Ecological Communication, Niklas Luhmann PDF

Ecological
Communication
NIKLAS LUHMANN
Translated by John Bednarz, jr.
The University of Chicago Press

Originally published as Okologiscbe Kommunikation: Kann die moderne
Gesellschaft sicb auf okologiscbe Gefabrdungen einstellenr
1986 Westdeutsches Verlag GmbH, Opladen.
The University of Chicago Press, Chicago 60637

Polity Press, Cambridge
1989 by Polity Press
All rights reserved. Published 1989
Printed in Great Britain
98 97 96 95 94 93 92 91 90 89 54321
Library of Congress Cataloging-in-Publication Data
Luhmann, Niklas.
Ecological communication.
Translation of: Okologische Kommunikation.
Includes index.
1. EcologyPhilosophy. 2. ManInfluence on nature.
3. Environmental protectionPhilosophy. 4. Environ-
mental policy. I. Title.
QH540.5.L8313 1989 574.5'01 89-4843
ISBN 0 - 2 2 6 - 4 9 6 5 1 - 1 (alk. paper)
This book is printed on acid-free paper.
j iMUHiiHUHii; mnmii j
! FT IINlVFSKmiRE
I 10 SEP. 1990
jlAU5ANNE/Dori$"!
Contents
Translator's Introduction vii

Preface xvii
1 Sociological Abstinence 1
2 Causes and Responsibilities? 8
3 Complexity and Evolution 11
4 Resonance 15
5 The Observation of Observation 22
6 Communication as a Social Operation 28
7 Ecological Knowledge and Social Communication 32
8 Binary Coding 36
9 Codes, Criteria, Programs 44
10 Economy 51
11 Law 63
12 Science 76
13 Politics 84
14 Religion 94
15 Education 100
16 Functional Differentiation 106
17 Restriction and Amplification
Too Little and Too Much Resonance 115
VI Contents
18 Representation and Self-observation

The 'New Social Movements' 121
19 Anxiety, Morality and Theory 127
20 Toward a Rationality of Ecological Communication 133
21 Environmental Ethics 139
Glossary 143
Notes 147
Index 181
\
Translator's Introduction
Ecological Communication enjoys a unique position in the

extensive writings of Niklas Luhmann. As a late work it embodies
the most recent developments in a thought that has undergone
important changes. It is relatively short but also includes every
aspect of Luhmann's complex theoretical position, and it provides
the theoretical framework within which concrete social themes
and problems are addressed and handled. For these reasons
Ecological Communication is an excellent introduction to the
current position of Luhmann's thought and an indication of its
direction. But at the same time the aspects that make it such an
attractive work also make it difficult to master.
Ecological Communication is very compact and demands an
extensive background knowledge from its readers. The English-
speaking audience is at an immediate disadvantage here because
only a small fraction of Luhmann's writings are available in
translation. (One German reviewer, for instance, estimates that
there are at least 6,000 pages of Luhmann's work in print in
German today.) But even for the German audience what was
required for a complete presentation of his position was not
available until the publication of his most comprehensive work,
Soziale Systeme (1984). 1
In an effort to make Luhmann's position more accessible I
would like to make some general remarks to situate Ecological
Communication within it and, I hope, provide a more adequate
basis for its comprehension.
Luhmann's general position is a unique and extraordinarily
Vlll Translator's Introduction
powerful synthesis of several quite diverse intellectual traditions.

At least four of these can be distinguished: (1) the systems-
theoretical approach to social action found in the writings of
Talcott Parsons; (2) the cybernetic interpretation of the relation-
ship between system and environment; (3) a phenomenological
disclosure of meaning and its importance for the relationship of
the components of social systems; and (4) an autopoietic
understanding of system-organization. Ecological Communication
and indeed all Luhmann's recent works can be understood
only through a clarification of these four separate intellectual
traditions and the way in which he combines them. I would like
to consider these separately to examine what advantage Luhmann
draws from each and how this is exploited in Ecological
Communication.
It is well known, certainly at least since Tbeorie der Gesellschaft
oder Sozialtechnologie: Was leistet die Systemforschung? (1971),2
that Luhmann's plan for a unified social theory is to be understood
as a theory of social systems. Furthermore, this theory makes a
claim to universality - and hence the reproach of ideology that
is so often leveled against it. But in this case Luhmann makes
clear that universality does not represent a claim to the exclusive
possession of truth vis-a-vis other, competing theories. Instead,
he means that this theory includes every aspect of the social
domain within itself and not just certain ones like stratification
and mobility, conflict and conformity, models of interaction, etc.
The influence of Talcott Parsons has been well documented in
this respect and has already been submitted to detailed criticism.
Its most lasting effect has been felt through the interpretation of
social action as interaction. According to Parsons, interaction
requires a plurality of actors whose actions are distinguished as
a unity (system) vis-a-vis an environment through symbolically
mediated structures of expectation.
This cursory presentation of a central aspect of Parsons's social
theory reveals how important it is for Luhmann's general position
because it focuses on the important concepts of complexity and
contingency that are involved in all interaction. Complexity
signifies the 'potential for high degrees of differentiation among
the components which . . . constitute . . . systems',3 while contin-
gency, or rather double contingency, denotes the 'complementarity
of expectations in the process of human interaction'. 4

Both these concepts assume a central place in Luhmann's theory
of social systems too. But Luhmann emphasizes the integrative
role of meaning.
The significance of Parsons's systems-theoretical approach for
Ecological Communication is its interpretation of ecological
problems from a general-systems perspective, i.e., in terms of the
relation of a system in this case society - and its environment.
When ecological problems are approached in this way the
concepts and methods of systems theory especially its most
recent developments - can be exploited not only to reveal the
conditions for these problems but also ways in which their
resolution becomes a possibility.
The concept of complexity employed by Parsons as an integral
part of his systems theory of social action is also a corner-stone
of another systems-theoretical discipline. Cybernetics is the latest
development in the history of systems-thinking starting with
Aristotle. It interprets the relation between system and environ-
ment as a difference in degree of complexity. Because a system's
environment includes everything other than the system itself, its
complexity is always much greater than that of the latter. This
means that systems are constantly confronted with new and
different environmental states. To deal with these they have to
bring their own complexity into a relation of correspondence
with that of their environments. Systems do this through
establishing system structures that reduce the complexity of their
environments and thereby obviate point-for-point correlations
between their own changes and changes in their environments.
The ecological significance of complexity appears as the problem
of a societal environment that can always change in more ways
than society itself. The latter, however, still has to react to these
changes. In Ecological Communication Luhmann uses the concept
of 'resonance' to designate this system/environment interplay. But
while complexity - environmental complexity is always a
problem for a system, at the same time it is the key to its solution
through the increase of the system's own complexity. 1
Complexity, which is both the central problem in a~cybernetic
systems-theory as well as the key to its solution, means that
system components are related, connected together. Social and
Vlll
psychical systems are unique because this connection is

accomplished through meaning. Now the concept of meaning is
nothing new to the investigation of the social domain. Parsons
for instance, following Weber, interpreted meaning as a subjective
property of actions in order to overcome a purely behavioristic
standpoint. Luhmann, however, understands it as a determinate
strategy amongst alternative possibilities. In other words, he no
longer understands meaning substantively as a property of system
components but functionally as the constituting and integrative
relations among them. This is the point at which Luhmann resorts
to phenomenological descriptions because, according to him,
only the latter can reveal meaning.^ Only a phenomenological
description of experience can reveal how the elements of social
systems refer to - are connected with - others, especially in the
past and future. It also reveals how the others are inherently
contingent, i.e. not necessary. As with Parsons, social action is
possible only when the meaningful perspectives of different actors
can be brought into agreement. System structures accomplish this
by reducing the complexity inherent in all social action.
The significance of meaning for Ecological Communication is
the essential recursiveness (self-referentiality) of system components
- in this case communications. Phenomenological investigations-
reveal that meanings constitute themselves self-referentially,
i.e., they refer exclusively to other meanings. They thereby
organize what Luhmann following Husserl calls 'horizons'
of further communicative alternatives. Communication can
communicate only what is meaningful because for Luhmann it
is not a 'transfer' of information but instead the common
actualization of meaning.
The concept of meaning leads directly into the fourth and final
aspect of Luhmann's complicated position: the concept of
autopoiesis. This concept was first introduced by the theoretical
biologists Humberto Maturana and Francisco Varela in De
Maquinas y Seres Vivos in 1972 (Autopoiesis and Cognition,
1980). They used it to capture the unique capacity of living
systems to maintain their autonomy and unity through their very
own operations. Maturana and Varela discovered that this is
possible only when the operations proper to a system (its complex
of component-producing processes) and its components (elements)
themselves are so related that they constantly (re)produce each

other. They expressed this relation as the organizational closure
of the system. By distinguishing systematically between system
organization and system structure, a model was established for
a dynamic system that was both closed (organizationally) and
(structurally) open at the same time.
Despite the initial difficulties encountered by Maturana and
Varela in extending this model beyond the biological domain,
Luhmann exploits both its analytic and explanatory power by
applying it to social systems where it also appears under the
name of self-reference. At the basic level this means that social
systems constitute themselves self-referentially. Everything that
functions as an element in the system is itself a product of
the system. Luhmann reveals how this occurs by carefully
distinguishing between (system) element and (system) relation,
i.e., structures and processes. He shows that the concept of
autopoiesis can be extended to the social domain only when the
elements of social systems are conceived as communicative acts
(events) and not as persons, roles, subjects, individuals, etc.
The importance and uniqueness of the concept of autopoiesis
for Luhmann resides in its ability to provide the theoretical
framework within which social systems can be differentiated as
constituting themselves self-referentially through the development
of their own separate symbolically generalized media of communi-
cation, for example, money/economy, power/politics, love/family
and truth/science. Yet, unlike the case of earlier systems-theories
where the system was either open or closed but not both,
autopoiesis accounts at the same time for the non-reductive
relation between the system and its necessary infrastructure
(environment). In this sense, whatever is not communication is
environment for society; even when this includes the conscious-
nesses psychical systems and lived bodies biological systems
- of human beings. After all, communication needs both psychical
and biological systems in order to occur.
Not until the concepts of autonomy (closure) and interpen-
etration (openness) are unified in one theory can any systems
theory as such, and any theory of social systems in particular,
be complete. By doing this the concept of autopoiesis provides
the synthetic unity necessary for the production of a systems
theory of the social domain.

The importance of autopoiesis for Ecological Communication
is the recursive organization of all social systems. When society
is defined as the social system that includes the possibility of
communication as such, this means that it cannot communicate
with its environment and ecological problems can only be handled
from within society itself, i.e., through its subsystems. /
The synthesis of these four diverse intellectual traditions is
certainly the source of the great strength and innovation of
Luhmann's position in general and of Ecological Communication
in particular. But at the same time it proves to be one of the
greatest obstacles to their comprehension because all of these
traditions developed separately and without any direct contact,
for example, between phenomenology and cybernetics on one
hand, or between the systems approach of Talcott Parsons and
autopoiesis on the other. But even between those that enjoy a
kind of conceptual relatedness - like cybernetics and autopoiesis
there is, to a significant degree, perhaps only a one-sided
awareness. Therefore, one cannot assume that a reader of
Ecological Communication - or any of Luhmann's recent works
for that matter - who is familiar with one or even two of these
traditions will also be familiar with the others and therefore able
to appreciate the full import of what Luhmann presents. This
becomes particularly evident in Ecological Communication where
Luhmann applies his theoretical principles to an extraordinarily
wide variety of social topics, ranging from law, politics, religion
and education to science and the economy.
As the most recent addition to Luhmann's general theoretical
position we have to start from the concept of autopoiesis in order
to analyze and explain the central argument and themes of
Ecological Communication because this is the concept that
integrates, synthesizes the theoretical content of the other
traditions. Autopoiesis signifies the closure of a system's organiz-
ation, i.e., the self-reference of the complex of components
and component-producing processes that mutually reproduce
themselves and thereby integrate and unify the system. Organiz-
ational closure does not mean that a system cannot be affected
at all by its environment. But it does mean that, as an autonomous
unity, i.e., organizationally, it can react to its environment only
in accordance with its own mode of operation, the mode of

operation peculiar to it. Because society's specific mode of
operation is communication it cannot react to its environment in
any other way - and since its environment includes everything
that does not operate in a communicative way, society is eo ipso
prevented from communicating with its environment. Instead, it
can only communicate about its environment within itself.
In Ecological Communication the communicative reactions or
disturbances that society's environment produces within society
itself is called 'resonance'.
| Luhmann believes that societal communication affects the very
way in which the possibility of environmental dangers arise. So
the question around which Ecological Communication turns is
not one of how society can manage existing environmental
problems. (Luhmann recognizes that this can be accomplished
through a sufficiently powerful law for policing the environment.)
Instead, he is concerned with how society comes to the very
awareness of environmental dangers as such. How does it put
itself in the position to recognize environmental dangers at all?
After all, 'Ecological dangers may exist or not and no one may
know about them. But the exposure to ecological dangers exists
only when people communicate about the pollution of rivers
and the air and the deforestation of the land.'6 When this
communication occurs, the dangers in the environment can be
addressed only in the ways that society itself has established for
communication. Ecological Communication is concerned with
how these dangers shape up it talks about the 'contours' of
the problem of society's ecological adaption - within society
when communication about them takes place. But this means
that it is concerned only with 'how society, in fact, reacts to
environmental problems and not how it ought to or would have
to in order to improve its relation to the environment'^
Modern society, as opposed to those before it, is differentiated
according to subsystems that concentrate on one specific and
primary function. For this reason they are called 'function systems'
in Ecological Communication. The economy, law, science, politics,
religion and education are examples of these. Even if all the
communication that occurs within society does not have to be
ascribed to one of them, Luhmann argues that the socially most
consequential communication does.

A particularly thorny problem, however, infects autopoietic
systems. Self-reference means that their operations are applied to
themselves. Ever since classical antiquity it has been known that
this leads to paradox. (The classical example is the famous 'liar
paradox' of Eubulides.) In recent times attempts to solve
paradoxes of self-reference have produced the theory of types
(Whitehead and Russell) and meta-languages (Tarski). These
solutions involved the introduction of some form of difference
for example, the difference of types or linguistic levels which,
in Luhmann's terminology, 'interrupts' or 'unfolds' the (unity of
the) self-reference. Function systems employ binary coding and
programming to introduce this difference. This means that
the unity (of self-reference, J.B.) that would be unacceptable in

the form of a tautology (e.g., legal is legal) or a paradox (one
does not have the legal right to maintain their legal right) is
replaced by a difference (e.g., the difference of legal and illegal).
Then the system can proceed according to this difference, oscillate
within it and develop programs to regulate the ascription of the
operations of the code's positions and counter-positions without
raising the question of the code's unity.
Function systems structure their communication through binary

codes that divide the world into two values (for example, true/false,
legal/illegal, power/lack of power, immanence/transcendence,
possessing/not possessing). The resonance created within society
by its environment is channelled into one of these function
systems and treated there effectively in accordance with one code
or another. This complicates the ecological problem because from
now on two system-references have to be kept separate: one
between society as a whole and its environment (society's external
boundary) and the other that exists within society itself for its
particular function systems (society's internal boundaries). The
problem is that environmental changes produce too little and too
much resonance within society at the same time.
Coding and the programs that accompany it (theories in science,
\ laws in the legal system, investments in the economy, party-
political alignments in politics, etc.) produce a sharp reduction
in what is information, i.e., differences that make a difference,
for society. They screen it off from its environment organization-

ally and obviate point-for-point correlations between system
and environment. ^Resonance within society, then, is always
improbable, i.e., occurs only in exceptional cases. According to
Luhmann, this implies that society produces too little resonance
vis-a-vis environmental dangers.! On the other hand, the situation
at the internal boundaries of society, i.e., where communication
actually takes place, is quite differentljNecessary communicative
interdependencies among function systems can literally produce
too much resonance within society even if environmental disturb-
ances produce too little resonance.
What Luhmann has in mind here can be illustrated, for instance,
by the relation between the economy and science or between
politics and law. Furthermore, he also wants to indicate that
insignificant changes in one of these function systems can
trigger an 'effect-explosion' in another or others. For example,
'Theoretically insignificant scientific discoveries can have agoniz-
ing medical results', and 'Payments of money to a politician that
play no role in the economic process - measured by the billions
of dollars that are transacted daily - can become a political
scandal'. Again, the law can place 'the pharmaceutical industry
and physicians under the threat of liability to supply information
and to establish precautionary standards (that) can have medical
as well as economic consequences that are entirely unrelated to
what is legally important and might not even compromise part
of the legal decision itself'.
The specific codes and programs of the particular function
systems guarantee the disproportion in reactions to environmental
disturbances among them. It also means that no one of the
function systems can be substituted for another. Thus in the
functionally differentiated modern society there is no subsystem
(function system) that represents the whole of society from within
as was the case in earlier societies - differentiated through
stratification when one level of society was viewed as the top
or center of society.
The functional differentiation that produces a centerless society
provides the basis for objections raised by Ecological Communi-
cation against what it calls the 'new movements'. These use
strategies that deal with environmental dangers by appealing to
a new environmental moralityîAccordingly, right is on the side

of those who are against the self-destruction of society. A new
environmental ethics for society is all that is necessary, and moral
zeal - demonstrable through anxiety - can compensate for any
theoretical deficits in this ethics.
But without a consideration of the constraints placed on
ecological communication (including any environmental ethics)
by the functional differentiation of society, this discussion can
fall very easily into a rhetoric of anxiety.
Anxiety is a particularly attractive theme for ecological
communication because it can always be used as the basis for
moral justification when all else fails. But not much is gained
through the communication (rhetoric) of anxiety except perhaps
more communication (rhetoric) of anxiety, and although it can
easily be used to block society's incursions upon its environment,
it has to pay for this with unforeseen reactions on society that
only produce more anxiety. If, then, a specific function is to be
assigned to environmental ethics within the context of ecological
communication, Luhmann says this should be to proceed
cautiously in dealing with the morality of ecological problems.
John Bednarz Jr.

Preface
On 15 May 1985, at the invitation of the Rhenish-Westfallian

Academy of Science, I addressed their yearly assembly on the
theme, [^Can Modern Society Adjust Itself to the Exposure to \
Ecological Dangers ?'7\The restricted amount of time afforded by j
the address, however, did not allow for the presentation of what
could be said on the matter in its entirety. Above all, it was
impossible to deal with the very similar modes of reaction of the
various particular function systems despite all their differences.
The main argument of the address, namely that modern society
creates too little as well as too much resonance because of its
structural differentiation into different function systems, was
presented only in its main outline. Only from this insight does
it follow that the solution to this problem can be found in new
ideas about values, a new morality or an academic elaboration
of an environmental ethics."1
The work presented here supplements the argument of my
address and completes it for the most important function systems
of modern society, even if only in very broad outlines. What
emerges is a very similar picture of basic structure for the different
functions, binary codes and programs of 'correct' experience and
action.! This justifies ascribing ecological problems to society and j
not only to the failures of politics and the economy or to an I
insufficient feeling of responsibility^ I
Perhaps in this way a theoretically guided comparison can
clarify how social theory is challenged by ecological discussion
that has recently become topical - and how little it had to offer
formerly.
XV111 Preface
In many respects the analyses presented here diverge from the

premisses that have been introduced naively in the ecological
literature and used without any further justification. This is true
for basic systems-theoretical questions as well as for numerous
details. Indeed, the ecological literature itself is a product of
social communication, i.e., a part of the object that we are
investigating here, including our own investigations. Indifference
in the choice of words and absence of interest in significant
theoretical decisions are two of the most noticeable characteristics
of this literature - as if caring about the environment could
justify carelessness in the talk about it. Moreover, it turns out
that the literature that emerges with the claim to scientific
precision is produced mainly by those disciplines that, at the
same time, fulfill the reflective functions in those function systems
from which they originate. This means that jurists must concern
themselves with expanding the categories that deal with legal
cases and economists with expanding those models with which
economic data are observed and positive or negative economic
growth is revealed. Of course, I recognize that all this has its
significance.! But the problem for me lies much deeper in the
differentiation of the function systems themselves.^
Investigations that are inspired theoretically can always be
accused of a lack of 'practical reference'. They do not provide
prescriptions for others to use. They observe practice and
occasionally ask what is to be gained by making such a hasty
use of incomplete ideas. This does not exclude the possibility
that serviceable results can be attained in this way. But then the
significance of the theory will always remain that a more
controlled method of creating ideas can increase the probability
of more serviceable results - above all, that it can reduce the
probability of creating useless excitement.
Niklas Luhmann
1
Sociological Abstinence
Compared to the history of reflection on humanity and society

this theme - ecology - is not very old. \Only in the last twenty \
years has one seen a rapidly increasing discussion of the ecological
conditions of social life and the connection between the social
system and its environment. (Contemporary society feels itself j
affected in many different ways by the changes that it has j
produced in its own environmen\This is clearly shown by a '
number of these: the increasingly rapid consumption of non-
replaceable resources and (even if this would prove beneficial) the
increasing dependence on self-produced substitutes, a reduction in
the variety of species forming the basis of further biological
evolution, the ever-possible development of uncontrollable viruses
resistant to medicine, the familiar problem of environmental
pollution and not least of all over-population. Today these are
all themes for social communication. Society has thus become
alarmed as never before1 without possessing, however, the
cognitive means for predicting and directing action because it not
only changes its environment but also undermines the conditions
for its own continued existence. This is by no means a new
problem. It appeared in earlier stages of social development too. 2
But only today has it reached an intensity that obtrudes as a
'noise' distorting human communication that can no longer be
ignored.
As far as sociology is concerned, this discussion began like
so many others, unexpectedly - and caught it, as it were,
unprepared theoretically. Originally, sociology had been con-
2 Sociological Abstinence
cerned with the internal aspects of society. It entangled itself in

ideologies of the correct social order and then tried to extricate
itself from them. All this was done under the assumption that
its theme was society or its parts. The history of the foundations
of this discipline had already predisposed it in this direction.
Nature, on the other hand, could and indeed had to be left to
the natural sciences. What the new discipline called sociology
could discover and claim as its own field of study was either
society or, if this concept was unsatisfactory, social facts, for
example, faits sociaux in Durkheim's sense, or social forms and
relations in the sense of Simmel and von Wiese, or social action
in Weber's sense. Thus the delimitation of the discipline had to
be interpreted as a demarcation of a section of reality.
But in addition to 'grand theory', research in the domain of
the most diverse social problems is also attuned to the social
origins of these problems. This is precisely what forms the basis
of the researcher's hopes of being able to contribute something
to a better solution of the problem. The problematic is reduced
to structures of the social system or its subsystems, and if these
cannot be changed then at least, one can blame the circumstances.
The external sources of the problems are not even considered.3
And although every problem of the system is ultimately reducible
to the difference between system and environment, this is not
even considered.
Even for the earlier theory of societas civilis this was no
different, and the same is true for practical philosophy: what is
social was viewed as civitas, as communitas perfecta or as political
society, even if this included all of humanity. According to the
Stoic as well as the Christian theory, non-human nature was to
be used by everyonc^JDominium terrae thereby became a concept j
by which the sacralization of all of nature was prevented and
the specification of what was religious was secured. Nature in
this sense, so ecologically important today, was de-sacralized
nature^jThe ever present counter-opinions were never strong
enough to present the developing natural sciences with a problem.
Then in the eighteenth century the problem experienced a
dramatic reversal. The counter-concept (which, as is so often the
case, suggests the real interest) was changed. Civilization took
the place of the sacred (whose specifically monotheistic version
was retained) as the counter-concept to nature.4 Thereby nature

became, on one hand, an irretrievably lost history and, on the
other, society's field of research.
But even this version was not enough to determine the difference
between nature and civilization. It does, however, offer the first
chance at an awareness of the environment (for example, as the
consequence of the ancient doctrine that God is to be worshipped
in his creations). The eighteenth century discovered the meaning
of milieu,5 i.e., of being situated concretely, for example, as the
connection between climate and culture. Stimulated by progress in
agricultural technology, the early French economists (physiocrats)
saw property as a legal institution that is both economically and
ecologically ideal because it guarantees the proper treatment of
natural resources while it reconciles them with human interests.
It is noteworthy that at that time the internalization of the
consequences of actions and their inclusion in rational calculation
were viewed as a function of property. Today the converse is the
case: the consequences of actions are discussed in terms of
externalization and property is criticized for lacking in responsi-
bility for these consequences.
At first nothing resulted from all this. The French Revolution
led to an ideologizing of social debates linked to social position
and political goals, and the descriptions of social relations
occurred entirely within society. This is most clearly visible in
the way in which Darwin was carried over into the social sciences.
Instead of accepting the idea that the environment selectively
decides how society can develop, an ideologically tainted Social
Darwinism came into being that promised individuals, economies
and nations the right to success through the survival of the fittest.
But after a few years this too became bogged down in the mire
of a new social morality, and even today the theory of evolution
in the social sciences has not freed itself completely from this
disaster.6
Even where sociology presented itself as opposition or as
'critical theory' all that it considered was society and humane
principles that did not correspond to the society, or at least not
at that time. This found expression as insufficient freedom,
equality, justice or reason in any event, all bourgeois themes.
The part that sociology played within this social discussion was
the self-critique of society vis-a-vis determinate ideals, not

frustration regarding uncertain hopes and fears. But it was simply
too easy to reject this critique because ideals have a fatal tendency
to transform themselves into illusions. The theoretical background
for this discussion has long since disappeared even though the
'simultaneity of the non-simultaneous' still had to be reckoned
with for a long time. In this state of alarm the only question
must be how justified are specific hopes or fears? Or, from the
perspective of a disinterested observer, which factors determine
the readiness to accept risks and how are they distributed in
society?
Totally absorbed in its own object, sociology did not even
notice that a reorientation had already started among the natural
sciences, begun by the law of entropy. If this law that declares
the tendency to the loss of heat and organization is valid then it
becomes even more important to explain why the natural order
does not seem to obey it and evolves in opposition to it. The
answer lies in the capacity of thermodynamically open systems
those related to their environments through inputs and outputs
to enter into relations of exchange, i.e., environmental
dependency, and nevertheless to guarantee their autonomy
through structural regulation. Ludwig von Bertalanffy appropri-
ated this idea and used it as the basis for what today is called
'general systems theory'.7
It would be unfair, however, to say that sociology did not take
account of this at all because there are some programmatic
similarities.8 For example, research in the sociology of organiz-
ations, emphasizing the environmental reference of organizations,
has been successful.9 But here the environment always means
something internal to the society, for example, markets or
technological innovations, in other words only society itself. 10
This preoccupation with society itself can be avoided only
through a change in the theoretical focus of the paradigm.11 Such
a manoeuvre, however, has consequences that reach all the way
into the ramification of sociological thought. 12 This means that
radical incisions have to be made, and only after such an operation
if it is not refused to begin with does one learn to proceed
again slowly.
The surprising appearance of a new ecological consciousness
has left little time for theoretical consideration. Initially, therefore, !

the new theme was considered within the context of the old ii
theory. Accordingly, if society endangers itself through its effects |
on the environment then it has to suffer the consequences. The j
guilty should be found out, restrained and, if necessary, opposed )
and punished'.! Moral right, then, is on the side of those who
intervene against the self-destruction of society. In this way the
theoretical discussion surreptitiously becomes a moral question
and any of its possible theoretical shortcomings are offset by
moral zeal. In other words, the intention to demonstrate good
intentions determines the formulation of the problem.[So, by j
accident, as it were, a new environmental ethics enters the f
discussion without ever analysing the all-important system I
structures, j il
Whoever proposes a new ethics, brings the question of blame
into historical view at the same time. It has been advocated, for
instance, that the Christian West was disposed to deal with nature
in a crude and insensitive way, if not simply to exploit it, 13 while
on the other hand it was also argued that Christians always loved
and respected animals and paid homage to the Creator through
His creation. 14 When the question is put in such a simple and
naive way both of these arguments are valid. This historical
perspective serves only to provide contrast and does not concern
itself at all with the actual course of events. It merely helps to
bring the new ethics into view without raising the difficult
question of whether and how it is possible as such. 15
Despite this, it has become obvious that as scientific research
progressed respect for 'natural balances' increased, whether this
was in ecological relations, foreign cultures or even today in
developing countries and their traditions. But at the same time,
one's own society was exposed to an incisive critique that was
replete with demands for intervention, as if it was not a system
at all.16 Obviously this reveals a negative ethnocentrism, and it
is possible that a significant aversion to 'systems theory' has had
something to do with the critical restraint this theory has directed
against its own society.
At the very least this summary discussion lacks an understanding
of the theoretical structure of the ecological question, above all
of its fundamental paradox that it has to treat all facts in terms
of unity and difference, i.e., in terms of the unity of the ecological

interconnection and the difference of system and environment
that breaks this interconnection down. As far as the ecological
question is concerned, the theme becomes the unity of the
difference of system and environment, not the unity of an
encompassing system.1"
Therefore the systems-theoretical difference of system and
environment formulates the radical change in world-view. This
is where the break with tradition is to be found, not in the
question of a crude and insensitive exploitation of nature. Indeed,
historical investigations of the concepts of periechon, continens,
ambiens, ambiente and medium can show that what is today
called environment was viewed by the Greek and even the
medieval tradition as an encompassing body, if not as a living
cosmos that assigned the proper place to everything in it. 18 These
traditions had in mind the relation of a containment of little
bodies within a larger one. Delimitation was not viewed as the
restriction of possibilities and freedom but instead as the bestowal
of form, support and protection. This view was reversed only by
a theoretical turn that began in the nineteenth century when the
terms ' U m w e l f and 'environment' were invented and which
has reached its culmination today: systems define their own
boundaries. They differentiate themselves and thereby constitute
the environment as whatever lies outside the boundary. In this
sense, then, the environment is not a system of its own, not even
a unified effect. As the totality of external circumstances, it is
whatever restricts the randomness of the morphogenesis of the
system and exposes it to evolutionary selection. The 'unity' of
the environment is nothing more than a correlate of the unity of
the system since everything that is a unity for the system is
defined by it as a unity.
The consequences of this interpretation for a theory of the
system of society (and indeed for a system of society that
communicates about ecological questions) can be reduced to two
points:
1 The theory must change its direction from the unity of the
social whole as a smaller unity within a larger one (the world)
to the difference of the system of society and environment,
i.e., from unity to difference as the theoretical point of

departure. More exactly, the theme of sociological investigation
is not the system of society, but instead the unity of the
difference of the system of society and its environment. In
other words, the theme is the world as a whole, seen through
the system reference of the system of society, i.e., with the
help of distinctions by which the system of society differentiates
itself from an environment.19 After all, difference is not only
a means of separating but also, and above all, a means of
reflecting the system by distinguishing it.
2 The idea of system elements must be changed from substances

(individuals) to self-referential operations that can be produced
only within the system and with the help of a network of the
same operations (autopoiesis). For social systems in general
and the system of society in particular the operation of (self-
referential) communication seems to be the most appropriate
candidate.
If these two points are accepted then 'society' signifies the all-
encompassing social system of mutually referring communications.
It originates through communicative acts alone and differentiates
itself from an environment of other kinds of systems through the
continual reproduction of communication by communication. In
this way complexity is constituted through evolution.
The considerations that follow presuppose this theory - not in
order to provide a solution to the problem of the ecological
adaption of the system of society, but instead to see what contours
the problem takes on when it is formulated with the help of this
theory.
25
Causes and Responsibilities?
Once one acknowledges the phenomenon of evolved complexity

the focal point of the ecological problematic changes. The
customary way of treating ecological problems begins from causes
within society and then seeks responsibility for their effects. Thus
it follows the normal temporal pattern and argues that the results
will not occur if not preceded by the causes. Accordingly, the
problems are best eliminated at their source, for example, when
a chemical plant disposes of poisonous waste at a garbage dump
or into a river with the consequence that fish are îllfD ,QT that
the water-supply becomes contaminated. An enl^rceable"* legal
code suffices to handle such problems. But both the problem
typology as well as a systems-theoretical analysis require a change
of approach: a reconstruction of the problem from a systems
perspective, one that is sensitive to the effects of ecological
changes. Intercepting the causes is one of the possible ways of
taking care of their effects, but only one among many. The
problem of reaction to effects and the possible (almost limitless)
causes and effects of such reactions remains. In other words, the
'tragedy' of decisions is that the affected system is also the cause
of its own damage. But this is still not a formula for the solution
of problems.1
Politics and jurisprudence, for instance, ûse the 'causer-
principle' (Verursacherprinzip) to alcribe costs and to assign *
responsibility,2 and it is clear that this produces a problem of
selecting the causer. Usually, this problem is hanclfed reflexively
by appealing to the purpose of the selection itself.3 The covert
9 Causes and Responsibilities?
-'/' --S f s ' v ,
significance of the causer-principle, then, is not a causal statement

but, as so often, a statement indicating a difference: alternatives
(for example, subventions) are rejected because of the expense to
the general public.
Science has long since left this practically significant stage of
analysis behind. In the age of systems-theoretical analyses causal
interconnections have been viewed as extremely complex and, in
principle, opaque - unless their determination is simplified
through a more or less arbitrary attribution of effects to causes.
The last three decades' research into attribution shows that the
real problem resides in the attribution of habits and procedures
that illuminate and give importance to a selection of the many
causes and effects.4 More exactly, the determination of causes,
responsibility and guilt helps to identify non-causes (Nichtur-
sachen) and to determine innocence and the absence of responsi-
bility too. If the producers fall on the^ one side then consumers
must fall on the other. In this way the attribution procedure
shows its real importance in providing exculpation.
All this is accepted today and does not need to be proved. The
theory of self-referential systems alone, however, has realized that
the classical instruments of the acquisition of knowledge, namely
deduction (logic) and causality (experience), are merely forms of
simplifying the observation of observations. For social systems
this means forms of simplifying self-observation.5 Methodolog-
ically, this means that the point of departure has to be the
observation of self-observing systems and not the assumed
ontologic of causality. In other words, one cannot avoid a
decision about what counts as a cause and who is to be
held responsible. It also ,mqan;s/that morality and politics are
overburdened by the unavoidably of this decision. The question
then is how can this decision present itself so that the impression
arises that it has not?
Radical theoretical positions of this kind lie far outside what
social communication and ordinary consciousness accept today.
Their consequences would require a rethinking whose results are
unjfereseSa'ble. In any event,- a period of slow and tedious
development seems inescapable. But at the same time, the
corresponding circumstances are evident and precisely in
ecological discussion itself. Attribution and assigning responsibility
10 Causes and Responsibilities?
have consequences themselves. Political coalitions, for instance,

can come apirf and economic enterprises depending on them can
fail. Theories and calculations which subsequently prove false
can be used to justify decisions, a discovery that itself can trigger
new consequences. All this may be very clear and obvious. But
at present the problem is how to promote it publicly and
legitimize it? Whoever observes this telescoping (Engfiihrung) of
observations will coipe to ,the conclusion that 'tragic' decisions
of this kind have to conceal their own contingencies so that they
do not have to be revealed as decisions, or at least not in certain
respects.
It is well known that Walter Benjamin thought that the
difference between legislation and adjudication was used for this
purpose of concealment, especially in reference to his concept of
violence.6 This holds for politics as well as law. In the economic
system the same function seems to be fulfilled by the difference
between the determination of available quantities and the decisions
of distribution under the condition of scarcity.7 In both cases,
however, it is shown that every individual decision affects the
difference itself.^Nevertheless, in both cases it is also true that
V -i'v y % a V'.C . v v Tv~. i
this cannot be extolled as responsibility. The difference must be

presupposed in order to determine where decisions are to be
made. In effect, it replaces its own arbitrariness and resolves a
paradox of self-reference. Only then can one-assign responsibility.
At this point we are already in the midst*of system analyses.
In this chapter we have been concerned only with establishing
premisses. In the analyses that follow we will ignore the question
of guilt. Of course, this does not mean that its clarification from
the point of view of political representability or of the legal
propriety of standards is unimportant. On the level of our
analyses this question would lead to the discovery that society
itself is guilty - and we know this already.
\
28
Complexity and Evolution
'We really can change the whole thing', is a slogan that could
still be heard even quite recently. Courage is all that is needed -
and cybernetic guidance! Complexity has simply been exploited
insufficiently until today causing all kinds of mistakes and
problems for the system's output. Instead, the system has to use
variety (i.e., number of possible states) to control variety and in
this way, acquire the 'requisite variety for running the world'. 1
This kind of optimism seems to have passed. It underestimated
the much discussed problematic of structured complexity. Above
all, it did not understand that the concept of complexity itself
designates a unity that acquires meaning only in reference to
difference, indeed in reference to the difference of system and
environment.2
It is not saying much to state that the world or a system is
'complex'. From this point of view everything determinate results
from the reduction of complexity. Instead, one could simply say
that everything occurs only in the world. But not much is gained
in this way. Statements concerned with complexity become
productive only when they are turned from unity to difference.
The distinction of system and environment can be used to do
this. It enables one to make the statement with which we will
introduce the following discussion: that for any system the
environment is always more complex than the system itself.3 No
system can maintain itself by means of a point-for-point
correlation with its environment, i.e., can summon enough
'requisite variety' to match its environment. So each one has to
12 Complexity and Evolution
reduce environmental complexity - primarily by restricting the

environment itself and perceiving it in a categorically preformed
way. On the other hand, the difference of system and environment
is a prerequisite for the reduction of complexity because reduction
can be performed only within the system, both for the system
itself and its environment.
To make this matter even clearer, the question can be put as
follows: how can a restrictedly complex system exist in a much
more complex environment and reproduce itself? In so far as a
genetic explanation is desired, the question could be handed over
to the theory of evolution. Evolutionary selection can be used to
explain how system structures, which maintain themselves under
the pressure of complexity, develop.4 But the theory of evolution
itself has so far not provided a satisfactory explanation of this
because there are obviously several, if not many, solutions to this
problem and because the choice among them cannot be explained
satisfactorily as environmental selection or as an adaptive
performance of the system.
One possibility could be greater indifference and insulation of
the system, i.e., less environmental dependence and sensitivity by
restricting causal interdependencies. It is obvious, however, that
macrochemical as well as organic and sociocultural evolution
transcend this possibility,and it is difficult to understand how
they can be made to do so by Environmental complexity. Thus
the question arises of what other forms can replace indifference
and insulation as functionally equivalent. Again, the answer is
greater system complexity.5
'Greater system complexity' is not a simple quality and
therefore, the 'increase' cannot be understood in terms of one
dimension alone.6 This means that talk of 'more' or 'less'
complexity is necessarily vague. Despite this, universally valid
statements can still be formulated. Accordingly, systems with
greater complexity are generally capable of entertaining more
and different kinds of relations with their environments (for
example, of separating inputs and outputs) and thus of reacting
to an environment with greater complexity. At the same time,
they have to select every individual determination internally with
greater exactness. So their structures and elements become
increasingly contingent. This leads to another question: which
structures can meet such demands?

