its
consequences have been highly contested.
Critics mostly speak of an inhuman theor y
(e.g. Zeleny and Hufford 1992; Mingers
1995).
4. This point utterly contradicts the
opinion of many social theorists, such as
for example that of Emile Durkheim. In
Durkheim’s inaugural lecture at the Sor-
bonne, delivered when he become a pro-
fessor of sociology and education, his
opinion is summarized as follows: Edu-
cation ‘is above all the means by which
society perpetually recreates the con-
ditions of its very existence. Can society
survive only if there exists among its
members a suf cient homogeneity? Edu-
cation perpetuates and reinforces this
homogeneity by xing in advance, in the
mind of the child, the essential similarities
that collective life presupposes. But, on the
other hand, without a certain diversity,
would all co-operation be impossible? Edu-
cation assures the persistence of this neces-
sary diversity by becoming itself diversi ed
and by specializing. It consists, then, in one or
another of its aspects, of a systematic socializa-
tion of the young generatio n’ (Durkheim
1956: 123–4, italics are mine).
5. For a comparable discussion of unin-
tended consequences of (educational )
planning and steering, see Vanderstraeten
(1997, 1999a). These articles also deal
with the specific characteristics of the
educational system of our modern society.
6. For Merton, the functions of a social
practice are its ‘observable objective con-
sequences’ (1963: 24). Manifest functions
are those outcomes that are intended and
recognized by the agents concerned; latent
functions are those outcomes that are
neither intended nor recognized. Notice
that this conceptual denition refers to the
input-output schema of the ‘open systems’
model. The system’s function is dened by
its transformation of input into output,
and the internal conditions of this trans-
formative performance can be seen as
structure.
BIBLIOGRAPHY
Abbott, A. 1988 The System of Professions: An
Essay on the Division of Exper t Labor,
Chicago: University of Chicago Press.
Bailey, K. D. 1997 ‘The autopoiesis of
social systems: assessing Luhmann’s theory
of self-reference ’, Systems Research and
Behavioral Science 14(2): 83–100.
Berger, P. L. and Luckmann, T. 1966 The
Social Construction of R eality. A Treatise in
Sociology of Knowledge, New York: Double-
day.
Buckley, W. 1967 Sociology and Modern
Systems Theory, Englewood Cliffs: Prentice-
Hall.
—— 1968 ‘Society as a complex adaptive
system’ in W. Buckley (ed.) Modern Systems
Research for the Behavioral Scientist, Chicago:
Aldine.
Burkitt, I. 1991 ‘Social selves: theories of
the social formation of personality’,
Current Sociology 39(3): 1–229.
Durkheim, E. 1956 Education and Sociology,
Glencoe: Free Press.
Fuchs, S. 1988 ‘Translator’s introduction’,
Sociological Theor y 6(1): 21–5.
Geyer, F. 1992 ‘Autopoiesis and social
systems – 1’, International Journal of General
Systems 21(4): 175–83.
Autopoiesis and socialization 597
Gouldner, A.W. 1970 The Coming Crisis of
Western Sociology, New York: Basic Books.
Keren, M. 1979 ‘Ideological implications
of the use of open systems theor y in politi-
cal science’, Behavioral Scienc e 24(5):
311–24.
Luhmann, N. 1981 Soziologische Aufklärung
3. Soziales System, Gesellschaft, Organisation,
Opladen: Westdeutscher Verlag.
—— 1982 The Differentiation of Society, New
York: Columbia University Press.
—— 1984 Soziale Systeme. Grundriß einer
allgemeinen Theorie, Frankfurt a.M.:
Suhrkamp.
—— 1985 ‘Erziehender Unter richt als
Interaktionssystem’ in J. Diederich (ed.)
Erziehender Unterricht – Fiktion und Faktum?,
Frankfurt A.M.: GFPF.
—— 1986 ‘The autopoiesis of social
systems’ in F. Geyer and J. van der Zouwen
(eds) Sociocybernetic Paradoxes: Observation,
Control and Evolution of Self-steering Systems,
London: Sage.
—— 1987 Soziologische Aufklärung 4.
Beiträge zur funktionalen Differenzierung der
Gesellschaft , Opladen: Westdeutscher
Verlag.
—— 1995 Social Systems, Stanford: Stanford
University Press.
Luhmann, N. and Schorr, K. E. 1988 Reex-
ionsprobleme im Erziehungssystem, Frankfurt
a.M.: Suhrkamp.
Maturana, H. R. and Varela, F. J. 1980
Autopoiesis and Cognition. The Realization of
the Living, Dordrecht: Reidel.
Merton, R. K. 1963 Social Theor y and Social
Structure, Glencoe: Free Press.
Mingers, J. 1995 Self -producing Systems:
Implications and Applications of Autopoiesis,
New York: Plenum Press.
Morin, E. 1977 La méthode. 1: La nature de la
nature, Paris: Seuil.
Parsons, T. 1959 ‘The school class as a
social system: some of its functions in
American society’, Harvard Educationa l
Review 39(3): 297–318.
—— 1962 ‘Comment on “The oversocial-
ized conception of man” by Dennis
Wrong’, Psychoanalysis and Psychoanalytic
Review 10(2): 322–34.
—— 1966 Societies. Evolutionar y and
Comparative Perspectives, Englewood Cliffs:
Prentice-Hall.
Parsons, T. and Bales, R. F. 1956 Family,
Socialization and the Interaction Process,
London: Routledge and Kegan Paul.
Schmid, M. and Haferkamp, H. 1987
‘Einleitung’ in H. Haferkamp and M.
Schmid (eds) Sinn, Kommunikation und
soziale Differenzierung. Beiträge zur Luhmanns
Theorie sozialer Systeme, Frankfurt a.M.:
Suhrkamp.
Shannon, C. E. and Weaver, W. 1949 The
Mathematical Theory of Communication,
Urbana: University of Illinois Press.
Shanon, B. 1989 ‘Metaphors for language
and communication’, Revue Internationale
de Systémique 3(1): 43–59.
Vanderstraeten, R. 1995 Leren voor het leven.
Opvoeding en onderwijs als maatschappelijk
deelsysteem, Leuven-Apeldoorn: Garant.
—— 1997 ‘Circularity, complexity and
educational policy planning. A systems
approach to the planning of school pro-
vision’, Oxford Review of Education 23(3):
321–32.
—— 1999a ‘Educational expansion in
Belgium: A sociological analysis using
systems theor y’, Journal of Education Policy
14(5): 507–22.
—— 1999b ‘Versäulung und funktionale
Differenzierung. Zur Enttraditional-
isierung der katholischen Lebensformen ’,
Soziale Welt 50(3): 297–314.
von Bertalanffy, L. 1988 General Systems
Theory: Foundations, development, appli-
cations, New York: Braziller.
von Foerster, H. 1960 ‘On self-organizing
systems and their environments’ in M. C.
Yovits and S. Cameron (eds) Self-organizing
Systems, London: Pergamon.
Weick, K. E. 1969 The Social Psychology of
Organizing, Reading: Addison Wesley.
Wrong, D. 1961 ‘The oversocialized con-
ception of man in modern sociology’,
American Sociological Review 26(2): 183–93.
Zeleny, M. 1981 ‘What is autopoiesis?’ in
M. Zeleny (ed.) Autopoiesis: A Theor y of
Living Organizatio n, New York: North
Holland.
Zeleny, M. and Hufford, K. D. 1992 ‘The
application of autopoiesis in systems analy-
sis: are autopoietic systems also social
systems?’, International Journal of General
Systems 21(4): 145–59.
598 Raf Vanderstraeten