Evolution does not merely mean the selection, by a particular
environment, of the systems that are capable of survival or
increasing the adaption- and survival-capacity of systems to a
particular environment.7 This does not explain why the environ-
ment continually produces stimuli for variation and yet allows a
multitude of systems to exist completely unchanged. The theory
of evolution must therefore include systems theory in the
explanation. Self-referentially autopoietic systems are endogen-
ously restless and constantly reproductive. They develop structures
of their own for the continuation of their autopoiesis. In this
way the environment remains as the condition of their possibility
and as a constraint. The system is both supported and disturbed
by its environment. But it is not forced to adapt by the
environment nor allowed to reproduce only through the best
possible adaption. Even this is a result of evolution and at the
same time a condition of further evolution.
Only when this reformulation of the theory of evolution is
accepted can one use it to explain why, ecologically, the system
of society is not necessarily directed toward adaptation and can
even place itself in jeopardy. The system forms its own structures
in reaction to irritation from the environment in order to continue
the autopoietic process, or it simply ceases to exist. Thereby, it
acquires the often unrealistic ide'a that the environment adapts
to it and not vice versa. Very complex systems can develop in
this way if forms of organization can be found that are compatible
with greater complexity, i.e., make corresponding reduction-
performances possible. The dynamics of complex autopoietic
systems itself forms a recursively closed complex of operations, i.e.,
one that is geared toward self-reproduction and the continuation of
its own autopoiesis. At the same time, the system becomes
increasingly open, i.e., sensible to changing environmental con-
ditions. All this can proceed along two lines of development:8
the evolution of systems with greater but more reducible
(more operational) complexity of their own and the increasing
temporalization of autopoiesis. In the latter case autopoiesis no
longer has to deal exclusively with the preservation of the existing
system-state or with the continual replacement of elements
that have fallen out (for example, the replication of cells or
macromolecules within cells) but eventually creates systems from

events whose continual passing is the necessary cause of the
autopoiesis of the system.
Thus the exposure to ecological self-endangerment remains
within the context of the possibilities of evolution. Threatening
situations occur not only because a higher degree of specialization
in answer to environmental changes reveals itself as misguided.9
The possibility also exists that systems act on their environment
in such a way that they cannot exist in this environment later
on. The primary goal of autopoietic systems is the continuation
of autopoiesis without any concern for the environment. Typically,
the next step in the process is more important for them than the
concern for the future, which indeed is unattainable if autopoiesis
is not continued. Viewed from a long-term perspective, evolution
is concerned about reaching 'ecological balances'. But this merely
means that systems pursuing a trend toward exposure to ecological
self-endangerment are eliminated.
If this evaluation of the evolution of social complexity and
ecological problems is correct, then the question of the 'domination
of nature' has to be reformulated. It is no longer an issue of a
greater or lesser technological control over nature or even of
sacred or ethical road-blocks. Nor is it a matter of the protection
of nature or of a new taboo. To the extent that technological
intervention changes nature and problems result from this for
society, greater rather than less competence for intervention has
to be developed, but practiced according to criteria which include
reaction on itself. The problem does not lie in causality but in
the criteria for selection. The question that follows from this is
twofold: (1) is there enough technological competence for selective
behavior, i.e., does it give us enough freedom vis-a-vis nature?
(2) is there enough social, i.e., communicative, competence to be
able to carry out the selection operatively?
\
32
Resonance
Concepts like complexity, reduction, self-reference, autopoiesis

and recursively closed reproduction with environmentally open
irritability raise complicated theoretical questions that cannot be
pursued in all their ramifications in what follows. So we will
simplify the presentation by describing the relation between
system and environment with the concept of resonance. We will
also assume that modern society is a system with such a high
degree of complexity that it is impossible to describe it like a
factory, i.e., in terms of the transformation of inputs into outputs.
Instead, the interconnection of system and environment is
produced through the closing-off of the system's self-reproduction
from the environment by means of internally circular structures.
Only in exceptional cases (i.e., on different levels of reality,
irritated by environmental factors), can it start reverberating, can
it be set in motion. This is the case we designate as resonance.
One can imagine a dictionary that would define nearly all the
concepts that it uses by referring to other definitions and would
allow reference to undefined concepts only in exceptional cases.
An editorial committee could then be formed which would
supervise whether language changes the meaning of those
undefined concepts or, through the formation of new ones,
disturbs the closure of the lexical universe without determining
how changes in the entries are to be handled when this disturbance
occurs. The richer the dictionary, the more it is kept going by
the development of language, i.e., the more resonance it will be
able to produce.
16 Resonance
Physics can also be called on to help us. A differentiated system

can be made to resonate only on the basis of its own frequencies.
In the biological theory of living systems, 'coupling' is used to
indicate that there never are point-for-point correlations between
the system and environment. Instead, the system uses its
boundaries to screen itself off from environmental influences and
produces only very selective interconnections.1 If this selectivity
of resonance and coupling did not exist the system would not
be able to distinguish itself from the environment. It would not
exist as a system.
The same is true for the process of communication in the social
system.We can formulate the question of the ecological basis of
and danger to social life much more exactly if we look for the
conditions under which the states and changes in the social
environment find resonance within society. This is by no means
something that is more or less self-evident. On the contrary, it
is improbable according to systems theory. From the evolutionary
point of view one can even say that sociocultural evolution is
based on the premiss that society does not have to react to its
environment and that it would not have taken us where it has
if it proceeded differently. Agriculture begins with the destruction
of everything that had grown there before.
We find the problematic of a purely selective contact with the
environment and the use of boundaries for screening-off on the
level of individual system operations too. Society is a system, sit
venio verba, uncommonly rich in frequencies. Everything that
can be formulated linguistically can be communicated about. But
we remain bound to language (just like we are bound to the
narrow spectrum of what we can see and hear), and what is
more important and decisive, speech and writing have to be
ordered sequentially. Everything cannot be said all at once nor
can all statements be connected with all others. The general
structure of language (its vocabulary, grammar and the way it
uses negation) makes selections necessary. This means that all
selections themselves have to be ordered sequentially, i.e., appear
in a context of succession in which one phrase makes another
intelligible but never the whole. Even if no boundaries of the
social system were given, i.e., even if we could start from the
very beginning with society, communication's mode of operation
Resonance 17
would establish boundaries simply through its coming into being

and continuation. If communication takes place then this eo ipso
differentiates a social system whether anyone wants it and
approves of it or not.
These constraints on the social system's capacity for resonance
are attuned to the mode of information processing that society
and psychical systems apply in common: to the characteristics of
meaning. The possibilities of a meaningful grasp of the world
are themselves attuned to and then require the necessity of
a purely momentary grasp of the world at any time. Only very
little can form the actual focus of attention or be treated as an
actual theme of communication. Everything else, including the
world as a whole, is associated with this only by means of
references, i.e., accessible only sequentially and selectively. Only
one of these possibilities can be pursued at any time, and
every advance creates more possibilities than can be handled
subsequently.2 This is what Husserl meant when he described the
world as the 'horizon' of actual intentions. It is actual as a
horizon, never as a universitas rerum. One can see in this a
formula, as it were, for the insolubility of ecological problems,
even though, at the same time, it is known that every reference
leads to something determinate or determinable3 - that there are
no paradoxes.
In other words, meaning is a representation of world complexity
that is actualizable at any moment. The discrepancy between the
complexity of the actual world and consciousness's capacity for
apprehension or communication can be bridged only when the
scope of the actual intention is restricted and all else is rendered
potential, i.e., reduced to the status of mere possibility. There is no
such thing as a 'stimulus inundation' since the neurophysiological
apparatus already screens off consciousness drastically, and the
operative medium of meaning has to work very hard to permit
something that is well digested to become actual. So the established
view of anthropology has to be revised. We will put the idea of
the very restricted resonance capacity of meaningfully constituted,
operatively closed systems in its place.
In the case of meaning-processing as well as living systems4
autopoiesis has to be secured before all else. This means that the
system exists only if, and as long as, meaningful information
18 Resonance
processing is continued. We can designate the structural technique

that makes this possible as a difference technique.5 The system
introduces its own distinctions and, with their help, grasps the
states and events that appear to it as information. Information
is thus a purely system-internal quality. There is no transference
of information from the environment into the system. The
environment remains what it is. At best, it contains data. Only
systems can see the environment because this requires the seeing
of other possibilities, the presence of a pattern of difference and
the situating of items within this pattern as a 'this instead of
that'. In the environment there is no 'instead of that', no 'this'
as a selection out of other possibilities, i.e., neither a pattern of
difference nor information. To emphasize this once again: 6 system
boundaries have to be drawn so that the world acquires the
possibility of observing itself. Otherwise there would be pure
facticity alone.
In a somewhat different terminology one could say that system
differentiation makes possible the establishment and reduction
of complexity. The system can place possibilities within the
environment and view what is found there as a selection from
numerous possibilities. It can project something negative and use
this to identify something positive. It can form expectations and
be surprised. All these are structures for the operation of systems
themselves. They presuppose that the system can distinguish itself
from the environment.
If physical systems have differentiation and highly selective
resonance at their disposal then this is certainly the case
for meaning-constituting systems too, especially society. The
difference technique can be used by these systems because
distinctions, negations, possibility projections and information
are and remain purely internal and because, in this respect, no
environmental contact is possible. In this way the systems remain
dependent on autopoiesis, on a continual self-renewal of their
elements by their elements, but because information and infor-
mation expectations, i.e., structures, are obtained by means of
difference projections, this closure is openness at the same time.
For the system can experience itself as its difference from the
environment by means of the very same difference technique.7
This in no way changes the internal closure of the interconnection
Resonance 19
of its own operations. Instead, this equips it with the capacity

to react to whatever is environment for it.
This theoretical account brings us to the following question:
which concepts and distinctions in social communication help us
to deal with the exposure to ecological dangers? It excludes the
very obvious and ordinary idea that there are facts that call for
reaction or else damage will result. But even facts have
communicative effect only as facts, and the establishment of a
fact is the establishment of a difference.8 Therefore we have to
ask in which difference patterns are facts grasped, which desired
states bring states into relief and how do expectations become
accustomed to whatever appears as reality to them?
In addition to this so-called 'constructivist' perspective, the
social system's differentiation must be kept in mind. It is just as
suggestive as it is misleading to assume that 'the' system reacts
to 'the' environment, even if this is only to 'its' own idea of 'the'
environment. The system/environment difference is indeed the
presupposition of all observation of the environment. But this
does not mean that the system as a closed unity can react to the
environment. The unity of the system is nothing more than the
closure of its autopoietic mode of operation. The operations
themselves are necessarily individual operations within the
systems, i.e., several among many others. There are no all-
encompassing operations. Besides, complex systems like societies
are differentiated into subsystems that treat other social domains as
their (socially internal) environment, i.e., differentiate themselves
within the society, for example, as a legally ordered political
system that can treat the economy, science, etc. as environment
and thereby relieve itself of direct political responsibility for their
operations.
This differentiation theorem has far-reaching consequences. It
implies:
1 That important performances of the societal system are con-
stantly executed by subsystems because this is the only way
to achieve a sufficient level of complexity, and that in order
to explore how a society can react to the exposure to ecological
dangers the constraints on the possibilities of its subsystems
must be examined. These in turn depend on the form of social
differentiation.
20 Resonance
2 The system's unity can, if necessary, be represented within the

system itself, where the concept of representation is understood
as a representatio identitatis, and not as a taking-the-place-
of-something-else. Representation is the reintroduction of the
system's unity within the system itself. This creates a difference
within it, whether this is sought or not. 9 The presentation of
the system's unity within the system itself must therefore fit
the pattern of system differentiation. It may appear as the
'top' if the system is differentiated in a hierarchical manner,
i.e., presents itself as stratification. Or it may appear as the
'center' if the system is differentiated according to the center/
periphery pattern (for example, ci ty/coun try side). It cannot
choose any of these forms of presentation if none of these
forms of differentiation exists. We will also have to consider
whether there are further possibilities and whether the exposure
to ecological danger could be an occasion to develop other
possibilities.
3 Since every operation is only one among many, every

operation within the system is observable by others. Formally,
observation means being treated as information on the basis
of a pattern of difference, normally through expectations that
are fulfilled or not. In this sense, self-observation constantly
accompanies the operations taking place within society. This
observation creates additional effects of its own, often in
opposition to those that the operations themselves intend.
Thus, on the one hand, there can be an immediate stifling of
initiated plans and, on the other, an effect-explosion which
neither waits for nor depends on the operations reaching their
intended goals.
Further inspection reveals that the theory of self-referentially

closed and thereby open social systems leads to considerable
complications. The concepts of system differentiation, represen-
tation and self-observation indicate what in particular needs to
be clarified to understand whether and how society can create
resonance because of exposure to ecological dangers. But it is
already clear that the problems cannot be solved merely with
admonitions and appeals to more environmental consciousness.
Resonance 21
Instead, the observation accompanying all political, economic

and scientific operations, and precisely from these perspectives,
may trigger one of those 'effect-explosions' that change society
- entirely independently of whether the relation of the social
system to its environment is improved in this way and, if it is,
according to what criteria.
39
The Observation of Observation
System resonance, then, is always in effect when the system is

stimulated by its environment. The stimulation can be registered
by the system if it possesses a corresponding capacity for
information processing permitting it to infer the presence of an
environment. Similarly, the system registers the effects of its own
behavior on the environment whenever this behavior triggers a
stimulation within the range of the system's possible perceptions.
The environment is the total horizon of information processing
that refers beyond the system. Thus it is an internal premiss for
the system's own operations constituted within the system when
the latter uses the difference of self-reference and other-reference
(or 'internal' and 'external') to order its own operations.
As an internal premiss, the system's environment has no
boundaries nor needs any. Presenting itself as a horizon, it is the
system-internal correlate of all references that extend beyond the
system. This means that, whenever necessary, any operation can
push it back still further. The horizon always recedes when it is
approached, but only in accordance with the system's own
operations. It can never be pushed through or transcended because
it is not a boundary. It accompanies every system operation when
this refers to something outside the system. As a horizon, it is
the possible object of intentions and communication; but only in
so far as the system can present the environment to itself as a
unity - and this requires that it can differentiate itself as a unity
from it.
In a somewhat different, Wittgensteinian formulation one could
Observation of Observation 23
say that a system can see only what it can see. It cannot see what
it cannot. Moreover, it cannot see that it cannot see this. For the
system this is something concealed 'behind' the horizon that, for
it, has no 'behind'. What has been called the 'cognized model'1
is the absolute reality for the system. It has a singular quality of
being or, logically speaking, univocality (Einwertigkeit). It is what
it is, and if it turns out that it is not what it seems to be then
the system has made a mistake! The system can operate only
with two values when it uses the distinction of self-reference and
other-reference.
All this necessarily holds true for a system's immediate
observation of what presents itself as environment to it. Neverthe-
less, a system that observes other systems has other possibilities.
Even if it posits its environment apodictically, like every other
system, the observation of a system by another system following
Humberto Maturana we will call this 'second-order observation'2
- can also observe the restrictions forced on the observed system
by its own mode of operation. The observing system can discover
that the environment of the observed system is not constituted
by boundaries at all, but, perhaps, by constraints. It can observe
the horizons of the observed system so that what they exclude
becomes evident. Using this, it can clarify the mode of operation
of the system/environment-relations in a kind of 'second-order
cybernetics'.3
At present, second-order cybernetics seems to be the place
where the problems of the foundations of logic and epistemology
can, at least, be handled if not 'solved'. We will therefore have
to examine it briefly because these problems become increasingly
important when we deal with science as a part of society, i.e.,
as part of the object discussed by it (including these words as
part of this text too).
Since social systems in general and societies in particular
constitute themselves through autopoietic self-reference, every
observer is confronted with the question of how these systems
come to terms with the problems of tautology and paradox that
necessarily follow when a system operates through self-reference
alone, i.e., when it must ground all its operations in self-reference.
The classical answer to this problem (Russell and Whitehead,
Tarski) is well known. Such a system has to interrupt or 'unfold'
24 Observation of Observation 24
self-reference while it distinguishes several levels as a hierarchy

of types, i.e., while it separates object-language, meta-language
and, if necessary, meta-meta-language. This solution does not
work, however, because the concept of level assumes a plurality,
i.e., a reference to other levels. This means that operations capable
of interrupting the hierarchy through the performance of a
'strange loop' cannot be eliminated.4 The hierarchy of levels can
be saved only by an arbitrary fiat: the instruction to ignore
operations that disobey the command to avoid paradoxes.
Questions 'why?' are not permitted despite the constant temptation
to raise them.
Since this requires the rejection of universalistic theories (and
would put us in the embarrassing position of never being able
to discover that society can summon only limited resonance to
its environment) another solution has to be found. The fatality
of the arbitrary proscription of thematizations aimed at avoiding
paradox can itself be avoided only through the distinction
of natural and artificial constraints on self-reference.5 Those
contraints on a system's self-reference that appear as natural or
necessary are the ones that are the conditions of the possibility
of operations, i.e., that conceal tautology or paradox in the
performance of self-reference. Those for which this is not the
case are artificial or contingent.
This distinction is always to be treated as system-relative
although it can be viewed as variable too. The constraints that
previously seemed necessary become artificial when learning
processes reveal how they can be replaced as eliminators of
tautology and paradox. In this context second-order cybernetics
is important.
An observer who recognizes that an object is a self-referential
system notices at the same time that it is constituted tautologously
and paradoxically, i.e., is arbitrary and inoperable, unobservable.
This produces a paradox of its own for the observer of a
self-referential system: the arbitrariness and impossibility of
observation. The observer can avoid this embarrassment by
distinguishing natural and artificial constraints applied to the
system observed. Then it becomes clear that this system cannot
see what it cannot see. For the observer whatever is necessary
and irreplaceable in the system can appear as contingent. With
the assumption of let us say, a supermodal, observationally

dictated distinction of necessary and contingent the observer can
eliminate the paradox while providing the observation with an
operational object. This can be done in a way that implies
learning possibilities for the object through the possibility of
the displacement of the boundary line between natural and
artificial constraints of complete self-reference.
Our concept of resonance assumes this second-order cybernetics.
It implies constraints, presupposes a reality that triggers no
resonance at all within the system and nevertheless presents
an environment for this second-order observation. From this
perspective one can see that the observed system constructs the
reality of its world through a recursive calculation of its
calculations,6 and since this is the case on the level of living,
neurophysiological and conscious systems it cannot be different
for social systems either. Second-order cybernetics can be used
to prove this. Consequently, it can draw no other conclusion
than that this applies to its own observation too, but at the same
time it can still see that what cannot be seen cannot be seen.7
Sociopsychological investigations of attribution have come to
similar conclusions entirely independent of this biologicocyber-
netic research tradition. Here, research proceeds under the title
of causal attribution. The actor's mode of attribution (first-order
observation) is distinguished from that of the observer's (second-
order observation). While the actor finds the bases for action
primarily in the situation itself, the observer sees the actor-in-
the-situation, looks for differences in the interpretation of the
situation by different actors and makes attributions primarily in
terms of the personal characteristics of the actor. 8 Correspond-
ingly, sociology has always concerned itself with actors who
already knew why they acted and, therefore, had to justify an
additional, transcendent, 'critical', knowledge-guiding interest
(Erkenntnisinteresse) . 9 In all these cases the beginning has to be
and this constitutes the innovation vis-a-vis the naive faith in
science - the fact that second-order observation together with.its
theoretical apparatus is possible only as a performance of
structured autopoiesis, i.e., it is not 'objectively better' knowledge
but only a different knowledge that takes itself for better.
To analyse the problem of the exposure to ecological dangers
26 Observation of Observation 26
with the necessary exactness, second-order cybernetics must be

taken as the starting-point. If the starting-point were an 'objecti-
vely' given reality that, for the time being, was still full of
surprises and unknown qualities then the only issue would be to
improve science so that it could know the reality better. But then
the relations of the other systems to their environment - for even
within society there are many other systems - would not be
grasped sufficiently. Even science would not be able to understand
why with its 'better knowledge' it often finds no resonance within
society because what it comes to know its 'better' knowledge
- would have no value at all as reality in the environment of
other systems or is at best a scientific theory for them.
Not much is gained, therefore, by following an ontological
theory of reality (which corresponds to a first-order observation
of the environment) because this theory is not in a position to
grasp the problem as such. We have to choose a second-order
cybernetics as the point of departure. We have to see that what
cannot be seen cannot be seen. Only then can we discover why
it is so difficult for our society to react to the exposure to
ecological dangers despite, and even because of, its numerous
function systems.
To the extent that society can differentiate structurally an
observing of observing and explain this theoretically it finds itself
in the position of establishing the conditions under which it will
react through its respective (function) systems to whatever is
environment for them. This is not a question of creating a basis
for better possibilities of action. Nor is it even something like a
'domination-free discourse'. Inasmuch as this idea is an improve-
ment over the old, unsophisticated way of watering the tree of
freedom with the blood of tyrants, the real problem is to be
found neither in a lack of justifications nor in the pattern of
coercion and freedom. Nor is it even to be found in removing
barriers to rational consensus and harmonious coexistence. The
problem is the acquisition of a different kind of insight.
In many ways modern society has opened up possibilities for
observing and describing how its systems operate and under what
conditions they observe their environment. The only drawback
is that this observing of observing is not disciplined enough by
self-observation. It appears as better knowledge. But in reality it
is only a particular kind of observing of its own environment.10

Under these conditions the idea that rational consensus ought to
be attained is quickly trivialized. Those who think they know
that this is going to be a protracted enterprise use this idea and
test their willingness to make concessions according to their own
judgement. But every operation and every observation has
structural limitations, which is precisely what second-order
observation makes clear. A better evaluation of the situation is
attainable only when this insight is applied to itself, i.e., is
employed recursively. When this is done the constraints on the
ability to observe, describe and turn insights into operations have
to be analysed and compared. Any protest against such constraints
would be strangely naive and, as such, would merit observation
itself - if not by the protester then at least by others who observe
the protester.
45
Communication as a Social
Operation
In the following I will not consider the very limited, socially

dependent possibilities of the consciousness of individual psychical
systems and will make society the sole system reference. By
society I mean the most encompassing system of meaningful
communication. Any limitation, for example, to organizations,1
would restrict the investigation too much. So the question is how,
as an operatively closed system of meaningful communication,
does society communicate about its environment? More specifi-
cally, what are its possibilities for communicating about exposure
to ecological dangers?
We must be careful in our presentation of the concept of
exposure to ecological danger as long as we do not know what
it is about. So we will understand it very broadly. We will take
it to designate any communication about the environment
that seeks to bring about a change in the structures of the
communicative system that is society. It should be noted that
this is a phenomenon that is exclusively internal to society. It is
not a matter of blatantly objective facts, for example, that
oilsupplies are decreasing, that the temperature of rivers is
increasing, that forests are being defoliated or that the skies and
the seas are being polluted. All this may or may not be the case.
But as physical, chemical or biological facts they create no social
resonance as long as they are not the subject of communication.
Fish or humans may die because swimming in the seas and rivers
has become unhealthy. The oil-pumps may run dry and the
average climatic temperatures may rise or fall. As long as this is
Communication as a Social Operation 29
not the subject of communication it has no social effect. Society

is an environmentally sensitive (open) but operatively closed
system. Its sole mode of observation is communication. It is
limited to communicating meaningfully and regulating this
communication through communication. Thus it can only expose
itself to danger.
To formulate this important starting-point differently, one
could say that the environment of the social system cannot
communicate with society. Communication is an exclusively
social operation. On the level of this exclusively social mode of
operation there is neither input nor output. The environment can
make itself noticed only by means of communicative irritations
or disturbances, and then these have to react to themselves. Just
as one's own lived-body cannot announce itself to consciousness
through conscious channels but only through irritations, feelings
of pressure, annoyance, pain, etc., that is, only in a way that can
produce resonance for consciousness. Using concepts introduced
by Francisco Varela2 one can say that there is no coupling by
input, only coupling by closure.
This is also true - and this gives the argument a particularly
incisive importance for the relation of consciousness and
communication. Even the consciousness of psychical systems
belongs to the environment of the societal system. As such, it is
only a psychical, not a social fact. Of course, human consciousness
and human life, like so much else, belong to the indispensable
conditions of social communication. But this does not change the
fact that, as the production of ideas by ideas, the processes of
consciousness are not communications.3 (Husserl saw a proof of
the transcendentality of consciousness in this discovery; we merely
infer from it a different system reference.) Thus, once again, the
relation of conscious systems and social system has to reckon
with a resonance threshold that selects in a very rigorous way.
Whatever 'ecological awareness' may occur empirically within a
consciousness, it is still a long way from this to a socially
effective communication. Afterwards, this difference between
communication and consciousness can itself become a theme of
communication. But then the communication is about 'alienation',
'apathy', the resignation or protest of youth or similar themes
connected only indirectly with the exposure to ecological dangers.
Viewed realistically, we have to reverse the idea that a 'subject'

must first consciously resolve to communicate in order to act
communicatively.4 Only when (and for reasons that cannot be
attributed to a consciousness) ecological communication is set in
motion and begins to co-determine the autopoiesis of social
communication can one expect the themes of this communication
gradually to become conscious contents too. This simply means
that social communication changes its environment, in this case
mental states. What follows from this for society can be grasped
only through an analysis of possible communication, through an
analysis of the social system's capacity for resonance.
Conscious systems, therefore, are limited to producing irri-
tations, disturbances or evasive themes if they have not first
adjusted themselves to the social conditions of communicability.
In this case the sharp delimitation of what is communicable
socially signifies comprehensibility or noise. In triggering social
communication-processes consciousness is guided either by the
corresponding valid structures (including the structurally given
possibilities of changing the structures) or it merely creates noise
which, according to the possibilities of social communication, is
eliminated or transformed into something communicable. This
statement should not be misread as the assumption of a static
system. On the contrary, a communication system's structures
are highly flexible. They can be varied in their use and can even
be used in a counter-sensical way, for example, ironically or for
guiding deviant behavior. All this changes nothing about the fact
that the threshold of possible and possibly understandable or
even possibly successful communication works in a highly selective
way, i.e., rejects whatever cannot find resonance.
Not least important is the realization that conscious systems,
which are outside the domain of linguistic (thus communicable)
articulation and depend on perception and intuitive presentation,
are hardly in a position to order trains of thought into
temporalized complexity. Even if an 'ecological awareness' would
arise in some conscious systems or others it would have
properties that would be almost useless for society. It would be
overdetermined perceptually or intuitively. In any event, this is
what the underlying systems-theory concludes. It would be more
likely to present its ecological theme negatively through particular
proposals than to present it in communication as positive

knowledge about the environment. It would either tend to anxiety
and protest or to a critique of society that would be unable
to deal with its environment adequately. It would reach its
generalizations only in a negative way and lean toward an
emotional self-certainty typical of cases in which it no longer
knows. In addition, it would depend on socially given forms and
connection capabilities (Anschlussfahigkeiten) or persist in an
ever-possible negation that produces little of value.
49
Ecological Knowledge and

Social Communication
As surprising as it may sound we still adhere provisionally to the

idea that society can expose only itself to ecological danger. This
means not only that it changes the environment in a way affecting
the continuation of social reproduction on the contemporary
evolutionary level but also if we disregard the unlikely case of a
radical extinction of all human life, that it can endanger
communication only through communication. Consequently,
somehow and somewhere society will have to thematize the
connections between its own operations and environmental
changes as problems of continued operations, if only for the sake
of finding resonance within social communication. So the key
question becomes how society structures its capacity for processing
environmental information.
As far as I know this question has been raised and discussed
only for relatively simple societies, those living on an archaic
level.1 These societies were able to present supernatural matters
better than natural ones and therefore they sought ecological self-
regulation in mythicomagical ideas, i.e., in taboos and rituals
dealing with the environmental conditions of survival. The famous
pig cycle not the one of the political economists but that of
the Maring living in New Guinea2 - is a classical example of
this. Whenever the pig population increases too much and they
begin to ravage the food-supply strong ritualized justifications
come into play to arrange a great feast that re-establishes a
balance in the number of pigs and regulates the protein
consumption of the tribe. A surprisingly pragmatic attitude toward
Ecological Knowledge and Social Communication 33
sacred matters makes it possible to balance the environmental

conditions of the system without making this requirement explicit.
Among other things this means that other, functionally equivalent
solutions of the problem of periodic imbalances - for example,
when population increases, to cultivate better-protected gardens
and to increase the pig population at the same time - are not
even considered. A ritually regulated society's own structures do
not program it toward growth.
Of course, these archaic societies possessed the necessary
knowledge for survival and productive know-how. They certainly
knew that pigs ravaged gardens, and they also knew that excessive
use of the land decreases the harvest and makes the land unusable.
But the semantic organization of this knowledge and its connection
with the motivational guidance of human behavior was left to a
semantics of the sacred - precisely because supernatural matters
are easier to organize than natural ones. In this way uncertainties
and prescribed social reactions could be intercepted and transfor-
med into social certainties, and one could deal more or less
successfully with the circumstance that reactions to environmental
problems within society have diverse effects, i.e., favor or frustrate
one more than another.
Archaic social systems were also responsible for important and
irreversible environmental changes. The desolation brought about
by deforestation is a good example of this, and demonstrates
that the problem is not new. The extent of the possibilities as
well as the social pressure to exploit them, however, have grown
enormously. Besides, and perhaps even more significantly, the
latent premisses of a religious guidance of society have evaporated
in the transition to modernity, i.e., the premisses functioned only
with the help of mystifications. Religious semantics always had
to operate3 with secrecy or with strategically placed vagaries.
Ignorance and its accompanying uncertainty were exposed to a
process of semantic atrophy. Thus it contracted into a small
residue of vague indeterminacy (for example, Divine Will) and
acquired a form that could be ignored.
It is evident that modern society no longer treats ecological
problems in this way, no longer solves or at least modifies them
via a semantics of the sacred. The transformations of cultural
and religious semantics that occurred as a result of writing,
alphabetizing and printing hardly permit dealing with environmen-

tal problems via taboos and ritualization any longer.4
Powerful counter-movements tried to do this and at the very
beginning of modernity sought ultimate answers in the arcane.
Erasmus, for example, pleaded against Luther for the self-
confinement of religious texts in the interests of human freedom.5
Hermetics employed ancient tradition and staked everything on
what was concealed in secrets.6 But this was all in vain! Printing
and widespread dissemination confer a completely new form on
technological information, formulas and the explanation and
handing-down of directions. This means that knowledge now has
to be understandable in itself. It has to present itself as
differentiated and thereby increasingly exposes itself to compari-
son and correction.7 Reference to ancient secrets, distant authori-
ties or awe-inspiring mysteries are confronted with wanting-to-
know-more-precisely. The idea of the cosmos, inherited from the
Platonic tradition, as a large, visible and yet unfathomable
organism, collapsed. Because of writing's duplication of oral
language and the changing demands on important communication
it is no longer possible to reconcile knowledge with motivation
through mysteries and secrets. The traditional figures evoking
respect and awe are no longer effective and cannot take the place
of a knowledge that is exact and proven. 8 Mysteries are reserved
for Holy Scripture. In any event, their use in everyday life shows
a contempt for the communication partner. 9
Even if these hypotheses are correct, namely that writing, the
alphabet and printing actually stimulated profound changes in
the communication system of society they still do not supply an
adequate description of the contemporary situation and its
chances for dealing with ecological problems communicatively.
Theoretical tools that are more complex are necessary to describe
modern society. New dissemination techniques for communication
are an essential, but only one factor among these. Another is the
change in the primary form of society's differentiation from the
stratification of lineage, clans and families to the differentiation
of function systems. This means that, today, each of the most
important subsystems of society is directed to a specific and
primary function that pertains to it alone. This formative principle
explains the enormous growth of modern society's performance
and complexity. At the same time it reveals the problems of

integration, i.e., of the negligible resonance capacity among the
subsystems of society as well as the relation of society to its
environment. The theory of functional system-differentiation is a
far-reaching, elegant and economical instrument for explaining
the positive and negative aspects of modern society. 10 Whether
it is correct is an entirely different question.
53
Binary Coding
We can now formulate our question more exactly. How can

environmental problems find resonance in social communication
if society is differentiated into function systems and can react to
events and changes in the environment only through these? After
all, in such a system there is communication that is not
coordinated functionally or coordinated only ambiguously the
communication of the streets, so to say, or in somewhat more
high-sounding jargon: 'life-world' communication. 1 Communi-
cation that affects society, however, depends on the possibilities
of the function systems. We will therefore have to investigate
these first because this is the only way to consider realistically
what possibilities exist for communication in a society that
distances itself consciously from all function systems - either
through protesting, moralizing or through a blurring of differen-
tiation.
The most important function systems structure their communi-
cation through a binary or dual-valued code that, from the
viewpoint of its specific function, claims universal validity and
excludes further possibilities. The classical example of this is the
binary code of logic used by science. Analogously, the legal
system operates with a code of legal and illegal. The economy
uses property and money to distinguish clearly between possession
and non-possession so that long-term possibilities of the transfer
of commodities and money can be organized and calculated, and
politics is guided by the questions of power that accompany
governmental authority and which are put to the vote using
Binary Coding 37
ideological codes like conservative versus progressive or restrictive

versus expansive.2 The significance of these functional domains
for the modernization of contemporary society will become
evident immediately once we approach the problem of steering
communication through binary codes.
As one can see, from the standpoint of a second-order
cybernetics, i.e., from the observing of observations, every binary
code resolves tautologies and paradoxes for the system that
operates with this code. The unity that would be unbearable in
the form of a tautology (for example, legal is legal) or in the
form of a paradox (one cannot legally maintain that one is legal)
is replaced by a difference (in this example the difference of legal
and illegal). Then the system can use this difference to steer its
operations. It can oscillate within it, and develop programs that
regulate the coordination of the operations to the positions and
counter-positions of the code without ever raising the question
of the unity of the code itself. When this is achieved self-reference
can unfold itself and does not have to be enlisted immediately
and directly as unity (although within the code it comes into
play dialectically, as it were, since every position is identified in
reference to its counter-position). At the same time one should
remember that an observer - a position within which we
presently find ourselves can see through this entire manouevre.
Nevertheless, the possibility of observation arises for the observer
only because a system (or a hierarchy or other functionally
equivalent solutions of the problem) chooses a code to conceal
those aspects of its self-reference that would reveal the tautology
and paradox of its operational bases.
Binary codes are duplication rules. They form within the
communication process when information acquires value and is
exposed to a corresponding counter-value. The reality that is
treated according to the code is singular. But it is, as it were,
duplicated Actively so that every value can find its complement
and be reflected in it. Of course, there are no negative facts. The
world, after all, is what it is. But by coding communication
about reality one can treat everything that becomes a subject of
communication contingently and reflect it in a counter-value.
This complementarity is not a matter of increasing or decreasing
the number of facts, of an 'Another beer, please!', but of
38 Binary Coding 38
projecting a positive/negative distinction with whose help the

possibility and consequences of the opposite can be examined.
Thus it is neither a question of an exclusively communicative
accomplishment nor of a state of affairs in the world that needs
to be depicted in the communication.
Binary codes of this kind can be viewed as highly successful
and important evolutionary achievements that have only attained
their contemporary degree of abstraction and technical proficiency
after a long development.3 At least the most important character-
istics of this structure deserve to be mentioned.
1 Codes are totalizing constructions,4 i.e., all-encompassing

constructions having a claim to universality and possessing no
ontological limit. Everything that falls within their domain of
relevance is assigned to the one value or the other, tertium
non datur. Just as God Himself stood outside the Creation in
creating the difference of heaven and earth so a third possibility
exists for codes - at best only parasitically (parasitically is
meant here in the sense in which Michel Serres uses the term).5
2 As the reference to everything that can be treated as information

within the code, totalization leads to the contingency of all
phenomena without exception. Everything appears in the light
of the possibility of its counter value, as neither necessary nor
impossible. Any necessities or impossibilities have to be
reintroduced in a counter-move - perhaps to remove paradox
from the code (see 4 below) and therefore remain doubtful.
3 Codes are in so far-as-abstractions. They are valid only in so

far as communication chooses their domain of application
(which, by the way, it does not have to do). After all, not
every situation is a matter of truth or justice or property.
Thus the use of a code is a socially contingent phenomenon
since this is the only way it can totalize a schema that reduces
everything to two opposed possibilities.6
This produces a connection between coding and functional
specification in the process of evolution: certain binary codes
are used only when the operations to be coded occur in the
corresponding function-system. Just as, on the other hand,
social function-systems attain universal relevance for all the
Binary Coding 39
operations concerning them because they are specialized

according to the operations of a determinate code.
4 Codes, as mentioned already, resolve the paradox inherent in
the problematic underlying every self-referential relation. Yet
every coding leads to the problem of applying the code to
itself and thereby, sometimes, to a paradox. Logical antinomies
like the liar's paradox - 'This sentence is a lie' - are well
known. But other codes have similar problems too: for
example, by what right is the difference of legal and illegal
introduced and upheld? Another example of paradox can be
found in the increasing dependence of those occupying a
greater position of power on help from others. In the
parliamentary code of ruling versus opposition parties the
ruling party will often show an inclination to undercut the
opposition by anticipating its position and taking it, or capital
finds itself under the constant pressure to reinvest, i.e., under
the constant pressure to facilitate the consumption of others.
When the code is transformed into a contradiction it deactivates
this problem for the operations that the code regulates. 'A
because non-A' becomes 'A is non-A' and in this form the
problem is eliminated. Such contradictions - allowing for
some residual problems that can be left aside for special
treatment are avoidable. But precisely in this regard codes
remain sensitive to changes in the conditions of social
plausibility.7
5 Coding uses and reinforces the old adage that opposites
attract,8 or in the words of latin rhetoric, Contrariorum est
eadem disciplina. Difference integrates. The transition from
one side to the other is pre-programmed in the difference-
scheme and thereby facilitated. Negation is all that is needed
to accomplish this for the logical code. The operative proximity
of value and counter-value leads almost necessarily to the
differentiation of corresponding function-systems. It is simple
to transform property into non-property through exchange or
sale. It is much more difficult, however, to submit it to a legal
examination or apply it politically.
6 In binary coding the guiding value of the code (truth, justice,
property, etc.), has at the same time to surrender its right to
40 Binary Coding 40
serve as the criterion of selection because to do so would

contradict the formal equivalence of position and negation.
After all, the establishment of falsity can have a much more
positive effect upon the advancement of science than the
establishment of truth. Very often this is simply a matter of
the theoretical context. 9 Property can become a burden when
it is just an expense and not a source of revenue. This is
determined completely by the context of investment, and even
the government is inclined to renounce responsibility for
certain political decisions. This again is determined entirely
by the kind of political programs (policies). Criteria, therefore,
cannot be frozen into the abstractness of the code because
they are not designed to establish the possibility of functionally
specified operations. Instead, they serve much more concretely
to steer correct and useful operations. The code, therefore,
can outlast the change of criteria (and, in principle, of all
criteria) although it is hard to imagine that everything could
be changed all at once and that the code could be suspended
for the purposes of a completely new beginning.
7 The difference of code and criteria for correct operations (or

of coding and programming) makes possible the combination
of closure and openness in the same system. In reference to
its code, the system operates as a closed system; every value
like 'true' and 'false' refers to its respective counter-value
alone and never to other, external values. But at the same
time, the programming of the system makes it possible to
bring external data to light, i.e., to fix the conditions under
which one or the other value is posited. The more abstract
and technical the coding, the richer the multiplicity of the
(internal) operations with which the system can operate as
closed and open at the same time, i.e., to react to internal
and external conditions. One can also designate this as an
increase in resonance capacity. But no matter how 'responsive'
the system may be structurally10 and no matter how sensitive
its own frequencies, its capacity for reaction rests on the closed
polarity of its code and is sharply limited by this.
8 Coding effectively excludes third values although these may
be reintroduced into the system on the level of programming
Binary Coding 41
correct behavior - of course, only under the conditions valid

for this level. Despite the explosiveness of new themes, a
threefold code, perhaps of the type true/false/environment or
legal/illegal/suffering, is never a possibility. Nevertheless,
environmental problems can become the object of research
programs or human suffering and their prevention the object
of legal regulations. Thus the differentiation of coding and
programming makes the reappearance of the third value
possible; but only to co-steer the allocation of the code-values
on which it primarily depends.
9 Furthermore, coding signifies the bifurcation of operations

and the structures built on them with the well-known
consequence of the constitution of historically irreversible
complexity. The self-reflecting distinction established within
the code produces consequences that are based on the fact
that truth is not falsity and that property or political
power can be transformed into their opposites only through
determinate procedures like exchange or voting. The structured
complexity that comes into being in this way is not under the
system's control. Neither can it be grasped as a unity nor can
the code be applied to itself. In other words, within the system
one cannot decide whether all falsities are false, whether all
injustice is illegal and the expropriation of property is
conceivable only as revolution (and then realizable only as a
partial transferral within the economic system).
10 Coding channels all further information processing into its

domain and is guided primarily by its initial distinction because
this is the only way to generate information and co-ordinate
it with a function system. All further information processing
transforms differences into differences,11 for example, in
determining whether a particular investment of a capital sum
will be profitable and whether a corresponding demand could
be generated in the market so that a price can be fixed.
11 When coding possesses all these characteristics it is technically

the most effective and successful form of differentiating
function-systems. But this does not mean that such a univocal
coding is the only way to form function systems. The education
42 Binary Coding 42
system, for example, uses a rather unwelcome code to meet

the demands of its selections and has an entirely different
basis in the school's complexes of organization and interaction.
Similar questions could be addressed to the coding of religion.
But this does not mean that, historically, the code is established
first and then a corresponding system is formed. Evolution
creates its own conditions as it progresses and comes to a halt
when and as long as this does not succeed. For a description
of modern society one will have to admit that important and
distinctively modern function systems have become identified
through a binary code that is specifically valid for each of
them. They know, in any event, what their guiding difference
is and how it functions in the operations of the system.
Whether they can show what the meaning of their unity is
and formulate this as a theory about themselves and whether
such a self-description grasps their social function accurately
is an entirely different question. Their differentiation does not
depend on all this. Theories of reflection occur in such systems
only secondarily - only for the defense of their autonomy and
only on the basis of the demand for meaning that the system
already presupposes structurally. This has been confirmed in
a remarkable way by the development of scientific theories,
political theory, economics and by legal theory since the
second half of the eighteenth century.
12 Because function systems are not differentiated as regions of

being, collections or by means of unified viewpoints but
instead by means of differences, a high degree of reciprocal
dependence is possible. Such dependencies are often interpreted
as constraints on autonomy if not as symptoms of the reversal
of differentiation. Actually the contrary is the case. Functional
differentiation promotes interdependence and an integration
of the entire system because every function system must assume
that other functions have to be fulfilled elsewhere. This is the
precise purpose of the binary code: to differentiate its own
domains of contingency and its own procedures for creating
differences through differences - and not essentially for
differentiating exclusive orders of existence. Operations can
therefore switch very quickly from the legal to the political
Binary Coding 43
or from the scientific to the economic code. This possibility

does not deny system differentiation. Instead it is attainable
only on the basis of it.
My position is that binary codes having these characteristics

occur in social evolution and that, if they are put into operation,
corresponding systems tend to be differentiated. Traces of such
a development can be shown for ancient Greek culture in the
clear differentiation of the semantics of logic and epistemology,
politics and ethics, and economics and philia. 12 At first the
traditional models of social differentiation according to the
distinction of city/countryside and stratification predominated
and society presented itself within a scheme of religiously justified
moral communication. In this sense, both urban life and,
subsequently, the life of the nobility are ethical postulates.13 Not
until modernity did society gradually switch to the primacy of
functional differentiation that, since the middle of the eighteenth
century, has been accompanied by a corresponding problem-
awareness. Ever since, a plurality of functionally specified codes
has steered resonance to the environment instead of a socially
unified or upper-class 'ethos', and these codes lack integration
only to the extent that a positive valuation in one code, for
example, 'true' does not automatically entail a positive valuation
in the other codes, for example, as legally or economically
significant.
61
Codes, Criteria, Programs
The thesis of the functionally specific coding of the operations

of modern society is only a first step along the path toward a
more concrete description. We must always keep in mind that
not all the system's communications are ordered in this way, i.e.,
divided into one or the other code. Differentiation never occurs
as the decomposition of a given set of operations, but as the
separation of subsystems operating under the direction of a code
within society.
Binary codes begin as different, highly abstract schemas that,
at the same time, leave unclear how the operations of society are
actually regulated. At first glance they appear as the coding of
preferences. This would mean that truth is better than falsity or
legality better than illegality or that it is better to have than not
to have. But if the actual operations are observed along with the
preferences that are expressed in them something different is
revealed. The truth of the proposition that mice have tails is not
valued as highly as the demonstration of the falsity of important
physical theories. Within the legal system much effort has been
expended on shifting certain laws into the domain of non-
constitutionality (or in other words, there is no preference of
particular laws or statutes for constitutionality), and the same is
true for the economy. Many firms would be in a more fortunate
position and would have better business results if they did not
own certain plants.
In order to deal with such situations theoretically two levels
have to be distinguished in the analysis of system structures: the
Codes, Criteria, Programs 45
level of coding and the level on which the conditions of the

suitability of operations are fixed and, if necessary, varied. Or,
to repeat an argument of the preceding chapter, the code's values
are not criteria. Truth itself, for example, is not a criterion of
truth.1 Criteria refer to binary coding, according to the established
tradition of concepts like canon, criterion, regula.1 But they are
not a term of the code itself.
We will formulate this difference of levels with the distinction
between coding and programming.3 On the level of coding a
system is differentiated by means of a binary scheme. At the
same time it establishes itself on this level as a closed system.
This means that a value can be abandoned only for the sake of
its counter-value. Accordingly, the alternative 'true or false' is
permissible, but not 'true or ugly'. The codes are closed 'contrast
sets'.4 Programs, however, are given conditions for the suitability
of the selection of operations. On one hand, they enable a
'concretizing' (or 'operationalization') of the requirements that a
function system has to satisfy, and on the other, they have to
remain variable to a certain extent because of this. On the
program level a system can change structures without losing its
code-determined identity. On this level, therefore, learning
capacity can be organized to a certain extent, so through the
differentiation of coding and programming a system acquires the
possibility of operating as closed and open simultaneously. As a
result, this differentiation, together with its accompanying capacity
for articulation, is the key to the problem of social resonance to
the exposure to environmental dangers.
Viewed historically, this kind of differentiation developed very
gradually and became a necessity only when function systems
were sufficiently differentiated. Within the existing tradition of
political ethics and natural law the code-values (positive/negative)
and the generalized formulas for the conditions of the suitability
or usefulness of behavior could not be distinguished. Instead, the
unity of the good and the right was rooted in a religious semantics
that transcended the domain for which the difference of good
and bad was meaningful: the world. 'The good' is thereby
doubled. Transcendently it operates without a counter-value, but
in the world it operates with the counter-value of 'the bad'. 5 It
becomes logically ambiguous and forces the construction of a
multi-level theory, for example, a hierarchy of laws. Nevertheless,

Enlightenment thinkers like Rousseau followed this scheme even
if 'the good' was sanctified as nature and made its ambiguity
felt only when it had unfortunate consequences. The French
Revolution demonstrated this in a striking manner and with it
brought the history of wisdom to an end. All reflection then has
to start ah ovo taking this fact into account.
A kind of demoralization of the most important codes -
resulting from experiences with a monetary economy - had
already been attempted.6 Correspondingly, the history of economic
reflection began with a search for a functional equivalent for
morality, namely with Adam Smith's Wealth of Nations. By then
science had long since promoted the idea of an 'invisible hand'
to show that scientific progress was near and that it would bring
immeasurable good with it. 7 Of course, there was no reason for
the fear that things can go wrong which haunts us today. The
metaphor of the 'invisible hand' and the reference to successful
progress was therefore enough to refute the theories of transcend-
ence which in those days were rejected as 'dogma'.
For the history of ideas we now find belief in progress with
its supporting metaphysics in a period of transition. The new
order of functional differentiation opens up possibilities that
could not develop according to the levels of the old society. An
entirely new kind of theory of reflection becomes possible one
related to the autonomy, self-value and function of the individual
function-systems without considering their interplay. Whatever is
triggered structurally is, at the same time, still screened off
semantically. It has not yet made its presence felt and becomes
acute for the first time in the nineteenth century as the 'social
question'. But social reflection found no clear guidance from this
and, lacking any consideration of environmental problems, no
external support.
vEven if one defers the question of ecological problems for the
present, it can still be assumed that functional differentiation
if it develops as a consequence of the differentiation of codes -
leads to new ways of formulating problems and new theories of
reflection.]In this regard two things must be kept in mind. The
first is that the levels of coding and the programming of function
systems become more clearly differentiated. The second and
this compensates for the differentiation is that programs

determining the criteria of suitability are formulated only in co-
ordination with particular codes and are not transferred from
one code to another.
What this means for theories of reflection that try to describe
the system's unity can be seen with the help of the following
diagrams of the legal system. The development of a regulative
semantics goes from the hierarchically inclined composition in
figure 1 to the more rigorously differentiated composition in
figure 2. If a hierarchy remains then it can reside only in the
subordination of the programs to the codes. Accordingly, the
theory of law abandons the medieval interpretation (predominat-
ing in canon, Roman and even in English law) that decisive
authority is given only for legal operation and has no authority
in the case of illegal. In the nineteenth century, however, the
Eternal Law
(just law)
Natural Law
perfect use" corrupt use
positive law
legal illegal
Figure 1 Hierarchical construct of the legal system
mere fact of authority was the goal. But if this happens the legal
system is, as it were, coupled to the political system without any
awareness of its own function. Thus its unity cannot be reflected
adequately. The transcendent, external support supplied by
natural law is eliminated and (for the time being) not replaced.
So the use of natural law as the theory of law has to be rejected
within the legal system. Programs assume the task of the correct
handling of the values 'legal' or 'illegal'. The unity of all the
46
code program
unity p justice
operation legal/illegal valid legal norms
Figure 2 Differentiated construct of the legal system

conditions of the correctness of programs had been reflected
traditionally in the title of 'justice' (by which a virtue or general
norm was understood). With a more rigorous differentiation of
codes and programs directing the operations the theory of law
had to make its concept of justice more precise.8 But it still finds
itself in the embarrassing position of having to formulate an
expression of unity for the difference of legal and illegal that, as
a meta-norm, idea or ideal no longer fits the semantic domain
of justice. The allowance of greater contingency in system
differentiation and reflection on the structural and semantic level
leads to the still largely unsolved problems of a satisfactory self-
presentation of the system, to say nothing of a theory of the
encompassing system of society.
One can assume that this provides a chance to extend the
resonance capacity of society and its function systems. The
metaphysico-moral conceptual framework may have been too
narrow a context of possibilities. Although its rejection is not
necessarily to be regretted any attempts at a 'rehabilitation' ought
to proceed very carefully. All this requires further investigations.
any event we have to begin from the fact that resonance to
the exposure to ecological danger is created essentially through
these function systems and cannot be, or at most only secondarily,
a matter of morality. To put it more precisely, the internal
dynamics and sensitivity of function systems like politics,
economy, science or law are disturbed by environmental problems.
Sometimes this happens directly as when resources dry up or
catastrophes threaten. But it also occurs indirectly via socially
mediated interdependences when, for example, the economy is
forced to react to legal precepts even if it would attain better
results following its own ideas. J
-On one hand, functional differentiation is possible only through
the rejection of redundancy J Function systems cannot step in for,
replace or even simply relieve one another.9 All equivalences are
ordered according to a specific function, i.e., within the system.

Politics cannot take the place of science, nor can science the place
of law - and so forth for all relations between systems. The old,
multifunctional institutions and moralities are, therefore, dissolved
and replaced by a co-ordination of specific codes to specific
systems that distinguishes modern society from all those before
it.
On the other hand, functional differentiation triggers an
enormous internal dynamics within the function systems which
combines intense resistance with very specific sensibilities to
irritations and disturbances. But this impedes the theoretical
description of the social system as a whole.iÊach function system;
has to be analysed individually according to its own specific \
resonance capacity .^Nevertheless, social resonance as a whole is j
not merely the sum of the resonances of each of the specific !
function-systems. The subsystems are environments for one
another. They can produce a process of resonating disturbances
when one subsystem reacts to environmental changes and alters
the social environment of the other subsystems. In this way a
scarcity of resources can produce political problems in addition
to economic problems like price increases. It can even force (
scientific research into certain directions instead of others.\_An I
extreme political sensitivity to environmental questions may
burden the economy with additional costs, mean the loss of jobs ^
and lead to political problems of its own,i and this same political 1
resonance capacity may trigger a new wave in the flood of norms
overwhelming the specifically juristic mode of handling legal
questions. This wave then surges back into the political system
and it begins to operate schizophrenically, namely bringing more
under legal jurisdiction (Verrechtlicbung) when it wants to bring
less (Entrecbtlicbung).
But disturbances are not the only things transmitted and
thereby partially absorbed and reinforced. The working together
of function systems is also necessary in practically all cases. For
example, scientific research has made the construction of nuclear
power plants economically possible through a political decision
about legal liability limitations. The world is just not constituted
so that events generally fit within the framework of one function
alone. Functional specification is an effective as well as risky,
46
evolutionarily improbable achievement of complex systems. It is

purchased at the cost of extensive internal interdependences that
are controlled by system boundaries. But it would be quite
mistaken to infer dedifferentiation from these reciprocal depen-
dencies.10 Instead they are a proof that modern society differen-
tiates its primary subsystems with reference to specific functions
and thereby prevents them from stepping in reciprocally for one
another as functional equivalents. It also makes them depend on
one another whenever problems can be solved only through their
co-operation.
Society's resonance therefore has to be analysed on two levels
at the same time (and here, as always, the idea of different 'levels'
conceals a theoretical systems-problem). On one hand, resonance
is conditioned by society's differentiation into function systems
(instead of into levels with dismay in the lower and responsibility
in the upper). On the other, it is structured by the different types
of codes and programs of the subsystems that affect one another
according to the general model of system and environment. As
one can see readily, this produces effects within the system which
are unlike the changes in the environment that originally triggered
them. These in turn are observed with respect to their own degree
of danger and so are in need of control. But all this occurs, if at
all, only within function systems according to specific codes and
programs.
10
Economy
Among society's many function systems the economy deserves

first consideration.(By the economy we mean all those operations
transacted through the payment of money. Whenever money is
involved, directly or indirectly, the economy is involved regardless
of who makes the payment and whose needs are affected. This
occurs, for example, in the collection of taxes or government
expenditure for public goods^ It does not occur in the case of
pumping oil out of the ground except when this process is
regulated economically to produce a profit that can be expressed
monetarily.
This definition of the economy describes the modern one that
is differentiated through the monetary mechanism.1 Compared
with earlier monetary systems like the medieval, however, it
demonstrates a striking limitation on what can be bought with
money. Today neither salvation nor the special providence of
transcendent powers, neither political offices nor tax-rates, neither
government assessments nor similar sources of income can be
purchased with money. 2 ^This restriction is the indispensable
condition of differentiating the economy fronT\religion and
politics j i t is the condition of the unification of the economy and
its autonomous closure as a self-governing, operative function-
system of societÎn other words, only through limitation do< s
the economy achieve the immense internal complexity of a
monetarily integrated systemT|ln the same way it also increas< s
its capacity for encompassing the satisfaction of needs and
production and moves the traditional domain of the domestic
52 Economy
economy further to the margin, i Limitation, therefore, is the ;

condition of its expansion, and this expansion contains the much /
deplored consequences for society's environment]
Originally the economy was coded through property. This
forced every participant into the alternative of owning property
or not regarding all the goods capable of bein^ owned. Ownership
by one participant excluded all the othersj. This was the only
way that exchange was possible and property was capable of /
fulfilling certain eâiogical Junctions. ] A tolerant treatment of ^
nature is beneficial to property because one can defend it against
others and they can eventually be brought through legal channels
to replace any damage.3 In its pre-monetary form, property,
especially as the ownership of land, was not capable of being
differentiated sufficiently. It remained, for example, the quasi-
necessary foundation of political power (feudalism). Only the
binary coding of the economy through money and a change of
the code having/not having to the code payment/non-payment
led to the complete functional differentiation of the economic
system.
Today, because of its monetary centralization, the economy is
a rigorously closed, circular, self-referentially constituted system
because it effects payments that presuppose the capacity for
making payments (thus the acquisition of money) and manages
this capacity. Thus money is a uniquely economic medium. It
cannot be introduced as input from nor transmitted as output
into the environment. Its exclusive task is to mediate system-
internal operations. Since the system can negate, these operations
consist of decisions. Payments qualify themselves against the
background of the possibility of not paying (or not being able
to pay). To pay or not to pay - that is, quite seriously, the
question by which the existence of the economy is determined.""
Code and programs are inevitably separated according to this
basic operation. The code exists because it makes a difference
whether someone has (the right to) certain sums of money or
not. Only someone who has a certain amount of money and can
part with it is in a position to pay because payment is the
transformation of having into not having. The same is true,
conversely, from the receiver's point of view. The code is the
condition of the system's being set in motion and keeping it
Economy 53
going and for the constitution of the system by events, in this

case payments. Such events (payments) are meaningless by
themselves if reasons cannot be provided for motivating their
performance for example, for satisfying needs or improving one's
position for future payments. In this respect the system must be
thought capable of learning, i.e., capable of reacting to changes
in itself and the environment. Criteria of correct behavior, i.e.,
programs have to be created for this. Needs cannot be programmed
directly. They occur and make their presence felt. But for the
system they are environmental givens. The system remains
dependent on a regulation of the operations internal to it, i.e.,
on a programming of payments themselves. This is accomplished
by prices.
Prices permit a rapid determination of whether payments are
right or not. This merely requires a quantitative comparison, and
because this is so easy the question arises whether the prices
themselves are right. So the programming requires a programming.
Earlier economic orders (cf. figure 9.1 above) tried to do this
with the theory of the just price. But in their case just meant
'corresponding to the market situation' while unjust meant
'interdiction of a profit that was not justified by this situation'.4
The fixing of prices simply to improve one's own solvency or
for commercial cunning was forbidden. Since code-values and
programs were not distinguished this would amount to greed
(pleonexy).
In the transition to modern society with its capitalistic economy
this restriction disappeared and was replaced by restrictions
internal to the economy. The legal system that interprets the
justification of prices as contract reacted to this by removing
restrictions on contractual freedom.5 But this immediately affects
only the operations of the legal system itself, for example, the
ability to sue for payments. For economic calculation, prices are
regulated by the economic process itself and require no external
regulation (for example, natural law or morality). Unices are
determined by what people will be willing to pay in the market
and this is determined by the money supply available^The theory
of the just price was therefore abandoned in the eighteenth
century when it became evident that the economic system itself
places restrictions on the profit motive and commercial cunning.
54 Economy
The system's coding and programming became a purely internal

matter, while the environment placed restrictions on the system
which were expressed within it as prices and price changes. Peace-
making and the peace-promoting function of trade is an often
repeated theme in the seventeenth century. But at the same time
the securing of peace was left to the state and the international
balance of power even if the economy could enforce the prevailing
market prices without having to bother itself with the preservation
of peace when it determines prices. Politics and the economy are
functionally differentiated systems. Therefore, if politics intervenes
in the process of price formation (which, as is known, happens
to a great extent) it transforms economic problems into political
ones. But the difference between the two systems remains.
A basic and very far-reaching feature of this system and its
form of resonance is commonly designated with the concepts
'market' and 'competition'. An adequate market theory does not
exist, even in economics. What is noticeable is a great amount
of differentiation among contexts of competition, exchange and
co-operation. Viewed historically and sociologically, this is a
highly unusual structure. At present, co-operation and exchange
normally do not occur among those who are in competition. This
makes it possible to keep the competition of interaction and
communication open among the competitors and to reduce it to
a mere calculation of the social dimension of everyone's respective
behavior. As long as the system determines itself through
competition it can spare itself the difficulties and ramifications
of the conjunction of interactions. But in doing so it has to
renounce the possibilities and certainties of control that
accompany direct communicative agreement (whether these turn
out to be positive or negative). The system reacts via 'contagion
sociale',6 via a quasi-simultaneous processing of expectations via-
a-vis the processing of the expectations of others. The 'double
contingency' embodied in this does not lead to the formation of
systems. Instead, it is freed to make decisions, under the conditions
of uncertainty, about the chances of success that are contingent
on the decisions of others.
One of the most remarkable consequences of this - that more
than anything else determines the resonance capacity of the
economic system is an extraordinary increase in tempo. The
Economy 55
system operates so fast that it is limited to observing events and

cannot integrate them through structures any longer.7 Every
intervention therefore acquires the character of an event, an
impulse, a provocation, a stimulation or destimulation of changes
in the system itself, and the unforeseeable effects of this continually
produce new impulses of the same type.
How, under these circumstances, a directive intervention is
possible within the organized complexity of the market is
something that would require a detailed investigation. Above all,
the problem exists in the money market, which varies in almost
complete disregard of the environment but influences all the other
markets. Yet the quantitative dimensions of this market -
fluctuations of up to several hundred billion dollars daily - give
a person pause.8 Whoever, in view of such phenomena, still
wants to appeal to an environmental ethics has first to be
concerned with the financial instruments of this ethics.
One has to admit that this (turbulence-emphasizing) description
of the system is by no means complete. This calls for a closer
look at how payments and the operative autopoiesis of the system
are actually disposed of. The economy can be called 'capitalistic'
only to the extent that it connects payments with the reproduction
of the capacity to make further payments, above all from the
point of view of the profitability of investments. Capital is
necessary because there is a time-lapse that has to be bridged
between any payment and the reproduction of the capacity to
make further payments. Revenues are not yet available when the
means of production are purchased (for example, in the case of
seed-corn) or are left unsold. The more available capital is, the
greater and more indirect are the relations of production and
consumption that can be included in the economy. But in the
final analysis this means that capital investment must be calculated
economically, i.e., rationalized in terms of the preservation,
reacquisition and increase of capital.
In a money economy (an inevitability in modern society) a
capitalistic self-control exists only as a possibility. If this is
ignored and, for political reasons for instance, unprofitable
investments are made then someone has to assume responsibility
for this. Payments are made without reproducing one's own
capacity for making further payments. In this, so to say, counter
56 Economy
to the money-cycle way, the incapacity for making further

payments is passed on to others and they are forced to make
unprofitable payments (perhaps in the form of taxes). They have
to regain their capacity to make further payments then in other
ways - perhaps through increased prices. In the same way, private
households are forced to pay for goods that are consumed
immediately and do not reproduce the capacity to make further
payments. So even the private household is removed from the
capitalistic sector of the economy and made incapable of
reproducing its capacity to make further payments if it does not
look for other modes of income, for instance through labor.
This 'double cycle' of the capacity and incapacity for making
payments results directly from the idea that an economic system
consists of payments. Payments themselves are twofold: they
produce in the payee the capacity for making further payments
and in the payer the incapacity for making further payments
(with this money, of course). But such individual events are
possible only in a dynamic system, i.e., only under the condition
that the capacity and incapacity for making further payments
can be transmitted or passed along. The metaphor of the 'cycle'
means nothing more than that, in any individual case, it has to
proceed in the hope of 'continuation' and that no system operation
can escape its inexorable law. The 'identity' of the capacity to
make payments is conceivable only on the levels of systems
but not in such a way that the very thing that one gives up is
returned (circulated) back to the payer. The metaphor of the
cycle represents the unity of the system, and this means the
autonomy of the system's autopoiesis. The reality resides in the
conditioned operations themselves.
Under such conditions the economic system has to look in two
different directions at the same time to provide for the preservation
and reproduction of the capacity to make payments: to the left
and to the right, as it were. On one hand, the issue is profitability
and on the other it is the supplying of the economic conditions
for the fulfillment of public duties and the provision of work.
The credit mechanism provides for a certain margin by creating
the capacity to make payments precisely where this would not
occur as a result of circulation alone.9 This can be steered to a
certain extent by central banks that are always in the position
Economy 57
of being able to make payments. But it is questionable whether

there are criteria for this in the system - apart from the intention
of normalizing the system's present perspectives of the future. 10
In any event, the mere necessity of the safeguard of a central
money supply (which is nothing more than the arbitrariness of
the transition from the incapacity to the capacity of making
payments by the central bank) is still not a criterion for its use.
Even theoretical or political guidance from equilibrium theory or
multivariable models of optimization is simply the elimination of
the tautology of economic self-reference: an interpretation of the
history of the system (i.e., data) so it can continue to be written.
More important than the structural metaphor of circularity is
the fact that, since payments and the regeneration of their
possibility are events, time has to be built into the system. This
means that the economy is continually concerned with gaining
time and forming capital so it can have time available at all
times. Thus the system develops its own future/past perspectives,
its own temporal horizons and temporal urgency. One cannot
simply assume that this system-time accords with the temporality
of the processes in the ecological or even the social environment
of the system. For this reason there is limited resonance capacity.
Even if, for example, fossil fuels deplete rapidly it may 'still not
yet' be profitable to switch to other forms of energy. This time-
loss is what concerned ecologists regret most of all. Even if the
gradual exhaustion of resources or a forthcoming election can be
significant to economic calculation decisions in the economy
are made according to its own conditions.
One might think that preserving the time-consuming double
cycle of the transmission of the capacity and incapacity for
making payments would be enough to keep the system busy.
Therefore, resonance for environmental questions is possible only
when the exposure to ecological dangers is brought into this
double cycle, whether this comes about because one sees a
possibility for making a profit in them, opens new markets,
produces new or transferred incentives to buy and especially
because one increases prices and forces them on the market.jlt
can also come about because one makes unproductive payments,
increases the incapacity for making further payments and passes
this on. The economy has to realize both possibilities, and it can
58 Economy
reinforce both of them when it puts more money into circulation.

Whether it can still fulfill the (patently fictive) expectations of
its own theory, for example, attain an equilibrium or optimize a
welfare function - is more than doubtful. It is truly astonishing
that, to a large extent, the constant (1) selecting of profitable
payments; (2) financing of public expenditures; and (3) providing
of labor succeed over a very wide range of goods and needs.
I Accordingly, the concept that the economic system produces
for itself of the ecological environment (as distinguished from
humans and society) is limited by the possibilities of adding its
own operations to it. In this sense, Dieter Bender defines
environment, 'as the totality of all naturally provided, non-
produced goods and services that provide streams of profit to
the individual participants in the process of production and
consumption'. 11 /Although this definition suggests a direction of
flow it includes the absorption of economic declines because the
economic system draws profits from this absorption too. But
right from the very beginning this definition is calculated towards
compatibility with internal economic operations and not to the
particularity of the environment. Moreover, through precipitous
equalization it conceals the largely typical problem for an
environmentally minded economy of separating the problems of
levels and amounts from those of allocation and deciding each
separately.12 None of this is a shortcoming or a constriction that
should be criticized. Instead, it is the condition for the system's
ability to steer itself internally according to its difference from
the environment. In the same way it also provides the limits of
possible resonance.
Only to the extent that the environment is brought into the
economy in this way and internalized with the help of quantitative
or profit calculi can there be economic motives for handling the
environment with care as the property theory of the physiocrats
had intended. Resonance to environmental data and events is
then regulated through prices and what influences them. On one
hand, prices are a critical instrument in the discovery of
environmental opportunities. When prices rise so do the oppor-
tunities for the increase of production, including the extraction
of material and energy from the environment. When prices fall
activities that are no longer profitable are discontinued. Small
Economy 59
profits stimulate production too, even if they are accompanied

by distant, unperceived (by the market) risks of catastrophic
consequences, and even if businesses felt a sense of responsibility
for such consequences it would still be rational economically to
leave these out of consideration. This comes under the much
discussed asymmetry of internal advantages and external liabilities.
But it also demonstrates that it is not always possible to solve
the problem through forcing the internalization of costs.
On the basis of theoretical models the economic theory
representing the position of the system's self-regulation provides
relative optimism for the possibilities of ecological adaption. 13 ;It
assumes that self-regulation is determined by prices alone and
that this enables the best possible distribution of information
about the environment. One could also say it is determined by the
prices that result from demand or, somewhat more aggressively, by
the prices that the market will allow. But this is, as mentioned
previously, a system-internal theory of system-internal processes
that, so far at least, believed production was expandable
according to a price-expressed need. Environmental conditions are
considered only as constraints on what is technically possible and
economically profitable at any time.
\ Within the conceptual framework of economic theory it is
possible to determine that the marginal utility and marginal costs
of the protection of the environment ought to balance out 14 and
thereby derive a principle that both makes the resqnance capacity
of the economic system possible as well as limits it. But enormous
problems of measurement and practical problems of attribution
remain. Above all, one must realize that the decisions of the
economic system never decide for the whole system. Instead, they
are guided by the 'internal' environment of the economic system,
namely, by the market. 15 But the latter is prefiltered to such an
extent that an all-encompassing economic decision rule directed
at the environment would find no application. It is also difficult
to imagine that prices could be so manipulated by an external,
politico-legal dissemination of data that the subsystems would
decide about production and consumption as if they were guided
by ecologico-economic marginal utilities. If such a need for
regulation is taken as the point of departure - and this might
very well be accepted today - then it suggests that the political
60 Economy
system fixes amounts (above all, the amount or level of acceptable

environmental pollution or the amount of the final consumption
of irreplaceable resources or even negative amounts in the form
of costs) and that the economic system concerns itself with an
optimal distribution and use of these amounts. 16 This seems
compatible with a market economy. But is it?
In order to take this under consideration and eventually to
solve it let us go back to the general code-paradox of the
economy, the paradox of scarcity. It states that the elimination
of scarcity through the appropriation of scarce goods increases
scarcity. This paradox is disguised (the invisible hand) for the
market through widespread economic success and especially
through growth. The so-called 'external costs' are used to
accomplish this. But when one establishes a difference between
amount decisions and allocation decisions this establishes, instead,
a different, functionally equivalent form of paradox elimination,
namely, a self-difference, a hierarchy. The market had served to
conceal the difference between these kinds of decisions. Through
the establishment of a hierarchy the difference becomes evident.
This also brings with it the typical problems of hierarchical
paradox elimination, and what Douglas Hofstadter calls 'tangled
hierarchies' or 'strange loops' 17 are encountered in many places.
One intends to operate on one level and unexpectedly the
operations are on another. Decisions about amounts interfere
with the allocation process and create compelling grounds for
changing the pretext of the amount due to the kind of allocation. 18
But this does not mean that this is bad or that it leads to insoluble
problems. No system is destroyed by logic. One has to realize,
however, that a different market-strategy is being used to eliminate
the paradox; one that, to a great extent, exposes structural
contradictions and decisions.19
Finally, we have to consider entirely different kinds of
theoretical ideas, those guided by input/output models. We have
to discover externalized costs and reincorporate them in the
economic analysis: We also have to reveal the environmental
consequences of economic activity and put them into a form that
allows decisions to be made. So one part of the economic
literature requires that the goals of the economic system should
be extended to the ecological side-effects of economic activity.20
Economy 61
In connection with this a distinction has to be made: the economic

system itself has no goals because, as a closed autopoietic system,
it is not directed toward any output. The best that can be said
is that production organizations observe environmental protection
as a secondary goal, especially when they are directed by managers
and when they do not perceive an urgent reason for a market-
price guided dividend policy beneficial to shareholdersijSimilarly,
a consumer - if ecologically-minded - might be willing to pay
more for environmentally safe soapl Behavior can be modified
in an ecologically desirable sense m the economy but not
without affecting production costs, and consequently taxes and
preferences for goods. 21 ! It may make very good sense to
speculate on ecological enlightenment, improved clarity of causal
connections and on 'changes of mind' or 'value shifts'. But it is
impossible to tell, at present, how such changes will work out
in the economy and what still-unknown side-effects they will
trigger. ;
This self-referential type of economic information processing
leads to transforming problems into costs. 22 They then become
parts of the calculation that decides whether it is economically
rational to make the corresponding payments or not. In this form
the system distinguishes between solvable and unsolvable (not
financable or financially unprofitable) problems. Both the results
and the prospects themselves remain problematic on the whole. 23
The results remain problematic because the rejection of the
payment of costs transfers the unsecured amount into the cycle
of the transmission of the incapacity to make further payments
and can overburden it. The prospects remain problematical
because the definition of the problem does not reveal all the
aspects of the problematic through the concept of costs and the
schema of paying/not paying. Just as with conditional legal
programs this is a matter of a specific technique for dealing with
higher structured complexity - of a very efficient, barely
improvable technique based on a one-sided selection of its
starting-point. Here as well as elsewhere the solutions to problems
are not without consequences. Therefore whether the economy
solves its problems or not problems remain for other domains of
society.
Because of the highly selective resonance of its object domain
62 Economy
a theory of the interconnection of price, costs and production is

not capable of judging our society's exposure to ecological
dangers. It is not even capable of giving political advice on this
question. But it provides a good idea of the self-determined
resonance capacity of the economic system and its self-referential
closure. Formulated as a principle it says that whatever does not
work economically does not work economically. This is correct
mutatis mutandis for all the other systems too, even for politics.
For the economy, the question will always be with which prices
will the capacity to make further payments be passed on and
how can the incapacity to make further payments be transferred.
This is the only mechanism that combines autopoiesis, resonance
to the environment, continuation of production and the inclusion
of an unintelligible, noisy environment in this process.
The key to the ecological problem,, as far as the economy is
concerned, resides in the language of prices. This language filters
in advance everything that occurs in the economy when prices
change or do not change. The economy cannot react to
disturbances that are not expressed in this language in any
event, not with the intact structure of a differentiated function-
system of society. The alternative is the destruction of the money
economy with unforeseen consequences for modern society. I
This structural restriction to prices is, however, not only a
disadvantage, not only a rejection of other possibilities; it
guarantees that the problem, if it can be expressed in prices, must
be also processed in the system. As always, the reduction and
increase of complexity work hand in hand, and it is difficult to
see how, without such a restriction of resonance capacity, the
extensively compartmentalized, alternativeless possibilities of
reaction to environmental stimuli could be produced. Otherwise
the situation turns out to be just like that of a woodcutter who
sooner or later has to find out that there are no more trees to
be cut down.
10
Law
In the contemporary ecopolitical discussion the contrast between

the language of prices and the language of norms is as striking
as it is disarmingly simple.1 This corresponds to the long-standing
distinction of society and state and suggests a simple alternative
in the reaction to ecological dangers. This leads immediately to
the following consequence: whatever does not fit within the
language of prices has to be expressed in the language of norms.
Whatever the economy does not bring about on its own has to
be accomplished by politics with the help of its legal instrument.
Finally, the ecological problematic runs up against a residual
political responsibility that unexpectedly becomes the all-consum-
ing responsibility of a constant state of vigilance.
This alternative is formulated much too simply and leads in
the final analysis to the fact that more is required from politics
than it can perform. It leads not only to avoidable disappointments
but also to an overburdening of the political system with
unfulfillable demands and it is misleading. Politics is then plunged
into verbal debates. It is forced to offer hasty, false solutions,
defer problems or try to gain time, which inevitably leads to
radical disappointments with it instead of the realization that
this system is capable of resonance only within the context of
the frequencies of its own autopoiesis.
We had already rejected the conceptual conditions of this
account because it tends toward over politization and conse-
quently to political fiasco. We replaced it with the argument of
a functionally differentiated society. Above all, this means that
64 Law
politics as well as the economy are only subsystems of society

and are not society itself. It also means that the function systems
of politics and law have to be distinguished more clearly than
usual. Admittedly, law stands in close relation to politics since
law-making typically requires previous political agreement. It is
a system that is sensible to political situations and, as a result,
singularly capable of resonance. But at the same time it is a
closed system that can create law only on the basis of law, norms
only on the basis of norms and through its court apparatus makes
sure that this condition of its autopoiesis is observed.2
The legal system receives its autopoiesis through coding the
difference between what is legal and what is illegal. No other
system operates according to this code. This binary coding
of the legal system creates the assurance that if a person is in
the right then the force of the law is behind him or her.
Uncertainty about the law exists only in a form that can, in
principle, be rectified, namely, in reference to decisions that can
be made in the legal system itself. This assurance is attainable
within society only if the legal system alone assesses what is in
accordance with the law and what is not. Besides, only one
system of society may use this code,3 and since no other system
uses it, the legal system cannot import or export what is legal
or what is not. It tells only itself and not the rest of society (its
social environment) what is legal or not in any case. The
incontestable social effects of law rest on the occurrence of this
within the legal system.4
Since in any case only one of these two possibilities exists for
the legal system there is no third possibility the schema
contains a complete description of the world.5 For the performance
of operations and the reproduction of standards all that remains
to be determined is whether any particular case is in accordance
with the law or not. In other words, standard legal programs
must be provided that fix the conditions of legally correct
decision-making. These can be found in laws or ordinances,
statutes or procedural rules, in judicial rulings or contractual
agreements. On the level of programming the system is closed
and open at the same time. It is closed because a norm can be
obtained only from norms themselves (no matter how logical the
mode of inference or arguments may be judged). It is open
Law 65
because cognitive viewpoints also play a role in this. Cognition

is required both for the system's environmental orientation as
well as for its own orientation, both for the determination of the
empirical conditions of the application of norms as well as for
judging the adequacy of or the need to change the norms
themselves. The system operates completely 'open' for environ-
mental conditions and their possible change. In other words, it
can learn. But first the autopoiesis of the system has to be used,
i.e., it has to proceed according to the difference of what is lawful
and what is not according to legal programs - even if only
because otherwise it would never be possible to know that a
legal process is at issue.
Thus the code is also the autocatalytic factor driving the system,
both in terms of its need for supplementation and in the formation
of highly complex program structures. As internal structures these
programs work only on the internal processes of the system. The
system can know its environment only through them - even if
this is as a disturbing noise that can be corrected only by a
modification of the programs. This corresponds to the social
function, i.e., the socially internal function, of law to take
precaution against conflict and also to provide for stable
expectations in the case of disappointment. In this case the social
system's environment comes into consideration at best as the
occasion for conflicts (and consequently as the occasion of
precaution against conflict). The environment disturbs the smooth,
customary fulfillment of expectations through 'noise', and in
many ways property functions to transform this eventuality into
a partially economic and partially legal problem.
Therefore within the legal system the basic figures of legal
thought represent a social desire for order (and in such a self-
evident way that it almost goes unnoticed that this is a structural
selection, i.e., a selection out of a plurality of possible ideas of
order). This holds for all figures of reciprocity, exchange and the
distribution and generalization of the relevant valid conditions,
including the categorical imperative. It also holds for the semantics
of freedom and the restriction of freedom ever since the bourgeois
revolution. Legal ideas of order apply to socially internal relations,
and it would surprise jurists to think that someone thought it
necessary to advise them about this.
66 Law
But it is precisely the ecological debate that provides the

occasion for this. For even law reacts only in its own system-
specific way to the exposure to environmental dangers, and there
is no guarantee in advance of an adequate proportion or a
causally successful reaction to the danger. The form in which
law programs its code, i.e., translates it into the conditions of
correct action, is fixed in the future-perfect tense. It imagines an
action as completed in the future. This is the basis of the profound
relation of law to freedom. Events occur on their own. The law
simply states how they are treated if they occur, therefore the
basic form of law remains the conditional program, no matter
how much talk there is in environmental law about 'goals'. The
law can never attempt to capture all causal factors successfully
or to determine all processes legally. Of course, one can decide
for reasons of a political goal to make laws and to justify law-
making as a means to the goal. But it would be a mistake to see
a causal statement in this. Whatever degree of probability the
law contributes to the attainment of the political goal is a question
that depends on other factors. Legally, this is a matter of the
clarity and univocality of the answer to the question of what
would happen in the case of a conflict and of the possibility of
forming expectations for this.
This form of legal regulation can be proclaimed as the
protection of freedom, indeed as the promise of freedom. Viewed
more prudently it is a matter of a specific technique for dealing
with highly structured complexity. In practice this technique
requires an endless, circular re-editing of the law: the assumption
is that something will happen, but how it will happen and what
its consequences will be has to be awaited. When these
consequences begin to reveal themselves they can be perceived as
problems and provide an occasion for new regulations in law
itself as well as in politics. Unforeseeable consequences will also
occur and it will be impossible to determine if and to what extent
they apply to that regulation. Again, this means an occasion for
new regulation, waiting, new consequences, new problems, new
regulation and so on. Presumed foresight is, therefore, an
important auxiliary motive that keeps the process going. It results
in an extreme complexity that is comprehensible in the legal
system only historically.
Law 67
Even in handling ecological problems law is bound to its own

function, its corresponding technique for reducing complexity
and, especially, to its own distinctions. All of these co-operate in
programming the reduction of complexity. One can see this even
more clearly as the machinery of environmental law is already
in full swing.6 Thereby, problems result both from the historical
presence of law (without which law would not exist, could not
even react) as well as from its sociofunctional specification.
When the ecological problem was first raised not a single area
existed for law as unexplored territory to be formed and covered
with a net of new regulations. There is only one legal system of
society and this is always present. Law itself is always completely
formulated and can only be changed. Consequently, 'environmen-
tal law', accompanied by new kinds of formulations of the problem
(or with easily revisable formulations), cuts into customary legal
domains like area planning law, legal competence, police law,
business law, tax law and constitutional law. 7 As long as such a
co-ordination does not succeed innovations remain abstract. They
remain merely problematic ideas. The presently topical 'testing
of environmental compatibility', for instance, looks like a new
concept, like a legal innovation. But it is nothing more than the
idea of a better harmonization and possible adaptation of existing
regulations to different legal domains, understood and reorganized
to cut across existing systematizations. All further development
of law is bound to corresponding points of departure and has to
take into consideration that other regulations remain in effect.
Otherwise the idea of a relatively consistent, non-contradictory
treatment of the legal code would have to be abandoned.
It is equally apparent that law can be developed only as a
social regulative. The valid guiding principles for this may change,
for example, from reciprocity through contract to the weighing
of interests and the protection of trust or through numerous new
legitimizations for the rights to and limits on freedom (whereby
the need for limits can lead to the creation of freedom just as
freedoms can reveal the necessity for their limits). This is why
the distinction of freedom-rights and legal coercion have become
the dominant formula for legal discussion about the environment,8
although the distinction itself does not refer to the environment
as such.9 We are far from acknowledging rights 'to the
68 Law
environment' vis-a-vis society, far from granting rights to trees

or for punishing dioxin for its toxicity by burning it. Instead, we
face the problem that because of a widely compartmentalized
system of subjective rights, we not only individualize the
disposition over a domain of our own (our 'own house') but at
the same time, under changed conditions, we leave the estimation
of one's own interests and the willingness to assume risks,
including the willingness to sell this willingness, to the bearer of
subjective rights. Of course, the law can make counter-precautions
through limits on freedom. But the question is, what does this
granting and limiting - under the control of the constitution and
normal laws - produce within society when it tries to create
resonance to the exposure to ecological dangers?
If the reaction to environmental problems must be met with
an internally goal-directed conceptual apparatus then an essential
incongruence of legal categorization is to be expected. Despite
the proven learning capacity of the legal system, its laws and its
dogmatic theories, this incongruence cannot be rectified, for the
regulation of communication in the system is something different
from its reactions to changes in the environment. Like every
system, the legal one is capable of resonance only in accordance
with its own structures.
A sociologically more important indicator of this is that the
component of arbitrariness in environmentally related legal
decisions increases significantly. This holds in at least three
respects:
1 for the necessity of defining marginal values, thresholds and

units of measure for which the environment supplies no
determination;
2 for determining the system's willingness to assume or tolerate

risks when their transgression brings with it protective
measures and eventually those that discount the costs or even
proscriptions;
3 for fixing preferences regarding the extremely diverse conse-

quences of environmental changes that are in large measure
blocked, scattered and concealed by the price mechanism or
even for the protection of concerned interests that cannot be
Law 69
co-ordinated immediately because of the indirectness and

opacity of causal relations.
All of these are by no means new types of problems for the

legal system. But they acquire a new intensity and scope when a
new ecological consciousness of the problem begins to affect the
law. Natural law ceases to function precisely where it concerns
nature, and even consensus (a kind of ersatz natural law) seems
unattainable. At the same time it becomes questionable whether
problems of this kind can be analysed, factored and ultimately
solved satisfactorily with the standard legal method of handling
cases.
Along with an increase in arbitrariness therefore, we find
apparently contradictory, empty formula-like obscurities. These
leave all decision problems open and merely give the impression
that something is happening, at least on the verbal level. We will
cite two examples of this without holding it against their author
since they are typical of many others. 'Insofar as there is some
room given to individual policy-makers for development it is
recommended to concede to environmental protection a certain
primacy'. Also, 'Their [the administration's, N.L.] task is to bring
about a balance between the general public interests and individual
concerns and to reconcile these with the requirements of
environmental protection'. 10 Formulas of 'equalizing', 'balancing'
or 'proportionality' can be achieved only arbitrarily. If the law
has to resort to such formulas then a technically informed
arbitrariness is not the worst solution. It is just not a specifically
legal one. 11
The transition from empty formulas to arbitrary distinctions
is what makes it probable that any argumentative beginning in
the usual sense of legal practice reintroduces these problems in
more manifold and magnified ways. Such possibilities make their
presence felt most clearly in dealing with risks. The standard rule
of maximizing anticipated profit with a minimum of risk fails. 12
Anyway, it works only in the few cases where no uncertainty
exists regarding the probabilities. As a general principle it is too
risky. 13 Empirical research has shown that the willingness to take
risks very often includes individual personalities, social systems,
circumstances and previous experiences. 14 Therefore, maintaining
70 Law
any threshold of tolerance can be achieved only arbitrarily. Very

often risk is valued highly and sought. 15 Moreover, the subjective
factor in the estimation of improbability increases with an increase
in improbability. 16 This is a problem that involves complexity
too: distinct willingnesses to take risks cannot be added together
and, to a great extent, depend on voluntariness so that they
would be changed, if not nullified, if they were required by law.
Regulation can occur but only arbitrarily and not without
changing the consensus situation by the regulation itself. This
means that centrally ascertained and established risk estimations
and tolerances are unavoidable. 17 They cannot be based on
consensus but have to be exacted, which then automatically
reduces the willingness to assume risks.
Of course, the jurist never raises the question of how people
come to estimate risks. Empirical research about risks is just as
irrelevant as models of rational decision-making. One has to
make decisions in accordance with maxims that one has
ascertained personally. The fact that any residual risk ought not
to be avoided, i.e., has to be accepted, is therefore acknowledged
in practice as well as in theory. The justifications for this, of
course, are not of much help. 'Social adequacy' cannot be inferred
from 'unavoidability'. But even the appeal to an ethics that, for
all practical purposes, is equally helpless, is no longer of any
assistance since this too has to turn to reason and reason must
then 'help itself'. 18
Another consideration that remains unresolved concerns the
'balancing of goods' as the expression of ethical responsibility.19
With this, the function of law to safeguard expectations in cases
of conflict is not yet fulfilled. The rule applies to decisions in
individual cases, i.e., it merely states that the courts will make
decisions only after a careful examination of the situation. So,
in the final analysis, the process of argumentation merely
distributes the self-validation of legal decisions over many stages.
One could attempt to determine the probability of the
occurrence of desired or undesired consequences. In this way
legal regulation can extend to the choice of test procedures and
the statistical control of the probability of the positive or negative
determination of errors. 20 If a satisfactory determination of
probability can be provided then uncertainty would no longer
Law 71
exist in the language of decision theory, only risk in the narrow

sense.21 But this would not solve the problem of the acceptance
of risks, which would only be formulated more clearly. The
advantage of consensus that exists when everyone begins from a
different estimation of probability or when the entire question is
left indeterminate would no longer apply. This would result in
an unattainable agreement concerning whether positive or negative
risks with a determinate probability are acceptable or not.
It does not even help if this question is made to depend on
'trade-offs' or on the level of expected profit or loss. First of all,
no unanimity will be attainable in their estimation and secondly,
profits and losses are distributed unequally in society.
One may also consider testing the readiness to assume risks in
the market. At present there exists a clear inclination to proceed
in this direction.22 This happens when there are sufficiently
localizable connections between profits and possible losses - for
example, when in the case of childrens' pajamas with a high
resistance to fire one cannot exclude the possibility that they also
may be cancer-causing. Then a duty to reveal the risk may be
sufficient. But for diffuse risks that are supplied indirectly through
the environment, other forms have to be found that are directed
toward the economic decision of the producer. Taken to its
logical conclusion, this would mean that not only those who
create risks for others should be made to pay for them or be
obliged to provide insurance against them but also that those
who live in danger must receive compensation for this that the
price of land in the flight paths of airports or in the vicinity of
nuclear plants does not decrease but is kept constant or increases
in compensation for living in constant anxiety. 23
It is already evident that economic limits on the provision
against risks of fatality are generally accepted for others. But it
is hard to accept that one's own risks of fatality are accepted
against payment. If ethical and eventually legal decisions diverge
here then this is simply because in the latter case a decision (i.e.,
freedom), is demanded of the candidate of fatality. Moreover,
such freedom would be economically exploitable vis-a-vis passive
exposure. One could negotiate and increase the price. To the
extent that the willingness to accept risks is indemnified the
aversion to risks would be profitable - at least as a position of
72 Law
negotiation. In addition, this solution presupposes an antecedent

identifiability (for oneself and for others) of winners and losers
that is not guaranteed in most ecological risks because they are
distributed too widely and indeterminately. Finally, the legal
system is faced with the question of whether it is justifiable to
hold someone who has agreed to accept a risk to a preceding
agreement in the case of catastrophe - 'Tu I'a voulu, Georges
Dandin\ If not, then should not the possibility of being released
from the agreement be included in the calculation from the
beginning?
These problems - accompanying the very first steps in problem
management are added to those connected with the analysis,
operationalization and factorization of problems. Included in
these are disputes about the adequacy of measurement procedures,
the power of experiments to provide evidence, and indispensable
variables in simulation models. The problems of risk estimation
will have to reckon with information that is constantly new in
addition to changes in preferences. All that is needed is an atomic
accident of the most trivial kind - an<4 everything will have to
be decided anew. 24 Further research may, and most likely will,
demonstrate with a high degree of probability that when we
apply rigorous standards we know less than we thought.
Additionally, we have to mention a problem of great practical
importance that only touches our theme marginally. In the
introduction of new or the exclusion of established technologies
there are always direct social risks, i.e., direct social consequences
and effects that cannot be assessed precisely and are triggered
not by causes in the environment but from within society itself. 25
Here too in the case of marginal risks we find subjectivisms
in assessments and the probability of improbable reactions as it
were.
What does a concern with a 'rational' solution mean in view
of such problems, especially where 'rational' means 'capable of
or requiring consensus'? The structure of reason was directed at
socially internal problems; developed towards problems of social
agreement. One cannot ignore this for ecological problems
either, especially when this pertains to social reactions. But the
problematic does not reside in the agreement. It resides in the
still-uncontrollable relations of system/environment. Accordingly,
Law 73
the classical model of finding a political consensus fails. Neither

a liberal theory that would like to view solutions as the undisputed
function of private decisions, nor a collectivist theory that thinks
it will always know what the people want, offer convincing
answers.
The typical jurist is satisfied by the idea that such questions
'have to be decided politically'. The recommended 'practical
regulations' are only a variant of this idea. The third value that
is excluded from the code of legal and illegal, i.e., what for the
time being is neither legal nor illegal, appears in the legal system
as politics. Thus the legal legitimization of political decision-
making leads to the reintroduction of the excluded third value
into the system. In this way the legal system makes use of a
constitution and democratic legitimization to deceive itself that
politics can handle problems better than law and that all
arbitrariness can be transferred into this system for appropriate
treatment and reintroduced as a legal norm. This stunning logical
achievement of including the excluded third possibility may be
admirable, 26 but it leads the political system, on one hand, to
view law as its own instrument of implementation, 27 and on the
other to decisions within the legal system that are not decided
in a specifically legal way but are determined arbitrarily.
A closer analysis of these particular components of arbitrariness,
even if they appear only within environmental law, show very
quickly that there is no recourse to a 'nature of the matter' nor
to a basic consensus of all those who think rationally and legally.
This holds for all three decision problems mentioned above. The
threshold values to be determined do not find a secure basis in
nature. Ecological problems are simply too complex, interde-
pendent, circumstantial, unpredictable, determined by the 'dissi-
pative structures' of thermodynamic systems, the abrupt disturb-
ances of stability (catastrophes) and similar structural changes,
for this. The acceptances of risks - research in decision theory
has come to this conclusion are subjectively so different that
consensus cannot even be reached in the case of assessments that
agree. In addition, the scope and opacity of the causal connections
imparted by the environment makes every value consensus trivial.
The earlier rule of generalized reciprocity, the 'scratch my back
and I'll scratch yours', and the categorical imperative fail and
74 Law
appear in the historical context simply as socially internal maxims.

The same holds for the late rational-law attempts at developing
rules for establishing a hypothetical consensus, procedural consen-
sus and norms where consensus would be granted when someone
behaved freely in a way that implied the recognition of these
norms. 28 All this assumes that the problem has its roots in society
and therefore can be solved in the social dimension. But the
inclusion of the environment of the society in the genesis of social
problems changes these problems essentially. Only a preventative
consensus is attainable: an abstract agreement about preventing
all possible damage in so far as the costs of prevention do not
have to be accepted. In the meantime, cars race along the streets,
lungs fill with tobacco-smoke, people borrow money, get married
and risk criminal prosecution for tax fraud or antisocial behavior 29
- as if to prove that life without risks and without the assertion
of highly individual preferences is of little value.
Under these circumstances the precipitation of political activity,
social solidarity and the legal solution of environmental problems
remains as abstract as it is inconsequential. A legal categorization
of precepts for environmental concern can only be worked into
the law, if at all, with other concepts. Dogmatic legal learning
processes are slow and need decades, if not centuries, to boil
down case experiences into concepts and maxims and to
reformulate the law in terms of these. In addition, they assume,
for whatever reason, that courts can be used to do this. In the
case of environmental law this seems to be so only to a relatively
small degree, if we measure it by the collection of regulations
and the amount of literature. 30 The law is created for bureaucratic
handling and seems to be designed for this from its mixture of
arbitrary determinations and vague, empty formulas. Similarly,
the administration of justice develops in connection with very
heterogeneous areas of specialization. It remains to be seen
whether this amounts to encompassing categorizations or even
to the development of specifically legal modes of argument and
justification. For the time being it is noteworthy that the legal
system reacts to the desideratum of an environmental law with
a considerable increase and complication of the regulation
apparatus. Through the co-operation of the political and the legal
systems a resurgence of norms has appeared at their boundary.
Law 75
The political system finds itself in the need of having to profess

and to cope with the desire to decrease [Entrechtlicbung] and
increase [Verrechtlichung] the scope of laws at the same time.
Finally, a last problem concerns the enforcement of law, its
execution and the effective prevention of exceptions. Here research
is concerned, on one hand, with answering the universal complaint
about problems of enforcement in obviously insignificant and
self-evident difficulties.31 More important, however because of
the increasing attention given to environmental questions - is the
appearance of new modes of enforcement that are difficult to fit
into the law. These can lead to the reassessment of views
and experiences and, at present, are identifiable only through
tendencies of distorting existing institutions. On one hand, this
holds true for 'private' efforts at seeing to it that public law is
enforced which have to operate without the protection of
subjective rights and are relegated to supplying information that
forces public authorities to intervene.32 On the other, enforceable
law at present serves the administration largely as a negotiating
position from which it can, in part, attain non-enforceable
concessions, relinquish severity in enforcement and, in turn, put
goals back into a gray zone of legality.33 Obviously both
exceptions to traditional legal structures can collide with each
other when private legal pursuants cancel out or highlight legally
problematic aspects of agreements. It may be that both of these
new types of enforcement techniques can be brought into an
equilibrium of 'checks and balances' and that courts can develop
criteria for this. For the time being only a tendency to concede
to administrations a greater scope in making judgements is
noticeable. 34 One can observe, therefore, that ecological communi-
cation deforms classical structures of the legal system, and how
it does this, on more than just the level of the content of norms.
10
Science
Perhaps we have been barking up the wrong tree. Perhaps it is

not the economy or law but science or politics that is responsible
for environmental questions in the differentiated system of society.
Since politics repeatedly pins its hopes on science and establishes
programs for promoting research and technological developments
we will turn our attention now to the scientific system to examine
its resonance capacity.
We are not concerned here with questioning the productivity
of the natural sciences, or with examining and making a wholesale
judgement about them. As before, we are interested solely in the
systems-theoretical question of what is decided about the
resonance capacity of function systems when they are differen-
tiated through a specific code and when their coding and
programming are distinguished according to this.
The difference between true and false is what matters for
science's code.1 Research programs are usually called theories.
Even for these the differentiation of coding and programming,
of truth-values and theories has the expected consequences: the
concept of hierarchy as the form in which unity and rational
connection are presented is abandoned and a differentiation of
the scientific system into disciplines and subdisciplines takes its
place. 2 After this, any possibility of determining the position of
knowledge in reference to the unity of the system is relinquished.
The disciplines work within a loose, expandable, theoretically
(i.e., research-programmatically) non-integratable association that
can be developed further, if necessary, through splitting-off,
k.
Science 77
through subdivisions and new formations. The unsatisfied require-

ments of reflecting the unity of knowledge and its conditions of
possibility resulting from this reflection converge in a new
kind of theory of reflection of the scientific system, into an
epistemology. The more pronounced the need becomes to
distinguish between everyday and scientific knowledge the more
the unity of the scientific system is reflected in this system
boundary, the more exact the theory of reflection of the scientific
system becomes. If a message about the meaning of the unity of
the difference between true and false is ever to be found (see the
position of the question mark in figure 2, chapter 9) then this is
where it will be.
The standard humanist critique of science is directed at the
level of programs, for example, in the impressive later work of
Husserl, at mathematical idealizations, the horizons of the life-
world and at the loss of contact with the subjectively meaningful
acts of consciousness.3 Accordingly, the historical European
specialization in science amounts to a loss of meaning. This
neither raises nor answers the question of the unity of the code,
of the unity of the difference of truth and falsity. But the 'loss
of meaning' (which is undoubtedly meaningful, or so many books
could not have been written about it) can be explained by a
differentiation of function systems, especially with the differen-
tiation of a function system for scientific research. In this way
we reach a position that suggests applying the unity of the code
to the unity of the system that uses it. 4 Separate coding enables
the system to be differentiated just as the system can record,
standardize and subject the code to practical verification.
This information, however, remains very formal if it is not
concretized in terms of the particular characteristics of the
scientific code. On this score there are two arguments:
1 The code of scientific truth and falsity is directed specifically

toward a communicative processing of experience, i.e., of
selections that are not attributed to the communicators
themselves. The intervention of personal qualities and circum-
stances are treated as disturbing noises and, like other
'accidents', eliminated if they do not lead to valuable discoveries
of truths or falsities.5 Historically this means that the
78 Science
explanation of the causes of errors an important part of

former epistemologies is de-anthropologized, freed from
connotations like original sin, cognitive faculties and ideologi-
cal blindness and placed within the structures of the scientific
system itself, i.e., in its theories and methods. Errors are now
mistakes, often productive mistakes, in the programming of
communication and can be discovered and eliminated by
coded operations.6 The code remains universal. It applies to
everything that can be experienced, even to action. But it does
not serve the communication that wants to carry out, cause
or even recommend an action. It does not serve to select
action. From this point of view truth and falsity can also be
conceived as a symbolically generalized medium of communi-
cation that serves specifically to reproduce the improbability
of similar experience and that leaves the steering of action to
other media.7 Above all, it is this reduction to co-ordinated
experience that differentiates the scientific code from morality
and religion. After all, religion is not merely the experience
of God, so it is reconciled to doubt. One has to pray for
miracles.8
2 The scientific code of truth and falsity is directed specifically

towards the acquisition of new scientific knowledge.9 The
mere recording, preserving and retrieving of knowledge has
required little human effort ever since the invention of printing.
In any event, to take care of this is not an unusual, improbable
performance and would not require the code of a special
communication medium. What is new has to be freed from
the suspicion of being an exception or being false. This occurs
through an improbable, culturally historical preference for
innovation, indeed for curiosity (curiositas), that has to be
tested and standardized itself, i.e., find its boundaries within
itself (resonance) and not in its objects. 10 Everything can come
under consideration. Similarly, scientific analysis does not
serve to solve problems but to multiply them. It begins from
problems that are solved or from problems having a chance
of solution and inquires further.
Important features of modern science can be explained and
allocated with the help of both these specifications, namely:
Science 79
3 The way that science goes about its task rests on a

differentiation of theory and method. Theories (research
programs resulting from research programs) externalize the
internal results of scientific work, i.e., apply them to the world
that can be experienced by everyone. On the other hand,
methods apply the code, i.e., make sure that the results can
be distributed according to the values 'true' and 'false'. 11 In
this case the test procedures of decision theory, game theory
and statistics provide only provisional certainties.12 But even
this is a reflection of the difference of theory and method. So
theories represent the openness, while methods - through the
exclusion of third values - the closure of systems. Their
distinction itself has an exclusively internal meaning for the
system: it refers only to the system's own operations.
4 After the system worked for several centuries under these

conditions it became clear where it was leading. This is
something that idealization, mathematization, abstraction, etc.
do not describe adequately. It concerns the increase in
the capacity of decomposition and recombination, a new
formulation of knowledge as the product of analysis and
synthesis. In this case analysis is what is most important
because the further decomposition of the visible world into
still further decomposable molecules and atoms, into genetic
structures of life or even into the sequence human/role/action/
action-components as elementary units of systems uncovers
an enormous potential for recombination. At the same time,
however, this potential for recombination is something with
which science overburdens itself.13 Parsons's theory of the
general action-system presents this situation for the social
sciences.
5 For methodological and theoretical progress this development

moves the problem of the observation, description and
explanation of facts with structured complexity - quite apart
from the ways classical epistemologies proceeded - to the
center of interest.14 From the concept of the 'black box' to
the radical distinction of operation and observation, the theory
of self-referentially closed systems, the auto-logic of self-
observation and self-description and corresponding theories
80 Science
of intervention research encounters the resistance of reality

forcing it to understand itself in terms of structured complexity.
It follows only then that complexity is interpreted as the
measure of system-ignorance and as the prerequisite for many,
theoretically non-integratable descriptions.15 This means that
science has to understand itself as a system that observes
observing systems (cf. above, chapter 5). At the same time, it
realizes that it, too, is nothing more than an observing system
that depends on its own structures. It encounters itself as a
complex system that recalculates its calculations with a view
towards self-provoked disturbances from the environment.16
6 The problem of structured complexity is not the ultimate

problem (if we could solve it) that would clear the way for
omniscience. Omniscience is not only factually impossible but
also logically impossible (this is not to say theologically
impossible!) because it would have to include itself within
itself. Analogously, a system that tries to steer itself according
to its own complexity is only hypercomplex. It employs some
of its operations to reduce complexity and others to observe
and describe that reduction and how it occurs. Thus it
generates a second complexity that encompasses the first, plus
other things. This is unavoidable if our theme of social
resonance to the exposure to ecological dangers is to become
the theme of scientific research. If so, then science describes
all of society, including itself, as a subsystem of society. The
description of description of description - ad infinitum, except
in a self-description of science that applies to itself whatever
it postulates for all systems: limited resonance according to
its own frequencies and eventually according to binary coding.
This is like Baron Miinchhausen's story of his adventure in
the swamp where he pulls himself out by his own hair - but
with the possibility of seeing how the others do it!
Even the scientific system that concerns itself (with the help of
this code and with its corresponding reductions) with themes like
the environment, owes the system's openness and learning capacity
to the closure of its autopoietic self-reproduction. Even the
scientific system finds itself reduced to a self-structured resonance;
Science 81
otherwise it would not be in the position to recognize information

as scientifically relevant to classify it as true or false and to
accord it a self-transcending relevance through inclusion in
theoretical contexts.
The possibility of recognizing these limitations resides in forcing
them back into paradoxes, i.e., into undecidability. Thus, for
example, all theories are directions for comparisons. The more
different the thing to be compared, the more powerful the
theory. To this extent science is the pursuit of the possibilities of
treating as the same, things that are different. Its theories dissolve
the paradox although they work with it and transform it into
normal research.
Alternatively, science proceeds as the decomposition and
recombination of reality. The further the decomposition is carried
the more difficult, the more effective, the more 'catastrophic' the
recombination. Possible examples of this include the physics of
subatomic relations, modern genetics and the assumption that
human behavior is the result of processes of socialization whose
foundations in stratification have to be eliminated. Decomposition
and recombination are carried out as a unity and just like the
process of comparison, this unity is the condition of the
appearance of new knowledge, i.e., of the acquisition of
knowledge. Thereby, what is not intended has to be intended
too: the increasing probability of uncontrollable recombinations.
An unavoidable concomitant of this is the classic paradox of
ceteris paribus. The assumption of ceteris paribus is the condition
of isolating the objects of research, but just like the presuppositions
of model-formation it is a consciously false assumption.17 Only
through false assumptions can true knowledge be attained.18
Since the latter problem has central importance for ecological
research we would like to spend some time with it. Ecological
research usually speaks of ecosystems. This concept would be
appropriate, however, only if external boundaries could be
supplied. But this is not the case. 19 It does not even help to define
systems by means of self-regulation instead of by boundaries.20
Self-regulation presupposes system boundaries. If the system-
premiss is abandoned and if ecological problems are not defined
as internal problems of an encompassing system there still remains
the possibility of working with the Simon!Ando-theorem of near
82 Science
(almost complete) decomposability.21 This provides a somewhat

clearer overview of the limitations. One, then, can ignore
environmental influences on the sector under research for relatively
short periods of time, or the internal relations within its
component parts for longer periods. But one cannot ignore the
influences on the growth of the component parts themselves or
their other environmental relations. Even if one takes complete
decomposability as the starting-point and assumes that a corre-
sponding clarity of environmental differentiations reduces the
internal problems of the scientific system, the theorem still
demands the consideration of the objective and temporal limi-
tations of the horizons of research. Considered methodologically,
the theorem is a correlate of the problem of the opacity of
structured complexity and, viewed theoretically, a variant of the
paradox that truths can be obtained only through the allowance
of falsities. It demonstrates the way in which abstractly formulated
problems are transformed into concrete research plans and then
create resonance within science.
These very abstract statements already indicate clearly enough
the type of resonance capacity possessed by scientific research
and its boundaries. As never before, an almost endless capacity
for resolution has revealed unbounded domains of possibility to
society. Science produces a transparent world that, wherever it
concentrates, reflects itself and transforms the transparency into
access to something new. Imagination is given wings, new kinds
of combinations are conceivable - whether as technical artefacts
or as their unwanted, perhaps catastrophic side-effects. Everything
that is possible and everything that is, is selection. But only
selection is capable of being rational. So our question is, how
can this rationality be rational if it has to select one out of an
astronomical number of other combination-possibilities? This is
not the world-picture of an idealizing and quantitative mathemat-
ics that was deplored by Husserl. Nor is it the world-picture of
technical instrumentality deplored by Habermas. Instead, this is
a world sinking both inwardly and outwardly into the void, a
world that can grasp only itself but can still change everything
graspable, a world unsuited for social orientation. If we are
permitted to extrapolate from research development so far, future
research will not reduce this picture to what is concrete and
Science 83
tangible but will reproduce it everywhere it sets about its task.

This makes an ultimate foundation of rationality unattainable.22
At best the concept of rationality can be revised and adapted to
this situation.
There are also plenty of counter-philosophies. One can extol
moral responsibility for side-effects or emphasize the necessity of
doing something. One can say that despite all else we are still
doing well 'in a life-world sense'. But these are reactions of
defiance: defense semantics that, if they attain communicative
currency, are open to observation and then break down as a
result.
Obviously, society as a whole neither wants nor is in the
position to assume the scientific world-picture. This is and
remains a mere implication of research. What science really
exports is the consciousness of selection and technology: the
consciousness of selection in reference to still-indeterminate
recombination possibilities and technology as already determinate
and realizable. In this way, other function systems acquire the
task of sorting out what is usable and what is not. Only a
fraction of what is scientifically possible is ever realized. Most is
not feasible economically, legally or politically. The effects of
contingency spread and, in addition to problems that they create
for themselves, other systems are still not in the position to have
to want what is technically possible. In this situation the ability
to reject what is technically possible gains greatly in significance.
It can be used against the creation of ecological risks as much
as in the selection of corrective measures. It is more likely,
however, that it will be practiced in the economy with a view to
economic profitability, in law according to criteria of existing
law and in politics for reasons of political opportunity.
13
Politics
The analysis of the function systems of the economy, law and

science shows that in all three cases a code-closed, autopoietic
self-reproduction is the condition of the system's openness, i.e.,
the condition of resonance capacity and its boundaries. The social
differentiation of these function systems and this is something
they have in common - extended their horizons of possibility.
But at the same time it defined more exactly where the boundaries
of possible resonance in the individual function systems lay. It is
no different for the political system.
Even today politics claims a special place in society, just as it
always has. Traditionally going back to Plato and Aristotle
society has been understand as a politically constituted system,
as societas civilis. In this sense political guidance was an
indispensable structural condition of a communitas perfecta. In
the structure of social differentiation the function of politics was
connected to a determinate, distinctive system arrangement. In
the metaphor of the body, for instance, it was identified with the
head or the soul. In other depictions it assumed the position of
an apex or center. Even today we still expect politics to provide
social integration and the solution of otherwise insoluble problems.
Jean Baechler, for example, continues to view politics as the
center of society. But his definition of this center is enough to
cast doubt on the assumption that, 'The center of the social
system is the activity that combines maximum power with
maximum sensibility.'1
Such ascriptions were plausible for the predominant forms of
Politics 85
structural differentiation of society. The notion of an 'apex' made

sense in the context of stratified differentiation, that of a center
in the context of a parallel differentiation according to center
and periphery as was clearly evident in the distinction of city
and countryside and in the regional differentiation of Europe and
its colonies. These conditions, however, have changed. The
primary structure of social differentiation at present is connected
with the distinction according to function systems. Even the
extent to which contemporary world-society is still differentiated
according to stratification or according to center and periphery
is something that results from the functioning of function systems.
Political reasons have determined the regional segmentation of
the political systems of world-society in terms of states despite the
permanent threat of war. Economic reasons force differentiation
according to center and periphery, i.e., according to highly
developed and underdeveloped regions.
A political theory that does not adjust itself to the realities of
functional differentiation will oscillate between overestimation
and resignation concerning political possibilities and try to
conduct politics with promises and disappointments and the
politics that does not admit that it is incapable of doing something
is caught in this dilemma. As the force whose task it is to put
things in order, politics works mainly through removing the limits
to the appeal to it. It regenerates hopes and disappointments and
continues to thrive because the themes in which this occurs can
be changed quickly. The inclusion of ecological problems within
politics may reinforce this see-saw effect because through them
it becomes quite clear how much politics would have to accomplish
and how little it can. So the political system is constantly tempted
to try to do this through a different government, a different
party, eventually through a different constitution. With this
observation we are already in the midst of the analysis of the
specific resonance capacity of politics and its boundaries.
In reality even the political system is differentiated by rpieans
of a special code through which it attains the closure of its own
mode of operation and an openness to the environment and
change of political programs. The code is commensurate with
the centralization of political power in the state. Power is political
only in so far as it can be used to cover collectively binding
86 Politics
decisions, and the question concerning who is and who is not

entitled to this is defined by the holding of political office. Just
as with the utilization of money in a money economy, this leads
contrary to what was customary to a considerable restriction
of the semantics of politics.2 The channeling of decisions into
political offices establishes politics as difference, not as unity.
The holding or non-holding of positions of political authority,
especially those that regulate who has political influence in
different matters and how much, is what is important. This, of
course, does not mean that politics is exhausted by decisions
ascribed to the state. All the activity that leads up to it is political
if it seeks to influence any of the premisses of these decisions
whether these are programs, forms of organization, the personal
filling of offices or particular subordinate decisions. The structure
of state offices serves as the political code, indeed as the unified
code of all of politics. It defines a zero-sum principle and an
either/or: positions in parliament, government and administration
can only be held or not held. Politics is thereby coded according
to government and opposition depending on whether political
groups enjoy a parliamentary majority, occupy the presidency
and other important government offices or not. As long as
political movements, like the 'Greens', do not adjust to this either/
or alternative, but prefer to operate in the government and in the
opposition at the same time, they act without any understanding of
the structural conditions of the system and the best that they can
do is to make trouble.
Even in this functional domain the code's abstraction serves to
mobilize and adapt programs. The holding of state offices cannot
be legitimized by being held once and for all, for example, by a
dynastic family. Holding such offices is contingent, a process of
selecting persons and programs and is under continual examin-
ation. Political election and the formation of governments serve
to bring the code and program into agreement for a certain
amount of time, i.e., to hand over the government to those who
personally and professionally seem to offer the guarantee of
carrying out the preferred political programs. This presupposes
a structural uncoupling of code and program, i.e., the possibility
of opening access to other programs.
The political complexity that is attained in this way can best
Politics 87
be understood if, for the purpose of historical comparison, the

theory of 'reasons of state' (Staatsrason) is considered. By the
year 1600 the idea of 'state government' was already detectable.
But the organization of the state and the holding of offices
(especially the holding of the leading offices by a prince) were
not clearly distinguished in the concept of the state. Therefore
'reasons of state' were directed at a methodology of holding-on-
to-the-government, while the necessity of 'reasons of state' was
justified by the necessity of government itself. In other words,
the concepts of domination and state were not yet separated, so
that one could still say: 'L'Etat c'est moi\ The code function of
the leading offices - the fact that their occupation by one person
excluded that by another was not yet differentiated from the
program function, i.e., not distinguished from the question of by
whom and according to which programs is the government to
be executed properly. Only the separation of these questions
makes the popular will - expressed in political elections - function
as a criterion, while the code value of office-holding as well as
the code value of truth or the possession of property and money
loses its value as a criterion. In a clearer conceptual sense one
can say that the holders of governmental offices today do not
possess authority any longer, i.e., can no longer presume to act
correctly as office-holders.
According to the official presentation which Max Weber still
accepted it appears as if political authority and the bureaucratic
'apparatus' were given over to the office-holders as the means of
attaining those goals indicated by the popular will. Even the
Middle Ages thought this way, except that at that time the issue
was not the carrying-out of a contingent popular will but the
realization of a communitas perfecta. The consequences of this
theory, for example, the potestas delegata non potest delegari
still concern constitutional jurists today. The goals/means scheme
as well as that of delegation are, however, too simple to provide
an adequate structural description. This does not obviate the fact
that they can or have to be as simple as this for particular
observers. We will therefore replace them with the thesis of the
differentiation of coding and programming. The code of the
exercise of political authority ex officio guarantees that political
authority is always kept in particular hands at any time, thus
88 Politics
that the autopoiesis of a differentiated political system continues

because political power can be applied to all social power
according to its own conditions. At the same time it is clear in
any situation who holds the power (or who acts in its name) and
who does not. Because of this difference a political opposition,
bound to offering a different governmental program, can then
be organized. If this condition is not met - if political opposition
is not permitted (or permitted only covertly) - then it has to be
purchased at the cost of the establishment of a political
stratification of the social system. This restricts the differentiation
of politics. It operates then through organization instead of
coding, without the latter, thereby, gaining better conditions for
an ecological politics.
Of course, the differentiation of coding and programming does
not mean that there is no connection at all between them.
Connections are produced when the holding of office the
positively valued code position - cannot be pursued for its own
sake or for the sake of income.4 A person has to reveal how he
or she intends to exercise political authority and, thereby, what
he or she views as correct behavior. Secondary codings have been
established to bridge the difference of code and program.
Ever since the French Revolution a distinction of restorative
(conservative) and progressive politics has existed. Both of these
can be understood differently. But it is also clear that this coding
can hardly refer to the real dynamics of social change and therefore
remains 'ideological'. Consequently, tendencies to replace them
with the distinction of a more restrictive or expansive understand-
ing of the state have surfaced recently.^ In this way the binary
structure of the code is copied and, at the same time, a viewpoint
for selection of what is considered correct is indicated.
All this has only been a propaedeutic to our theme of the
resonance capacity of the political system. It shows that even the
operations of the political system follow the general course of
differentiated function-systems.6 Thus there is little sense in
attributing a special social position to the political system, like
a leading role or complete responsibility for the solution of
ecological problems. Even the political system cannot act outside
its own autopoiesis, its own code and its own programs. If this
happened then such an activity would not be recognizable as
Politics 89
politics at all. It would not offer the possibility for further

connections. It would necessarily be perceived as something else,
perhaps as a social movement, criminality, youthful immaturity
or as a fashionable or academic curiosity. Thus even the political
system is capable of resonance only within the context of its own
frequencies. Otherwise it would not be a system. It is limited
to executing a practicable politics alone. The conditions of
practicability have to be regulated and, if necessary, changed
within the system itself. It needs to be said that this is not as
in all other cases - a matter of the restrictions of natural law
and of the a priori conditions of impossibility but of the
consequences of autopoietic autonomy and functional differen-
tiation.
Political resonance arises because 'public opinion', as the true
sovereign, suggests the chance of re-election. Even legal offenses,
scandals, etc., are redirected back into politics in this way.7 But
we must also remember that politics' sensibility to themes can
be influenced by the selection of personnel too. There may be
more or less ecologically open, more or less committed politicians
and functionaries. This can cut across all parties throughout the
bureaucracy. In view of these system sensors it can become a
guide-line of political programmatics to look for the possibilities
of extending the resonance capacity of the political system, as
long as resonance is not mistaken for arguing [Rasonanz].
Otherwise an effect like the 'Green' parties results: they are
completely in the right with their principles, but one simply
cannot listen to them.
Fixed limitations of political resonance exist because the
system's own medium of political power, which is protected by
controls of physical force,8 has little chance for application in
highly complex societies. The crude application of such power is
almost useless for the regulation of ecological problems because
no one can be forced into any specific behavior that would
improve the relation of society as a whole to its environment.
Besides, political power, because it ultimately rests on the threat
of physical violence, i.e., on fear, possesses the real limitation
that it can neither prohibit nor prevent anxiety.9 Violence can
certainly be used to combat violence but anxiety cannot be used
to combat anxiety. The recursive application of anxiety on itself
90 Politics
produces peculiar situations of force within politics itself. Anxiety

over what is bad as politics returns as a political factor within
politics. This does not mean that the scope of resonance is
extended. On the contrary, there are only a few possibilities of
dealing with anxiety. It is not surprising then that the politics of
the 'Greens' as long as it operates this way does not seek a
rational attitude towards ecological questions but approaches the
objects of its fears directly. This amounts to the politics of
obstruction: no nuclear energy, no concrete pistes, no cutting-
down of trees, no razing of houses. The limitations of the political
system prevail as the restriction upon blockages, and this can be
covered only by 'principles', not by the responsibility for
consequences. Not least of all this means that environmental
parties have their hands full trying to qualify for the responsibility
of government within the context of the political code. For then
they would have to meet the requirements of the universality and
openness of the political system, i.e., they would have to be able
to develop programmatic guide-lines for all the questions arising
within the system.
A universally competent politics is restricted essentially to two
measures that require intervention in the legal system and in the
economic system. It can make laws under the condition of
compatibility with the legal order and it can spend money under
the condition that the incapacity of making further payments
which results from this, can be transferred. In other words, it
can use power to enforce new laws and it can use power to
procure money without a return for it. Both of these possibilities
increase the possibilities of order that the legal system or the
economic system could produce by themselves. But both of these
possibilities also require that the legal and the economic systems
remain functional and are able to regenerate their specific medium.
The legal system has to produce enough legal components to be
able to handle law-making, and the economic system has to
produce enough capitalistic components in order to be able to
handle the drain of money. In both cases this is not a matter of
a zero-sum game or even of a problem of the exercise of power
and resistance to it but of the conditions of the ability of system
performances to increase. The legal system is a system only if it
can co-ordinate legal communication and legitimize it by means
Politics 91
of recursive operations. For if such communication occurred ad

hoc and passed away with its particular situation then it would
not be recognizable as law but only as the pursuit of interests.
The economic system is a system only if it can produce payments
through payments - otherwise the acceptance of money would
cease and the autopoiesis of the economy would come to an end.
Any use of law or money as an instrument of politics encounters
these barriers although it is difficult to see and mostly only
with hindsight where they are overstepped. The political use
of law and money can break through these barriers because it
can use its power and threaten with force. There are many
examples of this. Viewed in the long term, this occurs at the
expense of the instrumentation of politics. Environmental politics
will require a long-term perspective.10
A further restriction arises out of the national, territorial
limitation of the coding of political power11 and, for the present
at least, from the lack of an effective international, legal regulation
of the transformation of ecological problems into national politics.
This restriction is hardly open to criticism because, since ecological
politics always requires a balancing with other interests and
viewpoints, it is meaningful to bind the decision about this to
the mechanism that articulates political responsibility and is
decided in political elections. Just as obvious are the disadvantages,
since the effects of society on its environment cannot be limited
regionally in many respects. The following are only some of the
examples that show how much society's political resonance
is restricted by territorial sovereignty: the disproportionate
consumption of energy by the United States, which can afford
it; final dumping of atomic waste at a nation's boundaries if
possible; the avoidance of stiff legal duties by transferring
production to countries which - because they have no ecological
politics - offer 'location advantages', disagreement over waste
gas.
But political resonance follows its own logic, not only in a
spatial but also in a temporal respect, and important political
restrictions reside in a wilful dealing with time. On one hand,
politics has to be ready for a short-term change of political
directions because of elections, and an ecologico-economic
accentuation of political programs to promote voter turn-out
92 Politics
would reinforce the effects of such a change even more. This is

in marked contrast to a constantly needed, long-term ecological
politics. Because of the initial situation, we will have to reckon
with a continuous change between primarily economic and
primarily ecological preferences, at the very least. On the other
hand, politically adopted regulations are often more stable than
beneficial, once they become operative. Even if their premisses
have long since been brought into doubt and their consequences
have long since been recognized, to question what is valid a
second time and to dissolve agreements again is difficult and
often inadvisable politically. One never knows in advance whether
something of equal value can be produced. From this point of
view it is hardly advisable to pay 'political prices' for environmen-
tal goods. Misallocations in the housing and agricultural markets
resulting from such policies ought to warn us.
As a result much is subjected to a change that is too fast while
the rest is subjected to a change that is too slow. In any event
and this is a consolation - the individual political themes have
very different temporal horizons that cannot be combined.
Thereby, the temporal structure of politics agrees only to a small
degree with the requirements of other systems, not to mention
with changes in the ecological environment. New exhaust
regulations for automobiles may be felt to be urgent and desirable
in view of the 'dying-off of forests'. But as far as the business of
politics is concerned this is only one argument among many. The
more politics depends on co-operation and the development of
consensus the greater is the probability of delays, of new
unexpected initiatives and of long-since obsolete bodies of
regulations.
It is proper, therefore, to ask whether a competitive democracy
of the type indicated here can introduce controversial environmen-
tal themes into politics. 12 With all the willingness to be
straightforward and to announce intentions, and despite the
spectacular career of the theme itself, not much of this has been
noticed so far. Agreement has been reached that something has
to be done. But the problems have to be allowed to become
urgent enough so that action can be taken without the chance
of losing any votes. Hitherto problems have been created by
bringing environmental themes into party rivalries themselves,
Politics 93
i.e., creating difference in the sense that one party commits itself
more than the others to long-term environmental programs even
at the expense of the economy (including the loss of jobs) and
campaigns with this difference. This procedure still seems too
risky to the so-called 'peoples parties' who regard voter-swings
of even small percentages as catastrophic. It is possible that this
is changing and that ecological themes, with their own harsh
demands, are supplanting the old sociopolitical ones. We will
always have to reckon with limited resonance. But it is by no
means settled that competitive democracy and the coding of
politics have to ruin themselves with ecological themes by keeping
access to the government open.
14
Religion
Theologians are included in the discussions involved with

environmental problems too. Their motives and interests are
not viewed with suspicion. They demonstrate argumentative
competence and are undeniably of good will. But their contri-
butions to the ecological discussion remain inadequate. To a
great extent they merely repeat what has been thought and
proposed elsewhere without the specific religious reference. What
they have to offer are mostly commonplaces that do not raise
the real problems. These are usually concealed in concrete
pictures, words, admonitions and appeals. After all, they propose
not to take technology, science and economic relations as the
sole prevailing vehicles of domination. Instead, they believe that
the latter ought to be auxiliary in the formation of a human
culture within the natural condition.1 Such things are better left
unsaid. They are inadequate, and of no greater help if theologically
reformulated by invoking God.
In view of this, how do things stand with the resonance
capacity of the religious system? What structures make the
resonance of this system possible, i.e., restrict it?
Much earlier than in the case of other systems (but therefore
much more precariously too) a differentiation of coding and
programming seems to have worked its way into religious
questions. In very early religions the sacred remains immediately
paradoxical, namely, enchantment and terror, attraction and
repulsion at the same time. It is recorded in rigid (but pragmatically
usable) forms of rituals, taboos, symbolically visible duplication
Religion 95
or even mythical stories. This background provides the explanation

for the success of the moral coding of religion that applies the
duality of joy and anxiety to the moral code of good and bad
behavior and thereby eliminates its paradox: good behavior
produces the positive feeling of nearness to God, while only for
morally bad behavior does He have to be feared. In this way,
salvation and damnation come into view. God is a good God
from Whom evil has mysteriously slipped away or serves
mysterious ends. Accordingly, the cosmos acquires a valid
principle from which the difference of good and evil is derived.
Before long this hierarchical form of paradox-elimination raised
doubts and questions. A religious observer like Job, for instance,
saw what appears necessary to the system as contingent and
looked for the reason. A mere glimpse of the paradox caused
problems for early religious reflection. The system-guiding
reflection had to concede that religion's purpose was not moral
bifurcation alone, i.e., the emphatic confirmation of the difference
of good and bad behavior. The coding of religion had to cut
across morality. Not that moral questions had no role to play!
Quite the contrary. But because of this religion could not stake
everything on the difference of good and bad behavior. As much
as it may have been overgrown at times with moral cosmologies
like the difference between heaven and hell, the moral classification
of persons had never been its real concern. Instead this was seen,
particularly in the late Middle Ages, as the work of the devil;
something that God, through the intercession of Mary, had to
counter. After all, the presumption of making a judgement about
good and evil itself was the devil's handiwork. So the idea that
morality stemmed from the devil always existed somewhere.
Consequently, the coding of religion can neither identify itself
with morality nor separate itself from it (for even the devil is a
power that is conditioned by religion). Its ultimate difference
resides in the distinction of immanence and transcendence.2
Transcendence is no longer understood in terms of another world
or as a separate and unattainably high or low region of the world
but as a kind of second meaning, i.e., as a complete, all-
encompassing second version of the world where self-reference
has meaning only as other-reference, complexity has meaning
only as implexity (Valery) and transcendence has meaning as
what cannot be transcended.
96 Religion
As is characteristic for all codes, the ever-present, socially self-

determining reality is duplicated by an implicit assumption. It is
identified by a distinction, i.e., indicated within the context of
this distinction. The unity of this difference (and not transcendence
as such) is religion's code. This code has many different semantics
that differentiate the religious system into different religions.
Take the example of creation myths. Through the creation of the
difference of heaven and earth God excluded Himself from the
world. Or take the example of the institution of a difference
between the sacred and the profane by which the desacralization
of nature is made a condition of the specification of religion and
so recorded irrevocably or the example of the historically
confirmed adoption of the belief that Jesus is the Christ, or the
late medieval/early modern binary coding of the moral disjunction,
according to which the sinner can be saved through remorse and
grace while the just person is lost precisely because of believing
him- or herself to be just.3 The pure difference of immanence
and transcendence is enriched in many different ways and
subjected to the evolution of conditions of plausibility. If society,
as it appears at present, encounters new situations then this does
not mean that one has to bring the coding of religion into doubt
immediately. But if this is done the semantic imposition of
cosmologies and theologies on the code ought to be examined
critically. Accordingly, we will examine the symbolic mediations
between coding and programming at the same time.
As the outline of the problem in figure 3 makes clear (cf. figure
2 in chapter 9 above) the questions here are somewhat different
from those in the function systems that remain.
code program
unity God revelation
operation immanence/ rules of Holy
transcendence Scripture
Figure 3
Religion 97
If it is true that the modern, functionally differentiated society

forces a greater differentiation of coding and programming, then
as far as religion is concerned, the problems do not reside in the
reflection of the unity of the code but in the reflection of the
programmatic unity of the system, its goals and the conditions
of the correct attribution of its values. This is where the system
comes into competition with formulas of correctness like justice,
prosperity and knowledge which are already functionally differen-
tiated. Because progress became a real possibility in this regard
in the seventeenth and eighteenth centuries the programmatic of
correct attribution became a problem for religion itself. The
morally rooted difference of salvation and damnation, of heaven
and hell, of the love and fear of God, receded. Hell completely
lost its plausibility. On the other hand, two ideas gained
acceptance as dogmatic articles of faith: (1) the intellectual step
from 'God' to 'revealed religion'; and (2) the guidance of access
to transcendence through the rules of Holy Scripture.4 This means
that de-differentiation [Entdifferenzierung] between the moral
foundation of the code, on one hand, and forcing the awareness
of contingency into the dogmatical, on the other, mutually
condition and require each other. The unity of the system - the
connection of coding and programming can be secured only in
a way that cannot be advocated as dogmatics any longer.
At the same time the religiously founded cosmology is discarded.
Even very old traditions had abandoned a directly religious
qualification of nature because this was the only way that allowed
for the emergence of separate religious formations that are
otherwise distinct from the world, for example, the desacralization
of nature as the condition of the specification of the religious.
In early modernity this desacralization of nature merely changed
its reference system. It was no longer a primarily religious but
became a primarily scientific or primarily economic requirement,
and religion could not intervene in this process because it had to
preach the same sermon.
At first an attempt was made to transfigure, glorify this
difference and, thereby, to resolve the paradox of a world-order
created by God that was confusing, misfortunate and inconvenient.
With the rejection of these solutions, however, the order itself
becomes a paradox. 'The disorder in the world is merely apparent.
98 Religion
And where it seems to be greatest is where the true order is still

even greater. It is simply more concealed from us.' 5 The difference
of immanence and transcendence is dissolved or transformed into
the difference of manifest and latent. The pointing finger
[Fingerzeig] of God (providentia specialis) becomes the 'invisible
hand'. This allows one to exploit the fact that manifest structures
are more sensitive to deviations than latent ones. This version of
the code provided possibilities of deviation and sensitivities and
thus saved religion.
In this way an optimistically progressive society could tranquil-
ize itself concerning the problems it produced.6 But in view of
contemporary uncertainty about the future it will hardly suffice
to imagine the grandeur of the world-order in what cannot be
seen. Even the question of theodicy, of how God can permit all
this why He did not create completely impenetrable atoms,
why He accepts chemical fertilizing and why He allows the heads
of big business to wheel and deal - does not help either. The
actual problematic is blocked by the question of the justification
of God or, in any event, is not addressed. The problem resides
in the impossibility of deducing the problematic from the code.
This requires a semantics of translation for any coding, for
example, a theory of theory formation (a theory of science) or
the legitimization of law-making or the economic superiority of
capitalism or socialism. The contemporary religious system has
nothing comparable to offer in this case, and the doubts that are
raised and justified by these semantics of translation produce
hesitation with every attempt.
As long as this does not change - and change seems impossible
- religion (or, in its name, theology), will have little to contribute
to the social resonance to the exposure to environmental dangers.
To be sure, it will be able to protest against deforestation, air
pollution, nuclear dangers, or the excessively medical approach
to the human body when these problems have acquired a certain
evidence. But it will not be able to intervene with a genuinely
independent form of the problem because it remains dependent
on an antecedent social awareness of the latter. Wherever
the assurance of meaning is required, wherever environmental
experiments with far-reaching, unpredictable consequences have
to be carried out is where theology admonishes and creates
Religion 99
uncertainty or remains silent. Strictly speaking, it has no religion

to offer.7 One almost gets the impression that religion today
develops as a kind of parasite on social problem situations -
parasite in the sense of Michel Serres8 as the reintroduction of
the 'tertium non datuf into the system. In other words, religion
profits from the binary structure and the exclusion of the 'tertium
non datuf in all other codes. It profits because it can provide a
formula of unity for its code and thus include the excluded third
possibility, the 'tertium non datur\ But does this mean that it
has to leave the programmatic of what is correct to the other
social function-systems and can provide only inferior programs
of its own, for example, in a fundamentalistic, concretist,
miraculous, eschatological way or in the form of a 'new myth'?
Once again, the entire problematic of social resonance to social
dangers is reflected here as if in a mirror. Resonance can be
created only through structural restrictions, the reduction of
complexity, selective coding and programming, i.e., only inad-
equately. For the time being at least, religion only seems to
confirm this through the rejection of its own reductions and thus
through the rejection of its own resonance. But if this were to
continue to be its presentation of the finitude of the human
world, would not everything depend, for the Christian religion,
on holding fast to the certainty of being accompanied by God?
This does not lead to an environmental ethics or even to a
theological exaggeration of environmentally political demands.
But it is conceivable that there are ecological as well as social
marginal states in which it is impossible for humanity to
experience the certainty of faith and to hope for redemption. At
the very least the fact that this must remain a possibility can be
justified by religion.
15
Education
One might look to the education system as a source of great

hope for the future because an interest in ecological questions is
a priority with today's young people. Would it not be possible
for the education system, especially in schools and universities,
to take up this interest and develop it toward a gradual change
in society's awareness of and attitude towards the environment?
We no longer enjoy the eighteenth century's pedagogical optimism
in education's ability to produce a complete change in humanity
over two or three generations. Some say that this ecological
interest is a political maneuver of one class to direct attention
away from the really important questions of poverty, injustice
and war. 1 Even if this is given due consideration one could still
hold that schools are the place where society should prepare itself
for an encounter with the environment.
But the education system is only one function system among
others, and in so far as it is not concerned with the education
of educators, it establishes attitudes and abilities that have to be
actualized in other function systems. It works in conjunction with
swings in public opinion. Perhaps it does so more slowly or more
persistently, but essentially it does so without the certainty of
being able to find connections with the activity of the other
systems. Nevertheless, the value of education may be rated very
highly if we remember how insignificant great problems appear
after passing through the filter of what is normatively possible,
what the costs are and how much depends on reacting to
environmental consequences. No function system is always
Education 101
capable of producing the only correct decisions. With many

practicable solutions it may be of increasing importance whether
and how the ecological ramifications of their effects are heeded.
But how can these problems be handled communicatively within
the education system? Even this system - like all the others - is
capable of resonance only with severe restrictions.
Although the primary purpose of the education system is not
to process communications but to change humanity, developments
parallel to other function systems are also found in it. For instance,
the education system has to react to its own differentiation through
a structural differentiation of coding and programming. Of
course, it is more difficult to recognize how this happens than
in the case of religion. The pedagogical literature is as unreliable
in its way as theological literature because both begin from a
programmatic orientation.
The coding of the education system is connected with its
function of making social selections. This is where we find that
technical bivalence that characterizes a code. A person can do
well or poorly in exams, be commended or rebuked, receive good
or bad grades, be promoted or not, be admitted to advanced
courses or schools or not. Finally, he or she can graduate or not.
Occasionally this bivalence can be broken down into scales. But
even then it functions comparatively (whether temporally for the
same person or socially in relation to others) as the bivalence of
better or worse.
The code of the education system results from the need to
develop a career, i.e., to construct a sequence of selective events
which, at any time, results from an interplay of one's own
selection and the selection of others and signifies the condition
of possibility and structural restriction for events connected with
this.2 Only if someone is admitted to a school can he or she
receive grades. Grades are important for promotion within the
school. The successful completion of a program of study is
important for entry into a career. Entry into a career determines
advancement. In all these cases the intentional or unintentional
non-fulfillment of requirements has value for the career as well,
only it is negative. Those who leave this process without
completing it may decide on no career at all. But they cannot
avoid this reference of their behavior to careers since careers are
102 Education
a given possibility and since they are the standard instrument of

inclusion by which persons are appointed to positions throughout
many systems.
The system's programs, on the other hand, have to do with
the content of what is to be learned or they describe the conditions
or capacities that are expected of a person as the result of the
education process. Through programs the education system is
connected to social demands. On this structural level it can also
be connected with ecologically important knowledge. Instead of
learning when Frederick Barbarossa was born and when he died,
people learn which thallium values according to the technical
guide for clean air - present the tolerance limits. After the
completion of an intensified course they also learn why the limits
have been set in this way, what could happen if these limits are
exceeded and which arguments could be used to effect a change
in these values, if necessary.
Beyond the question of studying Barbarossa or thallium, these
things have a second existence according to the selection code of
the education system. They can be known correctly or incorrectly
or not even known at all. More depends on them than merely
German history or the chances for the authorization of an increase
in the output capacity of a cement works. Once introduced into
the education system they have a career value - that is, whether
someone has learned this well or poorly or not at all. When there
is a differentiated education system, careers are unavoidable.
Their coding can take up and handle any themes at all - therefore
the code of social selection can and also has to outlast the change
of programs, educational goals and pedagogical fantasies. But
because of this it does not meet the requirements of the main
concern of programs, which is to produce something correct. It
merely creates a formal contingency that forces itself upon all
programs. This is not a contingency of knowledge, for example,
that Barbarossa did not have to be born or that the technical
guide for clean air could have been formulated differently but
one that is created by the differentiation of the education system
itself. In the education system's context of selection everything
can be connected with everything else, provided that it has been
assigned a value. Both Barbarossa and thallium enter into the
grade-average that prestructures what can be connected to this.
Education 103
If one takes this assumption as the point of departure it is easy

to explain why, even in the education system, tendencies to
separate and recombine coding and programming appear. On
one hand, the pedagogical profession assumes the task of selection
- the participation in examining and grading - as an unpleasant
collateral business that disturbs and impedes the real task of
educating. On the other, the binary code establishes itself here
as in the case of the other systems, and programs are selected
and used to enable the co-ordination of code values. To an extent
that could only be explained after careful analysis, the student is
treated as a 'trivial machine' (in distinction to a Turing machine),
i.e., as a machine that has to produce the only correct answer to
an input like a question or an assigned task (it makes no essential
difference if a determinate scope of reaction possibilities is
acceptable). Above all, this means that the momentary state of
the system tested in this way, for example, whether the student
is willing or attentive or is interested as such, plays no role for
the instructor.3 To the extent that the student is treated as a
trivial machine the ability to integrate coding and programming
is equally assured. If educators rise up in protest against this
description of their premisses and activity then this indicates, at
the very least, that they take the difference of coding and
programming more seriously than schoolmasters who take a more
'hands-on' approach. But at the same time the question becomes
more urgent: how can the quality of the work of self-referential
machines be fixed if Turing qualities like 'no trouble-makers' are
required of the participants?
If one takes this dominance of coding as the starting-point
then it becomes immediately evident that the career structure that
forms the basis of selection in and out of school transfers the
social pressure of problems back into the school and how it does
so. This can be recognized in school as 'performance stress' or
as creeping discouragement or despair over the uncertainty of
finding a desirable job because of school performance. In the
case of career possibilities the demands on performance and
discouragement increase proportionately when it becomes clear
that future prospects depend on particular performances and
that these are uncertain. It is very questionable whether the
determinations of attitudes that enter the system through the
104 Education
selection code can be balanced effectively on the level of programs

through a better agreement of interests and curricula. Of course,
this is ultimately an empirical question. But the theoretical
assumption of a differentiation and recombination of coding and
programming speaks against it. Within the system the curricula
have to be suitable to enable a just and relevant distribution of
code values, and this structural requirement is difficult to reconcile
with the claim of non-trivial, self-referential machines to be
taught only according to their own interests and conditions.
This necessarily brief and rough outline of only one of
the many structural problems in school education4 already
demonstrates that even the education system is not free from
indeed to a great extent is overburdened by its own structural
problems and operative requirements, which leave little room for
a simple reprogramming toward an increase in ecological
sensibility. Of course, it is easier here than elsewhere to arrange
things, whether thallium or Barbarossa. But we will have to
concede realistically that social systems cannot establish a rational
relation to ecological problems by this alone. One can just as
well imagine that, at the same time, a further, quasi-reflexive
kind of environmental pollution is created, namely, misplaced
ideas, to go along with misplaced matter.
The education systems works directly only on a particular
environment of the social system the bodily and mental
conditions of people. If the effects of this are felt in the social
system then this environment in turn has to affect society, that
is, be able to be connected with it communicatively. Thus the
education system offers perhaps the best chances for an extension
of intensified ecological communication - under the condition
that two thresholds of resonance are overcome: that of the
education system itself and that of all other function systems of
society in which new attitudes, evaluations and sensibilities to
problems are introduced through education.
It might be difficult to evaluate these possibilities realistically.
In times of need and as reserves of certainty, so to speak,
they might acquire greater importance than as well-coded,
programmatically organized daily life.5 Not least of all, the
success of environmental politics depends on whether and to
what extent coded communication can react to what participants
Education 105
assume as the opinions of other participants. We should also

mention another consequence of education, long-term effects and
value-changes: the presumption of consensus brings about social
movements that do not fit within the function systems and can
be experienced by them only as noise. We will come back to this
in chapter 18.
10
Functional Differentiation
The preceding chapters discussed the existence of ecological

problems and the ways in which they trigger resonance in the
function systems of modern society. But in the analysis of
particular systems the sociologist should not lose sight of the
unity of society. Indeed, the comparability of function systems
and certain agreements in the structures of their differentiation
- we examined the differentiation of codes and programs but
this is only one of many viewpoints - point to this. The unity
of the entire system resides in the way it operates and the form
of its differentiation. The more clearly social evolution approaches
a specific kind of operation, namely, meaningful communication,
and the primacy of functional differentiation vis-a-vis other forms
of internal system-formation the more obvious its corresponding
structures become. If one eliminates all anachronisms, the
conceptual and theoretical means by which society describes itself
in its scientific system in this case in sociology have to be
adapted to this.
Above all, one must realize that theories of hierarchy, delegation
or decentralization that begin from an apex or center are incapable
of grasping contemporary society adequately. They presuppose a
channelling of the communication flow that does not exist nor
can even be produced. Furthermore, the attempts to describe the
relation of state and economy according to the model of
centralization and decentralization and then, when it is politically
expedient, to praise the advantages of decentralized decision-
making and to warn against its disadvantages is unrealistic. In
Functional Differentiation 107
reality, the economy is a system that is highly centralized by the

money-mechanism but with a concomitant, extensive decentraliz-
ation of decision-making, whereas the political system organizes
the political organisation more or less centrally and handles
political influences according to entirely different models, like
those of social movements. These systems distinguish themselves
through the way in which they try to combine and reinforce
centralization and decentralization according to their respective
media of communication. But their independencies cannot be
understood according to the model of centralization and decentra-
lization.
Thus it is pointless to try to conceive the unity of modern
society as the organization of a network of channels of
communication, steering-centers and impulse receivers. One
immediately gets the impression that good intentions cannot be
realized because somewhere something is directed against them1
which frequently ends up in mythical explanations in terms of
capitalism, bureaucracy or complexity. With the help of a theory
of system differentiation it is evident, however, that every
formation of a subsystem is nothing more than a new expression
for the unity of the whole system.1 Every formation of a subsystem
breaks the unity of the whole system down into a specific
difference of system and environment, i.e., of the subsystem and
its environment within the encompassing system. Every subsystem
therefore, can use such a boundary line to reflect the entire
system, in its own specific way; one that leaves other possibilities
of subsystem formation open. For example, a political system
can interpret society as the relation of consensus and the exercise
of force and then attempt to optimize its own relation to these
conditions. On one hand, consensus and force are specific
operations, but on the other, they are also all-encompassing
formulas and horizons for social conditions and consequences
that can never be made completely transparent in the political
subsystem.
Every function system, together with its environment, recon-
structs society. Therefore, every function system can plausibly
presume to be society for itself, if and in so far as it is open to
its own environment. With the closure of its own autopoiesis it
serves one function of the societal system (society). With openness
to environmental conditions and changes it realizes that this has

to occur in the societal system because society cannot specialize
itself to one function alone. This is a matter of the operationaliz-
ation of a paradox. Presented as the difference of system and
environment the function system is and is not society at the same
time. It operates closed and open at the same time and confers
exclusivity on its own claim to reality, even if only in the sense
of a necessary, operative illusion. It confers bivalence upon its
own code and excludes third values that lurk in the environment's
opacity and the susceptibility to surprise. In this way society
reproduces itself as unity and difference at the same time. Of
course, this does not eliminate the paradox of unitas multiplex.
It reappears within the system as opacities, illusions, disturbances
and the need for screening-off - as transcendence in immanence,
to put it in terms of the religious system's selective coding.
This systems-theoretical analysis highlights the significance and
the preference of modern society for institutions like the market
or democracy. Such descriptions symbolize the unity of closure
and openness, of functional logic and sensibility. Of course, the
market is not a real one (as it could be seen to be from the
cousin's corner window)3 and democracy no longer means that
the people rule. This is a matter of a semantic coding of an
ultimately paradoxical state of affairs. It explains the meaning
and the illusionary components of these concepts, explains the
weakness of the corresponding theories and explains why, since
the beginning of the eighteenth century, a kind of self-critique
has accompanied this.
Yet the unity of this order is already necessarily given by
evolution, i.e., through the continual adjustment of possibilities.
Evolution does not guarantee either the selection of the best of
all possible worlds nor 'progress' in any sense. At first evolutionary
selection produces a very improbable, highly complex order. It
transforms an improbable order into a probable (functional) one.
This is exactly what concepts like negentropy or complexity
intend. But it does not mean that the improbability disappears
or is inactualized as prehistory. It is co-transformed and
'aufgehohen' in Hegel's famous sense. It remains a structurally
precipitated risk that cannot be negated.
Stratified societies already had to deal with problematical
consequences of their own structural decisions. These were

expressed, for example, as the constant conflict between inherited
honors and distinctions and new ones, as the unfulfillable
obligation to prescribe a class-specific endogamy and not least
of all as the conflicts that result from centralizing the control of
access to scarce resources, above all of the ownership of land.
Compared to modern society these are relatively harmless
problems for which historically stable solutions were found in
many cases. The transition to primarily functional differentiation
leads to a completely different constellation with higher risks and
more intensified problems resulting from structural achievements.
Society's self-exposure to ecological dangers is therefore not a
completely new problem. But it is a problem that, today, is
coming dramatically to the fore.
With functional differentiation the principle of elastic adap-
tation through processes of substitution becomes the principle of
the specification of subsystems. Its consequence is that, more
than ever before, functional equivalents can be projected and
actualized but only in the context of the subsystems and their
coding. Extreme elasticity is purchased at the cost of the peculiar
rigidity of its contextual conditions. Everything appears as
contingent. But the realization of other possibilities is bound to
specific system references. Every binary code claims universal
validity, but only for its own perspective. Everything, for example,
can be either true or false, but only true or false according to
the specific theoretical programs of the scientific system. Above
all, this means that no function system can step in for any other.
None can replace or even relieve any other. Politics cannot be
substituted for the economy, nor the economy for science, nor
science for law or religion nor religion for politics, etc., in any
conceivable intersystem relations.
Of course, this structural barrier does not exclude corresponding
attempts. But they must be purchased at the price of dedifferen-
tiation (Entdifferenzierung), i.e., with the surrender of the
advantages of functional differentiation. This can be seen clearly
in socialism's experiments with the politization of the productive
sector of the economy or even in tendencies towards the
'Islamization' of politics, the economy and law. Moreover, these
are carried out only partially. For example, they do not touch
108
on money (but, at best, the purely economic calculation of capital

investment and prices) and are arrested by an immune reaction
of the system of the world society.
The structurally imposed non-substitutability of function sys-
tems does not exclude interdependencies of every kind. A
flowering economy is also a political blessing - and vice versa.
This does not mean that the economy could fulfill a political
function, namely, to produce collectively binding decisions (to
whose profit?). Instead, the non-substitutability of functions (i.e.,
the regulation of substitution by functions) is compensated by
increasing interdependencies. Precisely because function systems
cannot replace one another they support and burden one another
reciprocally. It is their irreplaceability that imposes the continual
displacement of problems from one system into another. The
result is a simultaneous intensification of independencies and
interdependencies (dependencies) whose operative and structural
balance inflates the individual systems with an immense uncontrol-
lable complexity.
This same state of affairs can be characterized as a progressive
resolution and reorganization of the structural redundancies of
society. The certainties that lay in multifunctional mechanisms
and that specified systems for different functions and programmed
them to 'not only/but also' were abandoned. This is shown very
clearly by the reduction of the social relevance of the family and
morality. Instead, new redundancies were created that rested on
the differentiation of functional perspectives and 'ceteris paribus'
clauses. But this does not safeguard the interdependencies between
the function systems and the social effects of the change of one
for the other. Time, then, becomes relevant: the consequences
result only after a certain amount of time and then they have to
be handled with new means that are, once again, specific to the
system. This is accomplished without being able to go back to
the initiating causes. Complexity is temporalized4 and so are the
ideas of certainty. The future becomes laden with hopes and
fears, in any event, with the expectation that it will be different.
The transformation of results into problems is accelerated, and
structural precautions (for example, for sufficient liquidity or for
invariably functional legislation) are established so that such a
reproblematization of the solution is always possible.
The rejection of substitutability has to be understood essentially

as the rejection of redundancy, i.e., as the rejection of multiple
safeguarding. As we know, the rejection of redundancy restricts
the system's possibilities of learning from disturbances and
environmental 'noises'.5 This implies that a functionally differen-
tiated system cannot adapt itself to environmental changes as
well as systems that are constructed more simply although it
increasingly initiates concomitant environmental changes. But this
is only part of the truth. For, through abstract coding and the
functional specification of subsystems, functional differentiation
makes a large measure of sensibility and learning possible on this
level. This state of affairs becomes quite complicated when many
system levels have to be kept in view at the same time. Society's
rejection of redundancy is compensated on the level of subsystems,
and the problem is that this is the only place that this can occur.
Family households, moralities and religious cosmologies are
replaced by an arrangement in which highly organized capacities
for substitution and recuperation remain bound to specific
functions that operate at the cost of ignoring other functions.
Because of this the consequences of adaptive changes are situated
within a complex net of dependencies and independencies. In
part, they lead to unforeseen extensions, in part they are absorbed.
In such cases simple estimations and simple comparisons of the
efficiency of different social formations are insufficient and
inadvisable.
A further consequence of functional differentiation resides in
the intensification of apparent contingencies on the structural
level of all function systems. Examples of this are the replacement
of natural by positive law, the democratic change of governments,
the still merely hypothetical character of the validity of theories,
the possibility of the free choice of a spouse and not least of all
everything that is experienced as 'a market decision' (with
whoever or whatever may decide) and is subjected to criticism.
The result is that much of what was previously experienced as
nature is presented as a decision and needs justification. Thus a
need arises for new 'inviolate levels' (Hofstadter), for a more
rational and justifiable a priori or, finally, for 'values'.6 Evidently,
the strangely non-binding compulsion of values correlates to a
widespread discontent with contingencies as much as to the fact
108
that decisions become more exposed to criticism through structural

critique and statistical analyses than facts. Indeed, even if we
cannot determine that someone has decided (for example, about
the number of deaths from accidents or about the increase of the
rate of unemployment) decisions are still necessary to redress
these unsatisfactory conditions. To require decisions means to
appeal to values, explicitly or implicitly. Consequently, structural
contingency generates an order of values without considering the
possibilities of concretely causing effects, i.e., without considering
the attainability of the corresponding conditions.
It is probable that ecological communication will intensify this
inflation of values even more. For if society has to ascribe
environmental changes to itself then it is quite natural to reduce
them to decisions that would have to be corrected: decisions about
emissions quotas, total consumption amounts, new technologies
whose consequences are still unknown, etc. We already noted in
chapter 3 that such ascriptions are based on simplifying,
illuminating and obscuring causal attributions. This does not
prevent them from being carried out and communicated, but, if
nothing else, it permits values to surface.7
At first, one might think that the value of clean air and water,
trees and animals could be placed alongside the values of freedom
and equality, and since this is only a matter of lists we could
include pandas, Tamils, women, etc. But viewed essentially and
in the long run this would be much too simple an answer. The
problematic of the inflation of values as a symbolically generalized
medium of communication - an idea of Parsons's8 results from
its influence on society's observation and description of itself.
Actually the descriptions of society are steered by the problems
that result from structural decisions and, therefore, they have a
tendency to evoke values and see 'crises'. Contrary to the mature
phase of bourgeois-socialist theories in the first two-thirds of the
nineteenth century disadvantages are deferred for a time, are read
off in values and are understood as the indefinite obligation to
act. In any event, they are no longer understood as digressions
of the spirit or matter on the way to perfection. Instead, they
are the inescapable result of evolution. According to the theory
proposed here, they are consequences of the principle of system
differentiation and of its making probable what is improbable.
Moreover, the critical self-observation and description that

constantly accompanies society has to renounce moral judgements
or end up getting lost in a factional morass.9 Instead, a new
kind of schematism, namely, manifest or latent (conscious or
unconscious, intentional or unintentional) takes its place. Only
manifest functions can be used to differentiate and specify because
only these can be transformed into points of comparison or goal-
formulas. This means that the critique is formed as a scheme of
difference that also illuminates the other side, the counterpart.
Straightforward striving toward a goal is viewed as naive. This
even undermines the straightforward intention of enlightenment.10
A mirror is, as it were, held up to society, assuming that it cannot
look through it because that which is latent can fulfill its
function only latently. This is the way sociology, too, pursues
'enlightenment' [Aufklarung] and explains its ineffectuality in the
same process.11 In this sense ideology, the unconscious, latent
structures and functions and unintended side-effects all become
themes without a clarification of the status of this shadow world
note especially the reversal of Platonic metaphysics. One can
therefore use this distinction only to discover that society
enlightens itself about itself.
The problem of reintroducing the unity of society within society
or even of expressing it in it is extended to the forms of the
system's critical self-description. Equally symptomatic are all
attempts at judging and condemning society from the exalted
standpoint of the subject, i.e., ab extra. This signifies nothing
more than placing the unity of society in a principle outside
itself. 12 A systems-theoretical analysis of such attempts, however,
enjoys the advantage of being able to retrace this problematic
back to the structure of modern society (which changes nothing
about the fact that this must occur in society).
Essentially, every attempt within the system to make the unity
of the system the object of a system operation encounters a
paradox because this operation must exclude and include itself.
As long as society was differentiated according to center/periphery
or rank, positions could be established where it was possible, as
it never has been since, to represent the system's unity, i.e., in
the center or at the apex of the hierarchy. The transition to
functional differentiation destroys this possibility when it leaves
114 Functional Differentiation
it to the many function systems to represent the unity of society

through their respective subsystem/environment differences and
exposes them in this respect to competition among themselves
while there is no superordinate standpoint of representation for
them all. To be sure, one can observe and describe this too. But
the unity of society is nothing more than this difference of
function systems. It is nothing more than their reciprocal
autonomy and non-substitutability; nothing more than the
transformation of this structure into a togetherness of inflated
independence and dependence. In other words, it is the resulting
complexity, which is highly improbable evolutionarily.
17
Restriction and Amplification
Too Little and Too Much
Resonance
Detailed analyses of society's resonance to dangers from its

environment have to be connected with the renouncement of
redundancy that resides in the non-substitutability of function
systems. This forces a channelling of all disturbances into one
or several of these function systems. Whatever appears as
environmental pollution can be treated effectively only in
accordance with one code or another. Of course, this does not
exclude the possibility that an eclipse or an earthquake can prove
upsetting in other, unspecified ways. As already mentioned in
reference to general systems-theoretical and particularly in
reference to biological investigations, the renouncement of
redundancy leads to a restriction of the capacity of reacting to
disturbances (noise). On the other hand, structural restriction is
also a way of increasing resonance capacity, as is quite clear
from the case of the organism. Eyes and ears, nervous systems
and immune systems that, for their part, are capable of resonance
only within narrow but evolutionarily proven frequency ranges,
develop only through a considerable amount of rejection. These
reductions can then be balanced only through the organized
capacity to learn.
This is the path modern society seems to have taken with its
choice of functional differentiation. It would be pointless to ask
whether there might have been other possibilities. It would be
equally pointless to ask whether we could transport ourselves
into a 'post-modern period' or are even in the process of
completing this transition. The actual circumstances offer no kind
of support for this. Instead, such assumptions are only premature

inferences from simplistic theories. The only meaningful question
must be whether we can use the renouncement of redundancy
that inheres in functional specification better than before. After
the logic of functional differentiation and its resulting problems
are discovered we can determine how the restriction of resonance
operates through the coding of the individual function-systems.
On one hand, this excludes all point-for-point relevance and
renounces the establishment of 'requisite variety'. Coding causes
a sharp reduction for every function system. When environmental
changes trigger resonance in self-referential function systems this
is an exception. Only in these cases do environmental changes
disturb and change the conditions of the continual reproduction
of system-specific communication.
On the other hand, this reduction is the condition for noticing
and processing environmental changes within the system as such.
Coding is the condition that permits environmental events to
appear as information in the system, i.e., to be interpreted in
reference to something, and it causes this in a way that allows
consequences to follow within the system. Of course, these chains
of consequences are mediated by further conditions: by the system
programs, for example, theories, laws, investments or party-
political alignments. If ecological problems pass through this
double-filter of coding and programming they acquire internal
relevance for the system and possibly far-reaching attention
but only in this way!
Despite all this, there is no guarantee that society as a whole
will protect against or can even contend with exposure to
ecological dangers in every case. On the contrary, society can
react only in exceptional cases. This implies that it brings too
little resonance into play for the exposure to ecological dangers.
At present this inference matches what public opinion suspects.
So social communication is alarmed and stimulated to more
activity without, of course, being able to translate this requirement
into the language of the function systems. But this is only half
the problem. The other half is more difficult to discern and at
present overlooked to a great extent. There can also be too much
resonance and the system can burst apart from internal demands
without being destroyed from outside.
The problem of resonance capacity, since it refers to a

differentiated system, does not reside simply in one dimension
where 'too little' and 'too much' could be balanced against each
other. Instead, two system boundaries have to be distinguished:
society's external and internal boundaries. By means of external
boundaries society screens off its own autopoiesis, i.e., communi-
cation, from the enormous complexity of non-communicative
states of affairs. On this, the level of its own operations there is
neither input nor output. Society cannot communicate with
but only about its environment according to its capacity for
information processing. Society itself, thereby, regulates what
constitutes information for society. But it can also be influenced
in the selection and ordering of communication by irritation and
disturbances, particularly by the conscious processes of the
participants.
Entirely different circumstances pertain at the internal bound-
aries of the system. This is where there are communicative
interdependencies. The aggregate data of the economic system -
growth-rates, unemployment numbers, inflationary and
deflationary developments influence the political system. Even
if the function systems are differentiated according to their own
autopoiesis, codes and programs, they can be disturbed by
communication in a way that is entirely different from the way
society itself relates to its environment. It is therefore highly
probable that the turbulences of one function system are
transferred to others even if, and because, each proceeds according
to its own specific code. For example, the economy is at the
mercy of scientific discoveries and technological innovations as
soon as these find economic use. The same is true mutatis
mutandis for the relation of politics and law, for science and
medicine and for numerous other cases. There is no superordinate
authority that would provide for measure and proportionality
here. Through resonance small changes in one system can trigger
great changes in another. Payments of money to a politician that
play no role in the economic process measured by the hundreds
of billions of dollars that are transferred back and forth daily -
can become a political scandal. Theoretically insignificant scientific
discoveries can have agonizing medical results. Legal decisions
that hardly have any effect on other decisions in the legal system
itself can form road-blocks for entire political spheres. If law, for
example, brings the pharmaceutical industry and physicians under
the threat of liability to supply information and to establish
precautionary standards then this is something that can have
medical and also economic consequences that are entirely
unrelated to what is legally important and might not even
comprise part of the legal decision itself: the effects of anxiety,
uncertainties, increase of the necessity of experiments on animals,
the rise of costs or even the increase of the routine use of
experimental apparatuses. In all these circumstances there is no
supervening reason because every system can create resonance
only with its own code. This is something that follows almost
automatically if information triggers code-specific operations.
Furthermore, within their own domains, the function systems
depend on other functions being fulfilled elsewhere. Particular
deficits in performance can amount to unmanageable changes of
the social environment of other systems and thereby produce
disproportionate consequences. In this way, far-reaching economic
or political consequences may ensue if the legal system, for
whatever internal reasons, is incapable of developing rules for
the right to conduct labor disputes that enable the participants
to foresee the legal consequences of their behavior. The principle
of 'proportion' [Verhaltnissmassigkeit] within the legal system
may then have disproportionate consequences in relation to the
other systems. For similar reasons politically justified intervention
has, on more than one occasion, ruined entire economic domains
or made them dependent on constant political attention. The
relationship of politics and law shows the same thing in a
spectacular way. On the other hand, the stability of governments
depends on whether the economic production of wealth rises or
falls, i.e., on developments that are to a great extent out of their
control and which often do not work out as well or poorly for
the economy as they do for the political system.
For these reasons a much greater amount of resonance is more
likely to occur within society than to result from its relation to
the external environment. Function systems are differentiated,
coded and programmed for functionally specific high output.
They constantly scour their socially internal environment for
impulses and pick up what is offered to them. They are
endogenously restless and very sensitive. Their structural improba-

bility the incorporated risk can be released very easily. If
these systems describe themselves as an 'equilibrium' then this
also means that they have made instability their principle of
stability - or to formulate this in the terms of an eighteenth
century author: a kernel of corn thrown onto one of the scales
of a balance is enough to upset the system.1
The autonomy of the autopoiesis of the particular function
systems and the rejection of reciprocal substitutability are the
basis for the possibility of disproportionate reactions because
every system that is solely and completely responsible for its own
function regulates the conditions of the oscillation of resonance
independently. But at the same time it cannot control the
environmental occasions that trigger this. There seems to be no
general rule for such situations. They do not show up in all of
the intersystem relations, but in some of them more often than
in others. They fail to appear where they might be expected, for
example, in the relation between recent German divorce laws
and the willingness to get married - and they appear surprisingly
where they are unexpected. They can be observed and analysed.
They can be understood and described as a structural property
of modern society, but they cannot be anticipated.
In view of all this, the special position of the political system
of modern society in relation to the latter's ecological problems
would have to be examined more closely and become the subject
of empirical investigations. The political system's own method
resides in the production of collectively binding decisions. These
have no directly ecological, but only socially internal effects. They
facilitate and suppress communication, but at the same time this
system reacts very sensitively to itself, and although it cannot
regulate other systems via binding decisions it can influence them.
Under these circumstances it is highly probable that politics
becomes the place to start the business of addressing ecological
issues. Precisely because the system cannot do anything here
immediately it becomes increasingly probable that this is where
communication about ecological themes will find a home and
expand. There is nothing within the system to prevent this.
Viewed from a purely political point of view, there is nothing
that would correspond to legal, economic and scientific restraints
and would forthwith reduce communication to what is possible.

The system enables and promotes loose talk. As we can read in
the newspapers, nothing prevents a politician from demanding,
proposing or promising the ecological adjustment of the economy.
But a politician is not obliged to think and act economically, and
so does not operate at all within the very system that his or her
demand will ultimately bring to ruin.
Political communication is always concerned exclusively with
which political programs will or will not help the government
and the opposition to take over from the other. This is its code.
So communication cannot be taken to an illusory extreme because
this would be observed and judged by the voters. It must therefore
promote plausible decisions about law and money. It can do this
as long as the legal and economic systems provide the political
system with some room to maneuver. But at the same time as
we can see clearly it relies on the effectiveness of illusions and
carries out its business in this way.
Under these circumstances we must realize that politics is used
as a launching-pad and transmission system for ecological
desiderata when and wherever these enter the consciousness of
individuals and social communication. Then the political system
may function as a kind of continuous-flow heater. But this only
increases the probability that, on the occasion of the exposure
to ecological dangers, a socially internal intensification of
resonance will result that combines politically convenient and
acceptable solutions with functional disturbances in other systems.
Such an oscillation of resonance will probably have destructive
consequences within an evolutionarily highly improbable social
system, therefore any claim to political rationality would have to
include reactions to the effects of politics in its calculations.
17
Representation and
Self-observation
The 'New Social Movements'
It contradicts every principle of social differentiation to re-

establish the totality of the system within the system. The whole
cannot be a part of the whole at the same time. Any attempt of
this kind would merely create a difference in the system: the
difference of that part which represents the totality of the system
within the system vis-a-vis all the other parts. The presentation
of unity is a production of difference, thus the intention itself is
already paradoxical, self-contradictory.
Traditional societies however, have been able to live with this
paradox. The form of their internal differentiation was able to
accommodate it sufficiently. As long as these societies evolved
toward advanced culture they were differentiated either hierarch-
ically or according to center and periphery, mostly through a
combination of both. At least then there was no competition
with respect to the subsystem that represented the whole within
the whole. Only the apex of the hierarchy, only the highest
stratum, or only the center, the city and the urban form of
politically civil life came into consideration.1 Not until the Middle
Ages did discrepancies arise between aristocratic and urban life
that initiated a transformation process that led to functional
differentiation.
Besides, the traditional advanced cultures could always resort
to a religious justification of representation. This was not only
a mode of (always doubtful) legitimization for them but the
representation of the whole within the whole could also make
use of the code of religion to articulate the paradoxically created
difference. The difference that results from the attempt to

reintroduce the whole within the whole and to regulate the system
from an internal standpoint presents itself as the difference of
what is determined in this way and a transcendent indeterminacy.
'The split that cuts across human space is the generator of an
ultimate indeterminacy.'2 It was precisely the inaccessibility of
this indeterminacy that was used as a difference to create order
in this world. In the eighteenth century it became evident, in
view of the new, highly complex function systems, that all
'natural' representation rested on presumption and that religion
is misused if it persists in concealing this presumption. This
clarification reflects the transition from stratificatory to functional
differentiation. In the new order there are no natural primacies,
no privileged positions within the whole system and therefore no
position in the system which could establish the unity of the
system in relation to its environment.
Even if we concede all this, society nevertheless cannot do
without self-observation. Not all communication falls within the
context of the primary subsystems, for even if this were the case,
it could become the subject of communication in the next
moment. All order, every form of differentiation that is realized
in society, can also be observed and described in society. Every
binary code that excludes third possibilities for the coded
operations also makes it possible to introduce these third
possibilities. All reduction of complexity preserves complexity.
Other, unrealized selections are 'potentialized', transformed into
what is merely conceivable and preserved for communicative
reactivation.
Formally, the concept of observation indicates an operation
that designates other operations within the context of the
distinction of 'this instead of that'. Observation refers to
operations or complexes of operations (systems). It applies a
distinction as its own meaning-schema (for example, earlier/later,
useful/detrimental, fast/slow, system/environment) that is not
necessary and often inaccessible for the autopoiesis of the
operations itself. It brings much richer meaning-possibilities into
play to reduce them through selective designation. Therefore, on
the level of operations, autopoietic operation and observation
have to be distinguished by a scientific observer who applies this
theory, i.e., this distinction of operation and observation.4 On

the system level, however, at least for meaningfully operating,
conscious or social systems, one has to assume that these cannot
eliminate self-observation. Therefore, as soon as forms of
differentiation together with their consequences are revealed, it
is probable that they are observed and described in society itself.
The only question is, with the help of what distinctions?
Social self-observation has to be distinguished from other-
observation within society. Of course, system differentiation also
makes possible the observation of one subsystem by another. The
peasants observe the nobility, the nomads the settlers, the
politicians the economy, the law-makers politics, etc. One system's
distinctions especially its binary codes are thereby applied to
other systems that do not use these distinctions to observe. This
is nothing more than the normal technique of reduction in the
relation between system and environment, transferred to the
internal system level of subsystems and their environment. Social
self-observation occurs only on the condition that observation
does not distance itself from its object but co-intends itself.
The dogma of original sin was a schema of self-observation
unequaled and unsurpassed historically. It led, if not on the
psychological then at least on the communicative level, to moral
self-condemnation and therewith to a mitigation of moral
criticism. No one, for example, could recognize sin in another's
act of going to confession. Everyone had to do this. All classes,
even the clergy, were subject to this principle. It was designed to
be class-neutral and at the same time made it possible to work
out a class-specific catalogue of sins and dangers to salvation.
This was discussed as 'pollution' or 'hereditary pollution' of
souls. Only because this schema was increasingly undermined by
the personal attribution of guilt and the impossibility of knowing
an individual's state of grace could a religious moralism flourish
whose secular after-effects are still felt today. But a modern
functional equivalent for original sin is not on the horizon.
Since the nineteenth century the self-observation of society (as
an observation of this observation can determine) has been
connected with prominent consequences of function systems -
above all, with the resulting possibilities of the 'revolution' of
the political system and with the consequences of a money
economy (capitalism). This requires the causal attribution of these

consequences and leads, according to all the teachings of
attribution theory, to an irremediable conflict of attribution.5
The self-observation of society becomes ideological, i.e., in the
application of causal attributions dependent on valuations and
partisanship.
The semantics resulting from society's observation and descrip-
tion of itself has, in the meantime, become historical and is still
represented with the labels 'neo-' and 'post-'. In view of such
rapid changes the self-description of society is temporalized,
indeed contracted into a mere 'definition of the situation'.
Designations like 'industry', 'capitalism', 'modernity' are retained
but used only to characterize a historical difference. Of course,
this makes them more appropriate, but the question whether they
are valid or not becomes vacuous when one maintains that they
are no longer valid. This amounts to a weakly concealed paradox:
society is that which it is not and temporalization serves to
awaken the illusion that, despite this, it really means something.
With such standards of orientation 'theory deficits' should
come as no surprise. New kinds of social movements and social
protests look for new forms of articulation. Society is still viewed
as the cause and the object of protest, but the themes of protest
have clearly changed to ecological ones.6 Even the theme of peace
is viewed from this perspective in so far as it concerns armaments.
The politics of the military ought to be prevented from using
nature against humanity. In any event, the main concern is no
longer the legal theme of peace through authority.
The largely unorganized resonance of such themes is reducible
to the appearance of doubts in the experiences of daily life about
the meaning of nearly all function systems. The horizons of what
is possible have expanded so radically that every non-realization
must have causes in society. Intentions have unintended results
while good intentions have bad side-effects. Rationality appears
increasingly as perverted and in communication inevitably encoun-
ters mistrust and rejection. There is no doubt that such a mood
is fed by experience, but at the same time it is difficult to fix the
attribution to causes in a precise way, therefore the initial
situation generating the ideology no longer exists although
ideologies are functionally indispensable wherever an assessment

of values matters.
Perhaps these tendencies can best be characterized in reference
to the binary codes of the major function systems. Attempts are
made through human understanding to avoid the tension between
having and not having and to mollify the sharpness of the
difference between legal and illegal. Attempts are made to
bring the environment to bear against the functionally rational
codings of society. All in all, attempts are made to assume
the position of the excluded third (code value) and then to live
in society as the included excluded third (code value): as a
parasite.7
As far as sociological observation of this observation is
concerned it becomes attractive to imagine that this is ultimately
a protest against functional differentiation and its effects. But
even if this is the general name for a new kind of self-observation
of society, a corresponding self-description capable of fixing the
results is still lacking. The new social movements have no theory.
They are also incapable of controlling the distinctions in which
they record their observations. A simple, concrete fixing of goals
and postulates, a corresponding distinction of adherents and
opponents and a corresponding moral evaluation therefore
predominates. Implicit in this is the idea that a person has to be
able to live as he or she wants to even if in reduced circumstances
and foregoing luxuries (whereby these sacrifices are to a great
extent consensual because they are in agreement with the life-
world experiences of everyone anyway).
Even the so-called early socialists, not to mention Marx, had
much more to offer theoretically, but through a radical reduction
of society to economy they created for themselves an initial basis
that was much too narrow (and theoretically incapable of
realization). It would be unfair to measure this by contemporary
standards. Nevertheless, one can still see (from the position of
second order observation, of course), how this kind of social self-
observation operates with an inadequate semantics.
The most important effect is that observation cannot include
and reconstruct within its own concept that which it protests
against. This remains visible only as a resistance that results from
rejected valuations. Last but not least, this contains a renunciation

of its own semantics and structural stability that can be attained
- here again Marx is the great paradigm - through a theoretical
construction encompassing action and resistance. The blase moral
self-righteousness observed in the 'Green' movement conceals
only superficially the ever-possible relapse into resignation.
The problem seems to be that one has to recognize the dominant
social structure - whether seen as 'capitalism' or 'functional
differentiation' - to assume a position against it. Today this is
not as easy as in the nineteenth century because the hope for a
historical resolution of the difference, i.e., the hope for revolution,
no longer obtains. A functional equivalent for the theoretical
construct 'dialectics/revolution' is not in sight and therefore it is
not clear what function a critical self-observation of society
within society could fulfill. Nothing more than a resigned
comment on decline in the manner of Adorno or Gehlen is
detectable, and such positions offer hardly anything to hold on
to.
Fortunately, there is an essential connection between modern
society's (semantic) deficits of self-description and the (structural)
system-form 'social movement'. As a position for the description
of society within society the movement places itself in difference
to society. It seeks to affect society from within society as if it
occurred from outside. This paradox creates the instability of the
observation position and the dynamics of the social movement
makes allowances for this without realizing it. This may very
well lead to changes, to semantic or structural results that in one
way or another come to terms with the facts. Like the 'Reds'
(liberal theologians, according to Harnack) the 'Greens' will also
lose color as soon as they assume office and find themselves
confronted with all the red tape. This outlook may serve to
appease 'conservative' observers, but it ought not to conceal the
fact that the real problem resides in the question of whether
modern society is too dependent for self-description on the
entirely inadequate basis of social movements.
17
Anxiety, Morality and Theory
On the level of its autopoietic operations modern society is bound

to a functional differentiation, coding and programming, which
in turn can be seen and judged critically on the level of self-
observation. But since this society cannot represent itself within
itself it cannot bestow the normative sense for which one could
at least assume, if not attain, a thorough-going consensus.
Therefore, a self-observation cannot, as with the prophets who
enjoyed a privileged position, remind us of what is essential and
lament its downfall. Instead, anxiety is chosen as a theme,1 to
replace the difference of norm and deviation. This leads to a new
style of morality based on a common interest in the alleviation
of anxiety, not on norms for which all that matters is the
avoidance (or regulation, or regret) of deviation so that people
can live without anxiety.
This structure is closely connected with the differentiation of
coding and programming discussed in detail in the preceding
chapters. Throughout society that which is right or correct is
assigned to function systems and articulated exclusively as
interchangeable programs that organize the assignment of the
values of the respective codes. Anxiety, then, becomes the
functional equivalent for the bestowal of normative sense; and a
valid one since anxiety (as opposed to fear) cannot be regulated
away by any of the function systems. Panic cannot be prohibited
as Shaftesbury had already recognized in his day.2 Anxiety
cannot be regulated legally nor contradicted scientifically.
Attempts at a scientific clarification3 of the complicated structure
128 Anxiety, Morality and Theory
of the problems of risk and certainty only supply anxiety with

new nourishment and arguments.4 One can attempt to buy off
or to compensate anxiety with money. But whoever does so only
indicates that he or she had no anxiety: the commodity
disintegrates at the conclusion of the agreement. Religion, too,
would only devalue itself if it tried to present itself as a means
of removing anxiety. As its history shows, religion merely transfers
the anxiety to other domains of meaning.
So anxiety cannot be controlled from the function systems, but
is protected against all of them. As a matter of fact, better
functional performance can go hand in hand with increased
anxiety without being able to remove it. 5
In this way anxiety does not have to be present at all. The j
communication of anxiety is always authentic since a person can
attest to personal suffering of anxiety without fear of contradic- 1
tion. This makes anxiety an attractive theme for the kind of

communication that wants to observe function systems and
describe them from outside themselves but still from within
society. Anxiety resists any kind of critique of pure reason. It is
the modern apriorism - not empirical but transcendental; the
principle that never fails when all other principles do. It is an
'Eigen-behavior' that survives all recursive tests;6 one that seems
to have a great political and moral future. We are fortunate,
then, that the rhetoric of anxiety is not in the position to create
anxiety in fact, 7 even if it remains a disturbing factor within
society.
In many respects the rhetoric of anxiety is not a new
phenomenon. Its political use, directed against internal and
external enemies, has been known for a long time,8 but the new
ecological themes have changed its direction and displaced the
difference of the friend/foe schema into a system/environment
perspective. To the extent that war amounts to an intentionally
produced ecological catastrophe the anxiety over war suppresses
the anxiety over the enemy. The old social differentiations of the
national, class and ideological type that used to lead to war lose
their power to convince and are replaced by reduced tendencies
toward regional or cultural ethnogeneses. In reference to the 'real'
problems of our time new social solidarities are extolled and
postulated morally with great emphasis.
Above all, the new themes of anxiety enjoy a new property: a

person need have no anxiety over showing anxiety. They are
therefore capable of being extended. No one who shows anxiety
in 'crises' or over ecological developments, the consequences of
technology and the like, appears in a negative light, for there is
no individual capacity with which the danger could be met. This
enables opinion polls to record the increase of anxiety and direct
its results back into public communication. In this regard talk of
the age of 'unconcealed anxiety' has gained currency9 and anxiety
can claim to be universal: volonte generale.
Furthermore, it is remarkable that this type of anxiety inherited
not only non-contradictability but also its paradoxical constitution
from the rhetoric of principles. In other words, the attempt to
ease anxiety increases it. 10 Even official policies and the constant
concern with the improvement of circumstances can increase
anxiety, for example, increasingly detailed packaging labels on
medicines or intensive research and reports in the chemistry of
the health sciences lead to the impression that nothing is harmless
and everything is contaminated. The psychological basis for this
paradoxical effect seems to be that highly improbable risks are
overestimated and that risks to which a person is unavoidably
exposed are thought greater than those to which a person exposes
oneself intentionally.11 Above all, one has to realize that
communication about anxiety makes communication about
anxiety possible and in this sense is self-inducive. One can always
take a position with respect to anxiety. Some speak of 'hysteria'
while others of 'appeasement' - and both sides are supposedly
right.
All this reveals that not only socially dominant, functionally
related communication but also anxiety-related communication
is a principle of resonance that emphasizes certain things and de-
emphasizes others. This difference is only intensified by a public
rhetoric of anxiety. This rhetoric assumes the task of effecting
anxiety (which does not show up by itself, is not self-evident).
To do this it must proceed selectively.12 So, today, there is great
anxiety over nuclear dangers,13 but hardly any against medically
induced diseases, and there is an absence, at least in public
communication, of anxiety over the anxiety of others. The rhetoric
of anxiety is selective because it emphasizes the development of
what is worse and says nothing about any progress that has been
made, for example, in the medicine of the health sciences.14
Through public rhetoric anxiety has become stylized as the
principle of self-observation. Whoever suffers anxiety is morally
in the right, particularly if it is anxiety on behalf of others and
this can be assigned to a recognized non-pathological type.
Despite these clearly semantic contours no (sub)system can be
differentiated out for the management of anxiety. Even in view
of the tactful, considerate, understanding treatment of this
syndrome it remains to be seen to what extent this is merely a
matter of 'pluralistic ignorance'. 15 If no one really feared a
radioactive contamination of the water-supply but everyone
assumed that others feared this and, consequently, accepted this
argument, then how could anyone detect that the anxiety was
not contrived?
Of course, the social problem lies less in the psychical reality
of anxiety than in its communicative actuality. If anxiety is
communicated and is not contested in the communicative process
it acquires a moral existence. It becomes a duty to worry and a
right to expect participation in fears and to require standards for
defense against danger. Therefore, those who worry about
ecological matters do not, like Noah once did, equip just their
own ark with the necessary material for later evolution. They
become warners - with all the risks that this implies.16 In this
way anxiety infuses ecological communication with morality and
controversies become impossible to make decisions on because of
their polemical origin. Only the future can indicate whether the
anxiety had been justified - but the future is constituted anew
in every present.
Against a morality that propagates anxiety-related distinctions
theoretical analyses are in a difficult position. Anxiety, since it
transforms the uncertainty of the situation into the certainty of
anxiety, is a self-justified principle that needs no theoretical
foundation. It can, and indeed quite rightly does, ascribe theories
to the function system of science and distinguish, accordingly,
whether they sympathize with the anxiety or not. As a result of
the long-standing apriorism of reason, the position from which
an anxiety-based rhetoric and morality makes its observations
enjoys an unassailable self-certainty.
On the other hand it is difficult to see how from this vantage-

point the relation of society to its environment can be drastically
improved. Even anxiety limits and intensifies resonance. In fact,
it is more likely to stop the effects of society on its environment,
but it has to pay for this by risking unforeseeable internal
reactions that again produce anxiety.
If observing means distinguishing and designating one should
begin from the distinguishing capacity, i.e., to distinguish
distinctions. The systems-theoretical distinction of system and
environment, treated consistently, aims precisely at the ecological
problematic. With the help of the concept of 're-entry', it permits
the formulation of a concept of rationality.17 Accordingly, a
system attains rationality to the extent that it reintroduces the
difference of system and environment within the system and is
not guided by its (own) identity but by difference. Measured by
this criterion, ecological rationality would be attained when
society could charge the reactions to its environmental effects to
itself. This principle would then have to be reformulated with a
corresponding system-reference for every function system in
society. It would also have to be noted that all these rationalities
could not be added together because every function system
calculates its own rationality and treats the rest of society as
environment.
There are many reasons for handling this idea carefully not
the least of which is that it is only a scientific theory, i.e.,
according to its own self-description it is derived from one of
the function systems. Only if one realizes this can it also be
interpreted as a scientifically considered proposal for a self-
observation and self-description of society.18 No socially thera-
peutic effect to counter anxiety can be expected from it. Indeed,
it would be highly questionable if relief from pressing problems
were to be sought in a theory's narcotic of abstraction, but one
should not fear that a sensible handling of systems-theoretical
analyses will produce this. They lead more to the expansion of
the perspectives of problems than to their suppression. Whoever
doubts this ought to go back and read this book a second time.
That one should not expect recipes here - as if a net rational
improvement could be attained from the closing of nuclear plants
or from constitutional reforms effecting a change in the majority
rules - is, therefore, self-evident. On the other hand, the

alternatives that the rhetoric of anxiety offers enjoy the advantage
of being near at hand, even if they are unrealistic. In a way
that is almost impossible to explain they obfuscate social
interdependencies and the mediation of effects.
One has to concede, however, that both are highly topical
possibilities of our society's observation of itself, and one should
also like to hope that they could be brought into a communicative
relation.
17
Toward a Rationality of
Ecological Communication
Anyone who hoped that these reflections on the theme of

'ecological communication' would clarify how this kind of
communication could contribute to the solution of pressing
environmental problems will be disappointed. Our aim was to
work out how society reacts to environmental problems, not how
it ought to or has to react if it wants to improve its relation with
the environment. Prescriptions of this sort are not hard to supply.
All that is necessary is to consume fewer resources, burn off less
waste gas in the air, produce fewer children. But whoever puts
the problem this way does not reckon witli society, or else
interprets society like an actor who needs instruction and
exhortation (and this error is concealed by the fact that he or
she does not speak of society but of persons).
We have also forced a second reservation on ourselves one
regarding criticism. Ordinarily, criticism presupposes that one
knows how things are to be improved and then finds fault with
the fact that this is not the case. In the reflexive understanding
of the Frankfurt School this advance knowledge is no longer
assumed but replaced by the dream of the subject or, since
Habermas, by the idea that society can determine its own identity
in communicative discourse in such a way that this commits the
participating subjects 'internally' and binds them to the collective
identity. In the self-evidence of the grounds of the validity-claims
the collective identity is given in a way that convinces every
participant in the communication; and obviously not only in the
particularity of the respective actions, situations and grounds but
134 Rationality of Ecological Communication
with reference to identity too. If identity were to be interpreted

merely as the system's description of itself, the agreement of
subjective and collective identity that creates the certainty of the
identity's rationality would be lacking. The systems-theory that
maintains that 'modern societies cannot develop a rational identity
at all lacks any reference point for a critique of modernity'.1 But
this means only that the completed project of modernity, if it is
presented in this way, is not carried forward by systems theory.
This should not be a bone of contention. The critique of this
semantics of modernity says that if this is modern society's
ambition then it cannot be grasped sufficiently or, if so, it can
only be conceived as failure! It follows that it is hardly imaginable
that contemporary societyx-has got into ecological difficulties
because it has not taken the project of modernity seriously enough
or has not discussed validity-claims and grounds sufficiently.!
At the same time, Habermas's suggestion remains true: a
centerless society cannot assert a rationality of its own but has
to rely on the subsystem rationalities of its function systems. This
is also true if the protest directed against it is considered, for this
too can only be a partial phenomenon and can neither be nor
represent the whole within the whole. At best, it produces a
corrosive mistrust without a rationality of its own to which
others, then have to react. From the point of view of systems
theory this would have to be the case if rational individuals
anywhere came to an agreement about their validity-claims. For
why should it be left to them to determine what is good when
the starting-point has to be that others do not have valid grounds
to disagree with the procedure and even less with the proposed
consensus. ,
The answer to the question of the system's rationality (and
also to the question of the rationality of ecological communication)
must therefore be a change in the formulation of the problem. 2:
In every assumption of a differentiated unity there is inevitably
a paradox because the unity of the whole is not outside or above
the parts but is identical and not identical with the sum of them
at the same time. Besides, we know that it is not possible (or
only for specific purposes) to resolve this paradox through a
differentiation of levels or a hierarchy of types.3 One can also
show that in modern society every claim of a part to be the
whole or to represent identity is subject to observation and

contradiction since, for reasons of social structure, there are no
longer any non-competitive positions, for example, the apex of
the hierarchy or a center vis-a-vis a periphery. It is ill-advised
and leads to a peculiar Utopianism and hopelessness if one does
not recognize these restricting conditions on any attempt at
rationality but persists in a belief in direct access to it.
\ So falls the idea that the environment has a partner in society
or even that humanity itself is this partner. This is just a new j
version of the privilege of representing the whole within the
whole because 'the' environment is the correlate of 'the' system
and can be seen as a unity only from the unity of the system.
Even enthusiasm and the awareness of responsibility cannot"
privilege anyone in this way.l Only as a unity - and this means
as a differentiate^ unity - can society react to its environment^
Besides, since nond of its individual function systems is organized
and capable of making decisions as a unity, an organizational
co-ordination is not even attainable.
In addition to these difficulties we are faced with a further
problem that the preceding chapters indicated and clarified with
the distinction of coding and programming./The unity of every
function system resides in being guided by a bjnary code valid
for. itself alone. Its unity is its difference, incleecf\ difference that
robsihe system of the possibility of placing itself on the 'right'
side.: This also eliminates teleological rationalities (for example,
rationalities of action) that would enable the system to characterize
itself as a striving for truth, right, power, wealth, education or
the moral life and to be taken, at least as far as good intentions
are concerned, as rational. Instead, the difficult questions
concerning what really constitutes the unity of the difference
guiding such codes and what really constitutes the rationality of
a distinction have to be answered. Whatever appears as 'right'
within such a code depends on the coded acquisition and
processing of information and has meaning in reference to the
contingency that is disclosed and structured by this. Only whatever
is capable of being its opposite according to a given schema
under the condition of the exclusion of a third alternative can
be right, and since we must assume this schema as functionally
specific, a direct inference from it to social rationality is impossible.
If this has been an accurate description of the situation then

the problems of the rationality of society as a whole have to be
approached in an entirely new manner. Whoever still localizes
rationality in the reflexivity of reason - for example, like
Habermas, in the reflexivity of a discursively ascertained ration-
ality4 will find it impossible from now on to discover rationality
either here or in what follows. The traditional conceptual
arrangement should not immediately exclude the consideration
of how a rational society would have to be conceived if (1) the
concept of self-reference (reflexivity) is transferred to all empirical,
autopoietic systems; (2) the inference from self-reference to
rationality is surrendered; (3) rationality can no longer be found
in the self-reference of reason; and (4) all considerations of
rationality are required to agree with the paradox of differentiation
and the binary coding of the function systems.5
Obviously, social rationality can reside neither in the projection
of rationality of individual function systems (even science's) nor
in its total rejection as irrational. It has to be conceived, as it
were, free of any specific location - as a difference that can be
realized differently. A sufficiently general, at least hitherto
unsurpassed concept for this resides in the generalization of the
method of functional analysis when this is conceived as a method
for the creation of difference. Its status as rigorously scientific is
disputed and is certainly in need of restricting conditions (perhaps
from specific theories like systems theory) or from formal
requirements (perhaps those of mathematics). The general rule
of beginning from reference problems and looking for their
functional equivalents can be seen, to a greater extent, as a
generalizable principle that accepts unity only as a problem, i.e.,
only for the sake of the difference that can be created through
it. Even with all the practical difficulties involved in finding
functional equivalents for everything given, many possibilities of
practicing this form of orientation in other function systems, even
without the aegis of science, i.e., originarily, become clear.
This changes the focus of all previous theories of reflection
from unity to difference and enables them to acquire information
in ways that differ from those accepted previously^ Until now,
the reflection used by function systems, even where it existed in
a theoretical form, was guided by the values of correctness and
sought the unity of the system in thes^)Because these theories of

reflection are exposed to a functionally oriented comparison and
family likenesses that are reducible to the functional differentiation
of the social system, a new situation is created that upsets their
normative and evaluative self-certainty. A functional reanalysis
would have to begin from the system's description of itself within
the system. Then it would very quickly realize that every self-
description simplifies the system that it describes in terms of a
model, i.e., reduces complexity and creates difference: the
difference between the system it describes and its description of
itself. Every reflection creates an observation of the reflection
that works selectively perhaps in the sense of a recursive sorting-
out of the observation's semantic commitments to recall or in
the sense of the self-referential systems' 'own values'.6 These are
only examples of theories of reflection about function systems,
formed under guide-lines that try to attain a large measure of
environmental openness in the function systems and, in this way,
represent society for each of them.
Social rationality would naturally require that the ecological
difference of society and its external environment is reintroduced
within society and used as the main difference. We have to begin
from the idea that there can be no privileged place for this, no
authoritative organization, consequently no 'constitution' that
could transform the ecological difference into binding guide-lines
for further information processing. If such a place were to be
created the result would be a new internal difference in society:
the difference of this place from all the others in society. One
can promote this idea as a self-justifying Utopia. But then
wouldn't this be just another version of a self-justifying reason?
Because of the paradox of a differentiated unity, of a unitas
multiplex, its implementation would fail for all social formations,
and not simply because of functional differentiation!. The tautology
of rationality - correct is what corresponds to the concept of
what is correct - would suddenly turn into a paradox: correct
is what is impossible because it presupposes society only as a
unity and not as difference^ To formulate it more mundanely,
this Utopia entails foreseeable difficulties of implementation. This
problem is repeated, of course, on all levels of internal social
system-formation, but within the subsystems the possibility exists
of a hierarchical organization through which the difference of

system and environment can be transformed into internal system
directives.
These considerations leave the concept of system rationality
intact. This signifies the possibility of reintroducing the difference
of system and environment within the system, thus the possibility
of directing the system's information processing by means of the
unity of the difference of system and environment. The unity of
the difference of system and environment is the world. Within
differentiated systems, however, this reference to the world is
filtered not only by the external boundaries of the encompassing
system but also by further internal boundaries. This is what the
conditions of 'Western rationalization' in Weber's sense hang on:
the rationalization of firms and administrations. At the same time
these conditions for construction mean that system rationality
increasingly loses its claim to be world rationality. Guidance
from internal system environments, for example, markets or
public opinion, begins to dominate. To the extent that system
rationality appears more realizable it becomes less world-rational
and even less socially rational, but once this becomes clear one
can also see that this is not a matter of an 'iron law' but rather
of the costs of increasingly improbable complexity. The
precondition of all concern with rationality is the proper
understanding why it is - and remains - improbable.; Then it
makes sense to be guided by the Utopia of rationalityr-to see
whether and how individual systems can be used to provide
solutions to problems that are more rational and include further
environments. Today it is already clear that communication about
ecological themes is beginning to examine such possibilities. \
However one assesses these possibilities, it should remain clear
that the concept of rationality offered here never designates
system-states, therefore it cannot solidify into desired end-states,
goals or the like. Nor is it a substantial or a teleological
rationality. System rationality is never concerned with unity but
with difference and with the resolution of all unity into difference.
So we have to ask where the rationality of a distinction arises,
considering what can be designated with its help? We suggest
that it is out of a reference to the ultimate difference of system
and environment, i.e., out of the ecological difference.
21
Environmental Ethics
In conclusion I would like to make some remarks concerning

perhaps the most prevalent expectation today that ecological
communication should culminate in ethical questions and find
its justification there. In view of the given social situation a
change of consciousness is necessary, a new ethics, a new
environmental ethics. We have already touched on this require-
ment at different places - and not been able to get very far. Our
investigations led us in an entirely different direction. The problem
of an environmental ethics cannot be covered sufficiently in a
few marginal remarks, so in place of a summary we would like
to clarify the difference between a systems-theoretical, sociological
analysis and ethics - in the hope that further communication will
be guided by this difference and not simply by ethical postulates
and maxims.
First, it is important to distinguish morality and ethics (knowing
quite well that it is a widespread linguistic custom to use these
terms interchangeably). Morality is to be understood as the
coding of communication by the binary scheme of good and bad
(or, if subjectivized, of good and evil). The code is always
applicable when the behavior that is the subject of communication
is sanctioned by the bestowal or withdrawal of esteem or
contempt.1 Therefore it can also be said that morality resides in
conditioning the attribution of esteem or contempt. In this way
'morality' is an artificial aggregate because it is never necessary
or possible for communication that the totality of the conditioning
viewpoints be available in a way that sharply distinguishes one
140 Environmental Ethics
from another. The formulation 'morality', therefore, always refers

to communication that is already moralized (and, of course, also
to non-moralized communication).
It is unavoidable that the observation of morality encounters
paradox in its object domain. This is not a specifically moral
problem. Instead, as is well known from logic, all binary coding
leads to paradox when the code is applied to itself. Sociological
analysis discloses this empirically and confirms the practical
relevance of paradox in this way.
On one hand, morality seems to have a polemical origin, i.e.,
it arises out of uncertainty, disunity and conflict because this is
the only way of providing an occasion for using esteem or
contempt as a sanction. One could even say that morality is
concerned with pathological instead of normal cases including
those of supererogatory performances and merits that appear as
if this could be expected of everyone - and, therefore, has to be
isolated by means of a special esteem given to a hero, martyr,
ascete, or virtuoso of virtue.
On the other hand, morality also leads to conflict, whether
this is concealed or open. A person committed to morality can
relent only with difficulty because his or her self-respect is thereby
placed in jeopardy. A person who considers him- or herself to
be morally responsible falls very easily into a bind. Such a person
must proceed very carefully - except in very limited circumstances
when morality (despite its pathological origin) has become self-
evident. In the case of dissent this problematic leads very quickly
to conflict, notwithstanding the similarly conflictual expedient:
that morality requires not only loving the good but also hating
and combating the bad. 2
We have to realize, logically as well as empirically, that morality
is paradoxical or, looked at temporally, that it causes paradox.
As the unity of the difference of good and bad it works both
well and poorly. What is good can be bad and what is bad can
be good. Thus the observer of morality finds observation blocked.
In any event, the observer cannot make any moral judgement
about morality. 'In other words, everything is moral. But morality
itself is not moral!' 3
Only ethics can make a moral judgement about morality. At
least it trusts itself to be able to do so. Ethics (no longer in an
ethos-related use, thus in a sense that is at present valid) is to be

understood as a reflection theory of morality. Its function is to
reflect the unity of the moral code, the unity of the difference of
good and bad. If moral difference raises the problem of its unity
(and is not taken simply as nature) it generates ethics. Ethics,
therefore, must undertake the task of eliminating the paradox of
the moral paradox if it wants to be a moral theory of morality.
It can do this only if it does not know what it does because the
elimination of the paradox is, of course, a paradoxical undertaking
itself. Therefore ethics has to establish an ersatz-problem that
enables it to conceal what is primarily of concern. It could not
tolerate proceeding, as it were, in a 'carefully careless' way as
social morality has prescribed since the seventeenth century. As
a theory of reflection it is too obligated to the disclosure of the
principle of the unity of difference, so it chooses the unity of a
rule that separates the entire domain of morality (good) from
itself as the ersatz problem - as in the case of the categorical
imperative.
We cannot and do not want to pursue the possibilities of the
construction of ethical theories here any longer.4 We must be
content with uncovering the problem in whose avoidance ethics
has its latent unity. The secret of ethics, its arcanum, the source
of all the experiences that it cannot name and to which it cannot
return is the paradox of moral coding. Thus, from the point
of view of its function, ethics fails in its task to warn against
morality. This is left to a much overworked sociology. But here,
too, an observing of observing, a second-order observing is
required.
These considerations are generally valid both for earlier
experiences with a social ethics that derives its problem from
humanity's treatment of humanity as well as for a possible
environmental ethics that resorts to moral conditioning, i.e., to
moral resonance. We will also have to consider whether the
problems of paradox and its avoidance do not become more
acute when viewed from the perspective of environmental ethics,
but since this ethics does not exist at present as a theory of
reflection this will be hard to tell. We cannot even exclude the
possibility that all ethical reflection fails because of the non-
moral nature of particular problems of risk. But we will not
consider this. Even so, we still have to realize that the discrepancy
between social problems and environmental problems will make
itself known formulated according to systems theory, this means
discrepancies between the society and its environment. In the
case of morality and ethics the concern, naturally, is with a social
regulation, but precisely because of this we will have to ask
whether the conditions and forms of this regulation do not have
to change if they are extended to an unrelated domain, to non-
social sources of problems. It would be premature, however, to
take this question out of circulation by saying that even ecological
problems are ultimately caused socially or, at least, are of interest
only in the context of ecological communication. However correct
this may be, an entirely new dimension of complexity comes into
play through the difference of system and environment, and it is
improbable that this complexity just like the internal social
complexity of double contingency could be transferred to the
conditions of esteem or contempt.
We must, therefore, ask the question whether, even under these
circumstances, an ethics that abstains from paradox will develop
and be able to be practiced with moral responsibility. This could
also give us pause to wonder whether it is not the recognition
of paradox that is the way for ethics to go to do justice to the
new problem situation - for, even in the case of theories, a more
complex problem situation changes the conditions of adequate
internal complexity. It could very well be that the digestive as
well as the ruminating apparatus of ethics will have to be
equipped with more stomachs above all with one for paradoxes.
In any event, as long as such an ethics does not exist, ecological
communication itself will have to respect its distance from
morality. At present it is falsely guided by the instructions of
environmental ethics. To be sure, ecological communication will
examine ethical possibilities too and perhaps be able to prepare
a field of development for their reformulation. But if anywhere,
it is in ecological communication that society places itself in
question, and we cannot see how ethics can dispense with this
and remain available as something that can be relied on. On the
contrary, if a specific function is to be attributed to environmental
ethics within the context of ecological communication then this
might very well be to remain cautious in dealing with morality.
Glossary
The text uses a series of concepts in a way that is peculiar to it

and with a precision that depends on complex preliminary
considerations. Since a sufficient justification for the use of these
concepts was not possible within the text I will present some
definitions with brief explanations here.
Autopoiesis Refers to (autopoietic) systems that reproduce all
the elementary components out of which they arise by means of
a network of these elements themselves and in this way distinguish
themselves from an environment - whether this takes the
form of life, consciousness or (in the case of social systems)
communication. Autopoiesis is the mode of reproduction of these
systems.
Code Codes arise out of a positive and a negative value and
enable the transformation of the one into the other. They come
into being through a duplication of a given reality and with this
offer a scheme for observations within which everything that is
observed appears as contingent, i.e., as possibly different.
Communication Designates not simply an act of utterance that
'transfers' information but an independent autopoietic operation
that combines three different selections - information, utterance
and understanding into an emergent unity that can serve as
the basis for further communication.
Complexity A state of affairs is complex when it arises out of
so many elements that these can only be related to one another
selectively. Therefore complexity always presupposes, both operat-
144 Glossary
ively as well as in observation, a reduction procedure that

establishes a model of selecting relations and provisionally
excludes, as mere possibilities, (i.e., potentializes) other possibilit-
ies of connecting elements together.
Coupling This concept designates the reciprocal dependency of
system and environment which can be seen by an observer if the
latter takes the distinction of system and environment as basic.
The observer can even be the system itself if it is in the position
to observe itself when it uses the distinction of system and
environment.
Differentiation, functional In the text this concept refers to the
formation of systems within systems. It does not necessarily
designate the decomposition of an entire system into subsystems
but rather the establishment of system/environment differences
within systems. The differentiation is functional in so far as the
subsystem acquires its identity through the fulfillment of a
function for the entire system. \ f V- * '
Ecology Means in this context the totality of scientific investi-
gations that concern themselves, on whatever level of system
formation, with the consequences of the differentiation of system
and environment for the system's environment. The concept does
not presuppose any specific kind of system (ecosystem).
Elimination, of the paradox see Paradox.
Observation Is defined on the level of abstraction of the concept

of autopoiesis. It designates the unity of an operation that makes
a distinction in order to indicate one or the other side of this
distinction. Its mode of operation can, again, be life, consciousness
or communication.
Paradox A paradox occurs when the conditions of the possibility
of an operation are at the same time the conditions of the
impossibility of this operation. Since all self-referential systems
having the possibility of negating create paradoxes that block
their own operations (for example, can determine themselves only
in reference to what they are not, even if they themselves and
nothing else are this non-being) they have to foresee possibilities
of eliminating the paradox and at the same time disguise the
Glossary 145
operations necessary for this. For example, they have to be

able to treat the recursive symmetry of their self-reference
asymmetrically, either temporally or hierarchically, without being
able to admit to themselves that an operation of the system itself
is necessary for this transformation.
Program Refers to that of codes and, following a well-established

conceptual usage (canon, criterion, regula), designates the con-
ditions under which the positive or negative value of a specific
code can be ascribed to situations or events. In social systems
this is treated as a question of a decision (thus also decision-
programs) between true and false, legal and illegal etc.
Redundancy The multiple certification of a function, therefore

the appearance of 'superfluity'. The rejection of redundancy
means that multifunctional mechanisms have to be replaced by
functionally specific ones that are applied to (autopoietic) self-
certification.
Representation Is used to designate the presentation of the unity

of a system by a part of it (repraesentatio identitatis) and is
distinct from the rigorously legal sense that has legally effective
substitution in mind. In so far, representation is always paradoxi-
cal: intending to present unity, it creates a difference between
the representing and the other parts of the system.
Resonance Signifies that systems can react to environmental

events only in accordance with their own structure.
Self-reference Designates every operation that refers to some-

thing beyond itself and through this back to itself. Pure self-
reference that does not take this detour through what is external
to itself would amount to a tautology. Real operations or systems
depend on an 'unfolding' or de-tautologization of this tautology
because only then can they grasp that they are possible in a real
environment only in a restricted, non-arbitrary way.
Social Systems A social system comes into being whenever an

autopoietic connection of communications occurs and distin-
guishes itself against an environment by restricting the appropriate
communications. Accordingly, social systems are not comprised
146 Glossary
of persons and actions but of communications.

Society That social system which includes all meaningful com-
munication and is always formed when communication takes
place in connection with earlier communication or in reference
to subsequent communication (i.e., autopoietically).
Notes
1. Niklas Luhmann, Soziale Systeme (Social Systems), Frankfurt:
Suhrkamp, 1984.
2. Jiirgen Habermas/Niklas Luhmann, Theorie der Gesellscbaft oder
Sozialtechnologie: was leistet die systemforschung? (Theory of
Society or Social Technology: What Does Systems Research
Accomplish?), Frankfurt: Suhrkamp, 1971.
3. Talcott Parsons, Social Systems and the Evolution of Action
Theory, New York: Free Press, 1977, p. 118.
4. Talcott Parsons et al., Toward a General Theory of Action,
Cambridge, Mass.: Harvard University Press, 1951, p. 16.
5. Cf. Soziale Systeme, 1984, p. 93.
6. Hessicher Rundfunk, 7 May 1987.
Preface
1. The text of the address has been published in the Proceedings of
the Academy under the given title, RWAkW G278, Opladen 1985.
Chapter 1 Sociological Abstinence

1. i write out of a sense of alarm,' admits John Passmore, Man's
Reponsibility for Nature: Ecological Problems and Western
Traditions, New York 1974, p. IX, and he undoubtedly speaks
for many ecological writers.
2. For a summary cf., Josef Miiller, 'Umweltveranderungen durch
den Menschen', in Karl Heinz Kreeb, Okologie und menschliche
Umwelt: Geschichte - Bedeutung - Zukunftsaspekte, Stuttgart
1 9 7 9 , pp. 8 - 6 9 .
3. Cf., the last two chapters in Amand L. Mauss, Social Problems as
148 Notes
Social Movements, Philadelphia 1975, which are preceded by an

extensive enumeration of problem domains.
4. In unpublished writings on the early history of liberalism Stephen
Holmes calls this 'antonym substitution'.
5. For the absence of a developed 'milieu' concept in the eighteenth
century cf., Georges Canguillem, La Connaissance de la vie, 2nd
edn, Paris 1965, pp. 129ff. Cf., also Jiirgen Feldhoff, 'Milieu', in
Historisches Worterbuch der Philosopbie, vol. 5, Basel-Stuttgart
1980, pp. 1393-5; and Leo Spitzer, '"Milieu" and "Ambience": An
Essay in Historical Semantics' in Philosophy and Phenomenological
Research, vol. 3 (1942), pp. 1 - 4 2 , 169-218.
6. Cf., Richard Hofstadter, Social Darwinism in American Thought
1860-1915, Philadelphia 1945; Emench K. Francis, 'Darwins
Evolutionstheorie und der Sozialdarwinismus', in Kolner Zeitschrift
fur Soziologie und Sozialpsychologie, vol. 33 (1981), pp. 20928;
Niles Eldrege/Ian Tattersall, The Myths of Human Evolution, New
York 1982; Walter Buhl, 'Gibt es eine soziale Evolution?', in
Zeitschrift fur Politik, vol. 31 (1984), pp. 3 0 2 - 3 2 .
7. Programmatically in 'Zu einer allgemeinen Systemlehre', Biologia
Generalis, vol. 19 (1949). pp. 114-29, and with a far-reaching
effect in the context of the English language. Cf., Ludwig von
Bertalanffy, General Systems Theory: Foundation, Development,
Applications, London 1971. For a historical estimation cf., also I.
V. Blauberg/V. N. Sadovsky/E. G. Yudin, Systems Theory:
Philosophical and Methodological Problems, Moscow 1977.
8. Cf., for example, Walter Buckley, Sociology and Modern Systems
Theory, Englewood Cliffs, N.J. 1967; Kenneth F. Berrien, General
and Social Systems, New Brunswick N.J. 1968. In the important
anthology that documents the trend - Walter Buckley (ed.), Modern
Systems Research for the Behavioral Scientist, Chicago 1968 -
sociologists are hardly mentioned. Anyone familiar with the
theoretical history of sociology knows that this is connected with
the dominating influence of the structurally functional paradigm
of that time. Its main representative, Talcott Parsons, had accepted
some suggestions from general systems theory but developed a
theory in which environments played a role only as systemically
'internal'.
9. Especially since Tom Burns/G. M. Stalker, The Management of
Innovation, London 1961; and Paul R. Lawrence/Jay W. Lorsch,
Organization and Environment: Managing Differentiation and
Innovation, Boston 1967. Among the many textbooks cf., also
Notes 149
Howard E. Aldrich, Organizations and Environments, Englewood
Cliffs, N.J. 1979.
10. Moreover, the same is true for the theory of the political system,
above all for David Easton, A Systems Analysis of Political Life,
New York 1965.
11. Walter L. Buhl, 'Das okologische Paradigma in der Soziologie', in
Harald Niemeyer (ed.), Soziale Beziebunsgeflechte: Festschrift fur
Hans Winkmann, Berlin 1980, pp. 9 7 - 1 2 2 ; and Buhl, Okologische
Knappheit: gesellschaftliche and technologische Bedingungen ihrer
Bewaltigung, Gottingen 1981, p. 35, emphasizes this in view of
the superficiality of previous systems analyses. Of course, with
very different possibilities aimed at the ecosystem in mind.
12. Cf., Niklas Luhmann, Soziale Systeme: Grundriss einer allgemeinen
Theorie, Frankfurt 1984.
13. Cf., Lvnne White Jr, 'The Historical Roots of Our Ecological
Crisis', in Science, vol. l 3 5 (1967), pp. 1203-7; reprinted with
critical replies in Ian G. Barbour (ed.), Western Man and
Environmental Ethics: Attitudes Toward Nature and Technology,
Reading Mass. 1973, pp. 18-30.
14. Cf., Giinther Altner, '1st die Ausbeutung der Natur im christlichen
Denken begriindet?' in Hans Dietrich Englehardt et. al. (eds),
Umweltstrategie: Materialien und Analysen zu einer Umweltethik
der Industriegesellschaft, Giitersloh 1975, pp. 3 3 - 4 7 ; Robin
Attfield, 'Christian Attitudes to Nature', in Journal of the History
of Ideas, vol. 44 (1983), pp. 3 6 9 - 8 6 ; Attfield, The Ethics of
Environmental Concern, Oxford 1983.
15. A typical argument for the dominance of ethical viewpoints,
'results because the exclusivity of scientific, technological and
economic viewpoints have led to the crisis (sic\) described above'
(according to Heinhard Steiger, 'Begriff und Geltunsebenen des
Umweltrechts', in Jiirgen Salzwedel (ed.), Grundziige des Umwel-
trechts, Berlin 1982, pp. 1 - 2 0 , vol. 13 - as if such an exclusivity
had always existed! But even the sophisticated analyses of Hans
Jonas, Das Prinzip Verantwortung: Versuch einer Ethik fur die
technologischen Zivilisation, Frankfurt 1979, has inadequate
historical analyses because of its desire to make a clear contrast.
16. This remarkable inconsistency has also occurred to Louis Dumont.
Cf., Essai sur Vindividualisme: Une perspective anthropologique
sur I'ideologie moderne, Paris 1983, p. 203.
17. It produces considerable confusion when a widespread linguistic
custom, whose consistent application would have to lead to
150 Notes
economizing on the use of the concept of ecology, designates

ecological interdependencies or 'balances' as 'system' (ecosystem).
For example, according to Heinz Ellenberg, 'Ziele und Stand der
Okosystemforschung', in Ellenberg (ed.), Okosystemforschung,
Berlin 1973, pp. 1 - 3 1 , vol. 1, 'An eco-system is a context of
interaction [Wirkungsgefiige} including living beings and their
inanimate environment that is open but also capable of self-
regulation to a certain degree . . . Ecological systems are always
open, i.e., capable of being disturbed from outside and without
sharp boundaries.' The same is true according to Kreeb, ibid.,
(1979) and the prevailing opinion. But not every interconnection
is a system. A system exists only when an interconnection
distinguishes itself from an environment. (Exactly the opposite of
Buhl, ibid., 1980, p. 121. In this sense, for example, one can speak
of the physical system of the planet Earth in which the human
organism, the vocal transmission of human communication, the
microphysics of the human ear, etc. participate. This designates a
systems-theoretical but not an ecological problematic. In distinction
to a simple systems-theoretical problematic, a problematic is called
ecological only when it aims at unity despite difference or even at
unity through difference, i.e., because a system/environment
interconnection is structured through the system separating itself
from its environment, differentiating itself from it and on this
basis developing a highly selective behavior toward it. So the
ecological problematic cuts across the systems-theoretical problem-
atic (which, of course, does not exclude that investigations in the
one perspective cannot be relevant for the other). For the ecology
of human society countless systems are relevant (perhaps the closed
system of genetic inheritance) without the unity of this system and
its environment identifying with the ecology of society, i.e., with
the system/environment relation of society. At present it is an open
question, discussed under the title of 'sociobiology', whether and
to what extent the handing-down of traits is relevant for the
system of society and its environment.
18. Cf., Spitzer, ibid., especially the first part.

19. Theoretical motifs like this that require a boundary line in the
world for the self-observation and reflection of the world have
appeared in the philosophy of reflection and in cybernetics in
connection with Wittgenstein. Cf., for example, Gotthard Giinther,
'Cybernetic Ontology and Transjunctional Operations', in Giinther,
Beitrage zur Grundlegung einer operationsfahigen Dialektik, vol.
1, Hamburg 1976, pp. 2 4 9 - 3 2 8 (especially pp. 318ff.). Sociology
Notes 151
has only reached the point of self-fulfilling or self-defeating

prophecies.
Chapter 2 Causes and Responsibilities?
1. In a somewhat different sense - in reference to problems of scarcity

and their temporal operationalizations - Guido Calabresi/Philip
Bobbitt, Tragical Choices, New York 1978, speak of tragic
decisions. I think it comes closer to the classical concept of what
is tragic if one turns attention to the participation in causality.
One can always find the ultimate 'ground' of the tragic in its
providing 'too few' causal possibilities.
2. Cf., for example, Eckhard Rehbinder, Politische und rechtliche
Probleme des Verursachersprinzips, Berlin 1973; Dieter Cansier,
'Die Forderung des umweltfreundlichen technischen Fortschritts
durch die Anwendung des Verursachersprinzips', in JaJldlli!lJii^
Sozia111>issenschaftijuA^2.9. (1978), pp. 145-63; Robert Weimar,
'Zur Funktionalitat der Umweltgesetzgebung im industriellen
Wachstumsprozess', in Festschrift Bruno Gleitze, Berlin 1978, pp.
51126 (519ff.). To a great extent jurists focus on prejudices
about attribution and the real problematic arises for them only
with the question of whether there should be a further restriction
to additional legal consequences, perhaps by raising the issue of
guilt. This, of course, is unavoidable if punishment comes into
consideration. For economists, however, it is clear that the causer-
principle is easy to regulate technically but does not work optimally
for allocation. In a certain way, this reservation, too, indicates
that attribution rests on a simplification.
3. The selection is guided by 'how the greatest environmental quality
can be attained and which procedure appears as the economically
and administratively favorable solution,' according to Eckard
Rehbinder, 'Allgemeines Umweltrecht', in Jiirgen Salzwedel (ed.),
Grundzuge des Umweltrechts, Berlin 1982, pp. 8 1 - 1 1 5 (96ff.).
Cf., also Rehbinder, ibid., (1973), pp. 33ff. In other words, the
causer is whoever one can catch.
4. This means, for example, that the discussion of the idea that
'capitalism' and the unrestricted use of the profit motive are the
real causes of environmental damage is just as correct or incorrect
as any one-factor theory. Cf., for example, Gerhard Kade, 'Umwelt:
Durch das Profitmotiv in die Katastrophe', in Regina Molitor (ed.).
Kontaktstudium Okonomie und Gesellschaft, Frankfurt 1972, pp.
152 Notes
2 3 7 - 4 7 , or the contributions of Gerhard Kade and Volker Ronge

in Manfred Glagow (ed.), Umweltgefahrdung und Gesellschaftssy-
stem, Munich 1972. It needs no particular mention that greatly
differentiated points of departure for ecological analyses are to be
found in the work of Karl Marx. Cf., for example, Peter A. Victor,
'Economics and the Challenge of Environmental Issues', in Herman
Daly (ed.), Economy, Ecology, Ethics: Essays Towards a Steady -
State Economy, San Francisco 1980, pp. 1 9 4 - 2 4 0 (207ff.).
5. This is very clear in Heinz von Foerster, 'Cybernetics of Cyberne-
tics', in Klaus Krippendorff (ed.), Communication and Control in
Society, New York 1979, pp. 5 - 8 .
6. Cf., Walter Benjamin, 'Zur Kritik der Gewalt', in Benjamin,
Gesammelte Schriften, vol, II. 1, Frankfurt 1977, pp. 179-203.
7. Cf., for example, the reduction of ecological problems to the
relation of scarcity and allocation, in Horst Siebert, Okonomiscke t
Theorie der Umwelt, Tubingen 1978.
Chapter 3 Complexity and Evolution
1. Quoted from Stafford Beer, Designing Freedom, New York 1'974,

pp. 7, 10, 95.
2. More exactly, this means that there is no 'requisite variety' for
this difference; or in other words, no system can acquire enough
complexity of its own to be able to control the complexity in its
environment. Of course, this does not exclude the possibility of
planning models, machines or systems that have requisite variety
for the states of affairs to be controlled.
3. Cf., Niklas Luhmann, Soziale Systeme, Frankfurt 1984, pp. 47ff.
and 249ff.
4. This is a theoretical approach that has been used frequently in
recent times. Cf., for example, Gerhard E. Lenski, 'Social Structure
in Evolutionary Perspective', in Peter M. Blau (ed.), Approaches
to the Study of Social Structure, London 1976, pp. 1 3 5 - 5 3 ;
Philippe Van Parijs, Evolutionary Explanation in the Social
Sciences: An Emerging Paradigm, London 1981; Bernhard Giesen/
Christoph Lau, 'Zur Anwendung Dawinistischer Erklarungsstrate-
gien in der Soziologie', in Kolner Zeitschrift fur Soziologie und
Sozialpsychologie, vol. 33 (1981), pp. 2 2 9 - 5 6 ; Michael Schmid,
Theorie sozialen Wandels, Opladen 1982.
5. Among others cf., Andre Bejin, 'Differenciation, complexification,
Notes 153
evolution des societes', in Communications, vol. 22 (1974), pp.
105-18.
6. This is already incontestable for the concept of complexity. Cf.,
for example, Todd R. La Porte, 'Organized Social Complexity:
Explication of a Concept', in La Porte (ed.), Organized Social
Complexity: Challenge to Politics and Policy, Princeton, N.J. 1975,
pp. 3 - 3 9 .
7. An increasing criticism of this idea can be observed over the last
few years through the development of theories of the self-
organization of thermodynamically open systems and the self-
referential formation of systems. Cf., for example, Edgar Morin,
La Methode, vol. 2, Paris 1980, pp. 47ff.; Alfred Gierer,
'Socioeconomic Inequalities: Effects of Self-Enhancement,
Depletion and Redistribution', in Jahrbuch fur Nationaldkonomie
und Statistik, vol. 196 (1981), pp. 3 0 9 - 3 1 ; Gerhard Roth,
'Conditions of Evolution and Adaption in Organisms as Autopoietic
Systems', in D. Mossakowski/ G. Roth (eds), Environmental
Adaption and Evolution, Stuttgart 1982, pp. 3748.
8. Today, a sceptical, or at least a very circumspect opinion
predominates in this respect. For social systems cf., for example,
Mark Granovetter, 'The Idea of "Advancement" in Theories of
Social Evolution and Development', in American Journal of
Sociology, vol. 85 (1979), pp. 4 8 9 - 5 1 5 ; Walter L. Buhl, 'Gibt es
eine soziale Evolution?' in Zeitschrift fur Politik, vol. 31 (1984),
pp. 3 0 2 - 3 2 .
9. A standard interpretation that advises conserving the developmental
potential of whatever is unspecific. Cf., for example, E. D. Cope,
The Primary Factors of Organic Evolution, Chicago 1896, pp.
172ff.; Elman R. Service, The Law of Evolution and Culture, Ann
Arbor, Mich. 1960, pp. 93ff.; also in Elman R. Service, Cultural
Evolutionism: Theory in Practice, New York 1971, pp. 3Iff.
Chapter 4 Resonance
1. Cf., Humberto Maturana, Erkennen: Die Organisation und

Verkorperung von Wirklichkeit, Braunschweig 1982, for example,
pp. 20ff., 150ff., 287ff.; Francisco Varela, 'L'auto-organisation:
de l'apparence au mecanisme', in Paul Dumouchel/Jean-Pierre
Dupuy (eds), L'auto-organisation: de la physique au politique,
Pans 1983, pp. 147-64 (148).
2. For a more detailed analysis cf., the chapter on meaning in Niklas
154 Notes
Luhmann, Soziale Systeme, pp. 92ff.
3. indeterminacy means the necessary determinacy of a strictly
prescribed style,' as is said in Ideen zu einer reinen Phdnomenologie
und phanomenologischen Philosophie, vol. 1, Husserliana vol. Ill,
The Hague 1950, p. 100.
4. Cf., Francisco Varela, Principles of Biological Autonomy, New
York 1979; Maturana, ibid. (1982).
5. The theoretical sources of this idea are quite heterogeneous and
difficult to survey. The neo-dialectical tradition, especially Hegel,
comes readily to mind. Cf., also Ferdinand de Saussure, Cours de
linguistique generale, 5th edn, Paris 1962; Alfred Korzybski,
Science and Sanity: An Introduction to Non-Aristotelian Systems
and General Semantics, 4th edn, Lakeville Conn. 1958; George
A. Kelly, The Psychology of Personal Constructs, 2 vols, New
York 1955.
6. Cf., above chapter 1, note 19.
7. Cf., for this Helmut Willke, 'Zum Problem der Intervention in
selbstreferentielle Systemen', in Zeitschrift fur systemische Therapie,
vol. 2 (1984), pp. 191-200.
8. Cf., for example, Korzybski, ibid., pp. 386ff.
9. For the semantics of totalitarianism that tries to negate this cf.,
Marcel Gauchet, 'L'Experience totalitaire et la pensee de la
politique', in Esprit, July/August 1976, pp. 3 - 2 8 .
Chapter 5 The Observation of Observation
1. Cf., Roy A. Rappaport, Ecology, Meaning and Religion, Richmond

Ca. 1979, pp. 97ff.
2. Cf., Maturana, ibid. (1982), especially pp. 36ff. Maturana calls
the other-reference of first-order observation 'niches' and reserves
'environment' for that which reveals itself to second-order obser-
vation as the other-reference of the observed system. We have not
followed this terminology in the text because, although it is
univocal and clear, it would force us to use a terminology that
constantly deviates from the standard usage.
3. The cybernetic theory of Heinz von Foerster is based on this
viewpoint. Cf., Observing Systems, Seaside Ca. 1981.
4. This is the well-known argument of Douglas R. Hofstadter, Godel,
Escher, Bach: An Eternal Golden Braid, Hassocks, Sussex 1979.
5. Thus in the form of a question for which there is still no answer,
Lars Lofgren, 'Some Foundational Views on General Systems and
Notes 155
the Hempel Paradox', in International Journal of General Systems,
vol. 4 (1978), pp. 2 4 3 - 5 3 (244).
6. Cf., in particular 'On Constructing a Reality', in Heinz von
Foerster, ibid., (1981), pp. 2 8 8 - 3 0 9 , and 'Objects: Tokens for
(Eigen-)Behaviors', in ibid., pp. 2 7 4 - 8 5 . Cf., also John Richards/
Ernst von Glasersfeld, 'Die Kontrolle von Wahrnehmung und die
Konstruktion von Realitat', in Delfin III (1984), pp. 3 - 2 5 .
7. Cf., again Heinz von Foerster, 'Cybernetics of Cybernetics', in
ibid.
8. See for this Edward E. Jones/Richard E. Nisbett, 'The Actor and
the Observer: Divergent Perceptions of the Causes of Behavior',
in Edward E. Jones et al., Attribution: Perceiving the Causes of
Behavior, Morristown N.J. (1971), pp. 79-94, and as a new survey
of this 'fundamental attribution error' - i.e., the neglect of
situational factors - Lee Ross/Craig A. Anderson, 'Shortcomings
in the Attribution Process', in David Kahnemann/Paul Slovic/Amos
Tversky (eds), Judgement under Uncertainty: Heuristics and Biases,
Cambridge, UK 1982, pp. 129-52 (135ff.), or in a more developed
way, Francesco Pardi, Uosservabilita dell'agire sociale, Milan
1985. Social psychology noticed immediately that it also character-
ized itself through this insight: as the observer who must also
realize that the observed actor follows different principles of
attribution than his observer. Cf., for this Wulf-Uwe Meyer/Heinz-
Dieter Schmalt, 'Die Attributionstheorie', in D. Frey (ed.), Kognitive
Theorien der Sozialpsychologie, Bern 1978, pp. 9 8 - 1 3 6 . See
further, the interesting essay of Walter Mischel, 'Toward a
Cognitive Social Learning Reconceptualization of Personality', in
Psychological Review, vol. 80 (1973), pp. 25283.
9. Philippe Van Parijs, Evolutionary Explanation in the Social
Sciences: An Emerging Paradigm, London 1981, pp. 129ff., calls
this the 'principle of suspicion' and notes that this kind of analysis
encounters an 'authoritative self-knowledge' that it cannot resolve
into suspicion. This, too, is a variant of the general difference of
first- and second-hand observation.
10. If one reads the recent work of Jurgen Habermas, especially Der
philosophische Diskurs der Moderne, Frankfurt 1985, guided by
these considerations then it appears as a critique of the critique
of the self-descriptions of modern society, i.e., as a kind of third-
order cybernetics in the environment specific for it: literature. It
is consistent then to carry out the discourse as the discussion of
opinions that authors have expressed about other authors (Hegel
about Kant, Heidegger about Nietzsche, etc.). The pristine
156 Notes
transparency of these presentations can be gained only on the
basis of an extreme reduction of one's own theory that dismisses
the aporias of a self-clarifying reason and merely requires that one
presents testable validity claims in communication. Through this
simplification the description of the description of descriptions
acquires a considerable succinctness but at the same time an
unbridgeable distance from real social operations that are then
indirectly transfigured as life-world.
Chapter 6 Communication as a Social Operation
1. Thus, for example, with unproductive explanatory remarks, Eric

Trist, 'Environment and Systems-Response Capability', in Futures,
vol. 12 (1980), pp. 113-27.
2. Ibid., 1983.
3. Cf., Niklas Luhmann, 'Autopoiesis des Bewusstseins', in Soziale
Welt, forthcoming.
4. This has consequences for the 'de-subjectivization' of the concept
of communication that I have worked out elsewhere. See Soziale
Systeme, ibid., pp. 19Iff.
Chapter 7 Ecological Knowledge and Social Communication
1. Cf., for an older survey of the literature June Helm 'Ecological

Approach in Anthropology', in American Journal of Sociology,
vol. 67 (1962), pp. 6 3 0 - 9 . Cf., also Julian H. Steward, Evolution
and Ecology, Essays on Social Transformation, Urbana 111. 1977;
Roy A. Rappaport, Ecology, Meaning and Religion, Richmond
Ca. 1979. The question of ecological self-regulation that is
important for our comparison is distinguished from the predomi-
nant problematic that asks whether and to what extent ecological
conditions can explain differential evolution, i.e., evolutionary
progress as well as the retardation of social development. Today,
this kind of theory finds itself exposed to many critical objections;
cf., for example, Elman R. Service, Primitive Social Organization:
An Evolutionary Perspective, New York 1962, pp. 65ff., 72ff.;
Robert L. Winzeler, 'Ecology, Culture, Social Organization, and
State Formation in Southeast Asia', in Cultural Anthropology, vol.
17 (1976), pp. 623-32. Kent V. Flannery, 'The Cultural Evolution
of Civilizations', in Annual Review of Ecology and Systematics,
vol. 3 (1972), pp. 3 9 9 - 4 2 6 , pleads for explanatory models that
Notes 157
are more complex (including the processing of cultural information).
This line of discussion also confirms that one has to understand
society as an operatively closed, but self-reactive and thereby
environmentally open system.
2. Cf., Roy A. Rappaport, Pigs for the Ancestors, New Haven 1968.
3. Cf., again for New Guinea, Frederik Barth, Ritual and Knowledge
among the Baktaman of New Guinea, Oslo 1975.
4. Cf., Walter J. Ong, The presence of the Word: Some Prolegomena
for Cultural and Religous History, New Haven 1967; Ong,
Rhetoric, Romance and Technology: Studies in the Interaction of
Expression and Culture, Ithaca N.Y. 1971; Ong, Interfaces of the
World: Studies in the Evolution of Consciousness and Culture,
Ithaca N.Y. 1977.
5. De libero arbitrio, la 7ff., especially 10; quoted according to
Ausgewahlte Schriften, (ed. Werner Welzig), vol 4, Darmstadt
1969, pp. l l f f .
6. Cf. A. J. Festugiere, La revelation d'Hermes Trismegiste, 4 vols,
Paris 1950-4; Frances Yates, Giordano Bruno and the Hermetic
Tradition, Chicago 1964.
7. Cf. Micahel Giesecke, 'Uberlegungen zur sozialen Funktion and
zur Struktur handschriftlicher Rezepte im Mittelalter', in Zeitschrift
fur Literaturwissenschaft und Linguistik, vols 51/52 (1983), pp.
167-84.
8. In reference to tribal cultures Rappaport, ibid., (1979), pp. lOOff.,
says 'Because knowledge can never replace respect as a guiding
principle in our ecosystemic relations, it is adaptive for cognized
models to engender respect for that which is unknown, unpredict-
able, and uncontrollable, as well as for them to codify empirical
knowledge.'
9. Thus Thomas Wright, The Passions of the Minde in Generall, rev.
ed, London 1630, reprinted Urbana 111. 1971, p. 141.
10. Cf., also Niklas Luhmann, Die Funktion der Religion, Frankfurt
1977, especially pp. 255ff.; Niklas Luhmann/Karl Eberhard Schorr,
Reflexionsprobleme im Erziehungssystem, Stuttgart 1979, pp.
24ff.; Luhmann, Gesellschaftstruktur und Semantik, vol. 1, Frank-
furt 1980, pp. 9ff.; Luhmann, Politische Theorie in der Wohlfahrt-
staat, Munich 1981, pp. 19ff.; Luhmann, 'Gesellschaftsstrukturelle
Bedingungen und Folgeprobleme des naturwissenschaftlich-tech-
nischen Fortschritts', in Reinhard Low et al. (ed.), Fortschritt ohne
Mass}, Munich 1981, pp. 113-31; Luhmann, The Differentiation
of Society, New York 1982, pp. 229ff.; Luhmann, 'Anspruchsinfl-
ation im Krankheitssystem: Eine Stellungnahme aus gesellschafts-
158 Notes
theoretischer Sicht', in Philipp Herder-Dorneich/Alexander Schuller

(eds), Die Anspruchsspirale, Stuttgart 1983, pp. 168-75; Luhmann,
'Die Wirtschaft der Gesellschaft als autopoietisches System', in
Zeitschrift fur Soziologie, vol. 13 (1984), pp. 3 0 8 - 2 7 .
Chapter 8 Binary Coding
1.Cf., for example, Achille Adigo. Crisi di governabilita e mondi

vitale, Bologna 1980; Jurgen Habermas, Theorie des kommunika-
tiven Handelns, Frankfurt 1981, vol. 2, pp. 171 ff.
2. Cf., for this chapter 8 note 5 below.
3. See especially for the code of truth Niklas Luhmann, 'Die
Ausdifferenzierung von Erkenntnisgewinn: Zur Genese von Wissen-
schaft', in Nico Stehr/Volker Meja (eds), Wissenschaftssoziologie,
special volume 22/1980 of the Kolner Zeitschrift fur Soziologie
und Sozialpsychologie, Opladen 1981, pp. 102-39.
4. Antecedents to this are the ancient custom of designating totalities
with dual expressions - like 'heaven and hell' or 'court and country'.
Cf., for example, Ernst Kemmer, Die polare Ausdrucksweise in
der griechischen Literatur, Wiirzburg 1903; Adhemar Massart,
'L'Emploi, en egyptien, de deux termes opposes pour exprimer la
totalite', in Melanges bibliques, Paris 1957, pp. 3 8 - 4 6 ; G. E. R.
Lloyd, Polarity and Analogy: Two Types of Argumentation in
Early Greek Thought, Cambridge, UK 1966; Louis Dumont,
Homo Hierarchicus: The Caste System and its Implications,
London 1970, especially pp. 42ff.
5. Michel Serres, Le Parasite, Paris 1980.
6. Even for dualisms in earlier social formations it was true that they
were presented, typically, to indicate more than only one and used
or not depending on the situation. Only in this way was the
exclusion of third possibilities and the proscription of mixed forms
possible. Cf., among the extensive literature, for example, Edmund
Leach, 'Anthropological Aspects of Language: Animal Categories
and Verbal Abuse', in Eric E. Lenneberg (ed.), New Directions in
The Study of Language, Cambridge, Mass. 1964, pp. 2 3 - 6 3 ; Mary
Douglas, Purity and Danger: An Analysis of the Concepts of
Pollution and Taboo, London 1966, especially pp. 162ff.; Victor
Turner, The Ritual Process: Structure and Anti-Structure, London
1969, pp. 38ff.; Rodney Needham (ed.), Right and Left: Essays
on Dual Symbolic Classification, Chicago 1973.
7. Cf., for an analysis of such matters within the legal system Niklas
Notes 159
Luhmann, 'Die Theorie der Ordnung und die natiirlichen Rechte',
in Rechtshistoriscbes Journal, vol. 3 (1984), pp. 133-49.
8. Plato, Lysis, 215 E.
9. Even 300 years ago this meant that, 'To the Royal Society it will
be at any time almost as acceptable to be confuted, as to discover,'
according to Thomas Sprat, The History of the Royal Society of
London, For the Improving of Natural Knowledge, London 1667,
reprint St Louis-London 1959, p. 100.
10. Especially for the literature on law and jurisprudence. Cf., for
example, Josef Esser, Vorverstandnis und Methodenwahl in der
Rechtsfindung: Rationalitatsgarantien der richterlichen Entschei-
dungspraxis, Frankfurt 1970; Philippe Nonet/Philip Selznick, Law
and Society in Transition, London 1979; Gunther Teubner,
'Reflexives Recht: Entwicklungsmodelle des Rechts in vergleich-
ender Perspektive', in Archiv fur Rechts- und Sozialphilosophie,
vol. 68 (1982J, PP- 13-59.
11. According to Gregory Bateson, Okologie des Geistes: Anthropolo-
gische, biologische und epistemologische Perspektiven, Frankfurt
1981, especially pp. 515ff.
12. For the belief that alphabetized writing could have been the
stimulus for this cf., Jack Goody/Ian Watt, 'The Consequences of
Literacy', in Comparative Studies in Society and History, vol. 5
(1963), pp. 3 0 5 - 4 5 . Another explanation could refer to the high
degree of the Greek state's structural differentiation and to the
already widespread 'privatization' of religious participation. Cf.,
for this S. C. Humphreys, 'Approaches to the Study of Structural
Differentiation', in J. Friedman/M. J. Rowlands (eds), The Evol-
ution of Social Systems, Pittsburgh 1978, pp. 3 4 1 - 7 1 .
13. Distinctions like center/periphery and great tradition/little tradition
refer to this. Cf., Edward Shils, 'Centre and Periphery', in The
Logic of Personal Knowledge: Essays presented to Michael Polanyi,
London 1961, pp. 117-31; Robert Redfield, Peasant Society and
Culture: An Anthropological Approach to Civilization, Chicago
1956.
Chapter 9 Codes, Criteria, Programs
1. See, for example, Karl R. Popper, Objective Knowledge: An

Evolutionary Approach, Oxford 1972, pp. 13, 317ff.
2. See in regard to truth Sextus Empiricus, Adversos Mathematicos,
160 Notes
II 80, quoted according to Opera vol. Ill, Leipzig (Teubner), no
date, p. 100.
3. See also the corresponding distinction of values and programs in
Niklas Luhmann, Recbtssoziologie, 2nd edn, Opladen 1983, pp.
80ff.; Luhmann, Soziale Systeme, p. 434.
4. In the sense of Charles O. Frake, 'The Ethnographic Study of
Cognitive Systems', in Anthropology and Human Behavior,
Washington D.C. 1962, pp. 7 2 - 8 5 (78ff.). Cf., also Frake 'The
Diagnosis of Disease Among the Subanun of Mindanao', in
American Anthropologist, vol. 63 (1961), pp. 113-32.
5. See as one example the distinction of first eternal law and second
eternal law in Richard Hooker, On the Laws of Ecclesiastical
Policy, Book 1, III, 1, cited according to the edition of Everyman's
Library, Letchworth, Herts 1954, pp. 154ff.
6. This is very clear in Joyce O. Appleby, Economic Thought and
Ideology in Seventeenth-Century England, Princeton 1978.
7. According to Joseph Glanville, The Vanity of Dogmatizing, London
1661, reprint Hove, Sussex 1970, p. 180, 'Nature works by an
invisible hand in all things.' The origin of the metaphor of the
'invisible hand' is, as far as I know, still unclear. One could
suppose that the polemics against belief in miracles and divine
providence, against the 'pointing finger' of God, provided, through
the occurrence of unusual events, i.e., through arguments like the
ones promoted within the Royal Society, the occasion for the
transformation of the metaphor of the pointing finger into that of
the invisible hand. Cf., also Thomas Sprat, The History of the
Royal Society, London 1667, pp. 82ff.
8. My proposal for this is to aim in a purely formal manner at
consistency in making decisions in the legal system. Cf., Niklas
Luhmann, 'Gerechtigkeit in den Rechtssystemen der modernen
Gesellschaft', in Ausdifferenzierung des Rechts: Beitrage zur
Rechtssoziologie und Rechtstheorie, Frankfurt 1981, pp. 3 7 4 - 4 1 8 .
9. We will come back to this in chapter 16.
10. Actual discussions about dedifferentiation [Entdifferenzierung] and
interpenetration wrestle with problems that start from the necessity
of describing the process paradoxically, namely, presupposing
what it supposedly eliminates. Cf., Eugen Buss/Martina Schops,
'Die gesellschaftliche Entdifferenzierung', in Zeitschrift fur Soziol-
ogie, vol. 8 (1979), pp. 3 1 5 - 2 9 ; Harald Mehlich, Politischer Protest
und Stabilitat: Entdifferenzierungstendenzen in der modernen
Gesellschaft, Frankfurt 1983, especially pp. 122ff.; Richard Munch,
Theorie des Handelns: Zur Rekonstruktion der Beitrage von
Notes 161
Talcott Parsons, Emile Durkheim und Max Weber, Frankfurt
1982; Munch, Die Struktur der Moderne: Grundmuster und
differentielle Gestaltung des institutionellen Aufbaus der modernen
Gesellschaften, Frankfurt 1984 (both works contain many contri-
butions to 'interpenetration'). More circumspect and less decisive
is Peter Weingart, 'Verwissenschaftlichung der Gesellschaft -
Politisierung der Wissenschaft', in Zeitschrift fiir Soziologie, vol.
12 (1983), pp. 2 2 5 - 4 1 .
Chapter 10 Economy
1. c.f., for more detail Niklas Luhmann, 'Das sind Preise', in Soziale
Welt, vol. 34, (1983), pp. 153-70; Luhmann, 'Die Wirtschaft
der Gesellschaft als autopoietisches System', in Zeitschrift fiir
Soziologie, vol. 13 (1984), pp. 3 0 8 - 2 7 .
2. In many respects one will still, of course, be able to detect
'medieval' relations of an almost universal applicability of money
in developing countries. Cf., for this and for the (again medieval-
like) counter-movements, Georg Elwert, 'Die Verflechtung von
Produktionen: Nachgedanken zur Wirtschaftsanthropologie', in
Ernst Wilhelm Miiller et al. (eds), Ethnologie als Sozialwissenschaft,
special edition 26/1984 of the Kolner Zeitschrift fiir Soziologie
und Sozialpsychologie, Opladen 1984, pp. 3 7 9 - 4 0 2 (397ff.).
3. The modern discussion of 'property rights' has developed within
this context. Its extension to ecological goods, like 'rights' to
environmental pollution, has not been able to assume the former
preservative function of property because a right to pollution of
the air and water, whatever has been paid for it, does not enable
the proprietor to handle the air or water carefully and does not
give him the right to complain about the emissions of others.
4. Cf., for example, Raymond de Roover, 'The Concept of Just Price:
Theory and Economic Policy', in Journal of Economic History,
vol. 18 (1958), pp. 4 1 8 - 3 4 ; de Roover, La Pensee economique
des scolastiques: Doctrines et methodes, Paris 1971, see especially
pp. 59ff.
5. This demonstrates both a release of the economy for self-regulation
as well as the reinforcement of economic dependence on a
functioning legal system, i.e., the increase of the dependence and
independence of the economy on law. This was clearly seen by
Max Weber and has been developed extensively ever since. Cf.,
for example, James William Hurst, Law and the Conditions of
162 Notes
Freedom in the Nineteenth-Century United States, Madison Wis.

1956. See also Hurst, Law and Social Process in United States
History, Ann Arbor Mich. 1960; Hurst, Law and Economic
Growth: The Legal History of the Lumber Industry in Wisconsin
1836-1915, Cambridge Mass. 1964 (with indirect references to
ecological consequences); and Morton Horwitz, The Transform-
ation of American Law 1780-1860, Cambridge Mass. 1977.
Worthy of consideration is Warren J. Samuels, interrelations
between Legal and Economic Processes', in Journal of Law and
Economics, vol. 14 (1971), pp. 4 3 5 - 5 0 .
6. Thus Michel Aglietta/Andre Orlean, La Violence de la monnaie,
2nd edn, Paris 1984. Drawing on Rene Girard the concept is called
here 'imitative contagion' and designates the interconnectedness of
mimic behavior, i.e., manufactured scarcity, conflict and violence
as the conditions of order.
7. This is in blatant opposition to what jurists, planners and even
economists who require a 'basic order' normally think. The system
no longer reacts to structural goals with conformity/deviance but
only to structural changes that can be perceived and processed.
8. To be sure, very uncertain estimations! See, for example, Han-
delsblatt, of 28 February 1985; Borsenzeitung of 1 March 1985;
Herald Tribune of 4 March 1985; Frankfurter Allgemeine Zeitung
of 7 March 1985 - all of these relating to the problem of the
effect on the dollar of the intervention of the central banks.
9. One finds in the 'environmental economy', as with Hans Christian
Binswanger, 'Okonomie und Okologie neue Dimensionen der
Wirtschaftstheorie', in Schweizerische Zeitschrift fiir Volkswirtsch-
aft und Statistik, vol. 108 (1972), pp. 2 5 1 - 8 1 (276ff.), the
interpretation that the real reason for the expansion of the economy
and for the increasing burdening of the environment is found in
the creation of money. But just as with every ascription this is
problematic because it promotes the impression that one can
remedy the trouble at this point.
10. Therefore it is not without reason that one has spoken of the
'sovereignty' of money and of the ultimate arbitrariness of its
violence (although these concepts suggest an analogy with politics
that can be overextended and misunderstood). Cf., in connection
with Rene Girard, Michel Aglietta/Andre Orlean, La Violence de
la monnaie, Paris 1984, especially pp. 53ff.
11. Makrookonomik des Umweltschutzes, Gottingen 1976, p. 10.
12. One can also interpret this as the failure of a 'hierarchization'
with which the economic system normally weakens its fundamental
r
Notes 163
paradox. Or, formulated differently, the environmental economy
has to resort to other forms of paradox-elimination and asymmetriz-
ation.
13. Less optimistic are observers who simply begin from the facts.
Cf., for example, Brock B. Bernstein, 'Ecology and Economics:
Complex Systems in Changing Environments', in Annual Review
of Ecology and Systematics, vol. 12 (1981), pp. 3 0 9 - 3 0 and in
relation to the boundaries of the consequences of a 'moral suasion',
of a change of values, of a change of consciousness William J.
Baumol/Wallace E. Oates, Economics, Environmental Policy and
the Quality of Life, Englewood Cliffs, N.J. 1979, pp. 282ff.
14. Cf., for example, Karl-Heinrich Hansmeyer, 'Okologische Anfor-
derungen an die staatliche Datensetzung fur die Umweltpolitik
und ihre Realisierung', in Lothar Wegehenkel (ed.), Marktwirtschaft
und Umwelt, Tubingen 1981, pp. 6 - 2 0 (9).
15. I have not been able to understand and to translate into a
sociological language what economists understand by the 'market'.
The crucial systems-theoretical insight is that the market is not
a 'subsystem' of the economic system but its system-internal
environment or section of this environment viewed from the
perspective of the individual subsystems. Cf., especially Harrison
C. White, 'Where Do Markets Come From?' in American Journal
of Sociology, vol. 87 (1981), pp. 5 1 7 - 4 7 . If one begins from this
then there is no difficulty in discovering such system-internal
environments even in socialist economies with a state-run pro-
duction apparatus. Whether one calls this a 'market' or not is
then primarily a matter of ideology.
16. Reflections on theoretical models can be found in Horst Siebert,
Okonomische Theorie der Umwelt, Tubingen 1978.
17. Cf., Douglas R. Hofstadter, Godel, Escher, Bach: An Eternal
Golden Braid, Hassocks, Sussex 1979.
18. For further problems resulting from the distinction of level-
decisions and allocation-decisions see also Joachim Klaus, 'Zur
Frage der staatlichen Fixierung von Umweltstandards und Emis-
sionsniveaus', in Wegehenkel, ibid., pp. 969.
19. Compared with the legal system, the parallel is clear with the
similarly hierarchical difference of levels between law-making and
law-application that has to be viewed as a strategy of paradox-
elimination but which always fails in practice. This, however,
occurs only in individual cases and in a way that is acceptable.
20. See perhaps Bender, ibid., (1976); Sieber, ibid., (1978); or in
reference to the level of management science Udo Ernst Simonis
164 Notes
(ed.), Okonomie und Okologie: Auswege aus einem Konflikt,

Karlsruhe 1980.
21. It is then an empirical question whether and how far these effects
can be compensated by new demands. In any event, the theory
does not require that increased expense for environmental consider-
ations has to be detrimental to the entire economy.
22. In this sense it is meaningless to speak of 'non-economic' costs.
This is only a metaphorical way of speaking that transfers the
specificity of the economic mode of thinking indiscriminately to
other social domains.
23. Thus we must reject the interpretation that can be found
occasionally in the economic literature (for example, Abraham
Moles/Elisabeth Rohmer, Theorie des actes: vers une ecologie des
actions, Paris 1977, p. 57), which says that through the calculations
of costs a social integration of action in terms of scarce resources
can be reached.
Chapter 11 Law
1. For the controversy cf., the detailed presentation of William J.

Baumol/Wallace E. Oates, Economics, Environmental Policy and
the Quality of Life, Englewood Cliffs, N.J. 1979, pp. 230ff.
2. For more detail cf., Niklas Luhmann, Rechtssoziologie, 2nd
edn, Opladen 1983, pp. 354ff.; Luhmann, 'Die Einheit des
Rechtssystems', in Rechtstheorie, vol. 14 (1983), pp. 129-54.
3. Wherever this condition is not met - in the slums of the larger
cities of Brazil it is said that the people live according to their
own laws, not according to the laws of the state - the certainty
that the law is not illegal does not exist.
4. For the sake of clarification it should be said that, in opposition
to a widespread belief (see Lawrence M. Friedman, The Legal
System: A Social Science Perspective, New York 1975), we do not
restrict the legal system to its organizational and professional
working complex (law-making, judicial procedure, attorneys) but
include any communication that guides itself by the difference of
legal and illegal in the juristic sense.
5. That the 'complete' description of the world remains logically
incomplete because third possibilities have to be excluded remains
a persistently recurring problem of the legal tradition. Cf., Niklas
Luhmann, 'Die Theorie der Ordnung und die naturlichen Rechts',
in Rechtshistorisches Journal, vol. 3 (1984), pp. 133-49.
Notes 165
6. Correspondingly, introductory and class texts have been in
production for more than ten years. See, for example, Michael
Klopfer, Zum Umweltschutzrecht in der Bundesrepublik, Perscha
n.d. (1972); Peter-Christoph Storm, Umweltrecbt: Einfiihrung in
ein neues Rechtsgebiet, Berlin 1980; Juergen Salzwedel (ed.),
Grundziige des Umweltrechts, Berlin 1982.
7. Michael Klopfer, Systematisierung des Umweltrechts, Berlin 1978,
serves as a good survey.
8. For this it is already of secondary importance whether particular
freedom-rights, like the right to use the air, fall under constitutional
freedom-guarantees or whether they are viewed as legal positions
that are initially guaranteed by the law-makers and afterwards are
modifiable if necessary.
9. The picture is no different if, instead of this, one has in mind the
more formal concepts of permission and prohibition or indicates
that the consideration of interests or the distinction of centralized
and decentralized standards of selection play an important role.
Even in these cases the distinction that structures the juristic
discourse politically or the interpretation of the law is not the one
of system and environment.
10. From Robert Weimar/Guido Leinig, Die Umweltvorsorge im
Rahmen der Landesplanung Nordrhein-Westfalen, Frankfurt 1983,
pp. 20 or 40.
11. Of course, there are many other reasons for the often lamented
'performance deficit'. Cf., Karl-Heinrich Hansmeyer (ed.), Vollzug-
sprobleme der Umweltpolitik: Empirische Untersuchungen der
Implementation von Gesetzen im Bereich der Luftreinhaltung und
des Gewasserschutzes, (project director Renate Mayntz) n.d. 1978.
A good case-study is Bruce A. Ackerman et al., The Uncertain
Search for Environmental Quality, New York 1974. In these
details one immediately sees how difficult it is to present
'thoroughgoing' proposals for improvement.
12. Cf., Robert D. Luce/Howard Raiffa, Games and Decisions, New
York 1957, especially pp. 278ff.
13. Cf., Aaron Wildavsky, 'No Risk is the Highest Risk of All', in
American Scientist, vol. 67 (1979), pp. 3 2 - 7 . Cf., also Peter
Gardenfors, 'Forecasts, Decisions and Uncertain Probabilities', in
Erkenntnis, vol. 14 (1979), pp. 159-81, with reference to the
significance of the distinctive quality of prognoses - a question
whose omission in decisions according to the rule of the
maximization of profit makes these decisions too risky.
14. Cf., Nathan Kogon/Michael A. Wallach, 'Risk-Taking as a Function
166 Notes
of the Situation, the Person, and the Group', in New Directions

in Psychology III, New York 1967, pp. 111-278. Coming from
entirely different points of departure, investigations in decision
theory provide the same impression. Cf., Harry J. Otway,
'Perception and Acceptance of Environmental Risks', in Zeitschrift
fiir Umweltpolitik, vol. 2 (1980), pp. 593-616 (with reference to
groups that are more strongly committed) or Baruch Fischhoff et
al., Acceptable Risk, Cambridge, UK 1981.
15. And this in the upper and lower strata, with or without calculation.
Cf., for its scope, Jaques de Cailliere, La Fortune des gens de
qualite et des gentilhommes particuliers, Paris 1664, pp. 307ff.;
Hunter S. Thompson, Hell's Angels, New York 1966.
16. For extensive research cf., David Kahneman/Paul Slovic/Amos
Tversky, Judgement under Uncertainty, Cambridge, UK 1982.
17. Cf., for this argument Chauncey Starr/Richard Rudman/Chris
Whipple, 'Philosophical Basis for Risk Analysis', in Review of
Energy, vol. 1 (1976), pp. 629-62. A survey of (very inadequate)
empirical methods for the ascertainment and collection of social
preferences for risk are found in William D. Rowe, An Anatomy
of Risk, New York 1977, pp. 259ff.
18. Thus Heinhard Steiger, 'Verfassungsrechtliche Grundlagen', in
Salzwedel, ibid., pp. 2 1 - 6 3 , for 'risk-remainder', pp. 37ff., quote
p. 41.
19. Thus Hasso Hofmann, Rechtsfragen atomarer Entsorgung,
Stuttgart 1981, for risk-remainder especially pp. 336ff. Or
pluralized according to organized interests as with Karl-Heinz
Ladeur,'Abwagung' Ein neues Paradigma des Verwaltungsrechts:
Von der Einheit der Rechtsordnung zum Rechtspluralismus,
Frankfurt 1984.
20. Cf., with examples from American legislation, Talbot Page, 'A
Generic View of Toxic Chemicals and Similar Risks', in Ecology
Law Quarterly, vol. 7 (1978), pp. 2 0 7 - 4 4 . Cf. also Lawrence H.
Tribe, 'Trial by Mathematics: Precision and Ritual in the Legal
Process', in Harvard Law Review, vol. 84 (1971), pp. 1329-93.
21. The history of the concept of risk is still unclear. The occasion
for the appearance of a special concept distinct from the general
concept of danger could also have been the need to view risks not
only negatively as dangers but to consider them as the object of
an intentional undertaking and the will to pay for their absorption.
22. The focus of the discussion lies in the question of the creation of
marketable emissions laws. For the already extensive consideration
of the advantages and disadvantages cf., the many contributions in
Notes 167
Lothar Wegehenkel (ed.), Marktwirtschaft und Umwelt, Tubingen
1981; further Werner Zohlhofer, 'Umweltschutz in der Demokra-
tie', in Jahrbucb fur Neue Politische Okonomie, vol. 3 (1984), pp.
101-21.
23. For the problem of the quantification of the readiness to take
fatality risks into consideration and for the difficulty in determining
ethical standards for this cf., Ronald A. Howard, 'On Making
Life and Death Decisions', in Richard C. Schwing/Walter A. Albers,
Jr, (eds), Societal Risk Assessment: How Safe is Safe Enough?,
New York 1980, pp. 89-106. Obviously the problem is significant
not only for technological or ecological risks. But it acquires an
actuality and recognition through them.
24. Of course, even this is not true unconditionally, but only, as
reactions to the Three Mile Island accident indicate, to very
different degrees. Cf., Ortwin Renn, Wahrnehmung und Akzeptanz
technischer Risiken, vol. Ill, Jiilich 1981, pp. 20ff. For the time
being there is no explanation for these distinctions.
25. Bryan Wynne, 'Redefining the Issues of Risk and Public Acceptance:
The Social Viability of Technology', in Futures, vol. 15 (1983),
pp. 13-32, makes this clear.
26. For the cybernetics of the oscillating between internal and external
perspectives in the system cf., Stein Braten, 'The Third Position:
Beyond Artificial and Autopoietic Reduction', in Kybernetes, vol.
13 (1984), pp. 1 5 7 - 6 3 .
17. That problems result from this transferring of questions that
cannot be decided in a purely juristic way into the political system
can be shown with a consideration of the Green parties. We will
come back to this in chapter 8.
28. Cf., the widely discussed remarks of John Rawls, A Theory of
Justice, Cambridge Mass. 1971 or Jurgen Habermas, Theorie des
kommunikativen Handelns, 2 vols., Frankfurt 1981.
29. See the above-mentioned study of Hunter Thompson, Hell's
Angels, New York 1966. See also Erving Goffman, 'Where the
Action Is', in Goffman, Interaction Ritual: Essays in Face-to-Face
Behavior, Chicago 1967, pp. 149-270; or for another extreme
case, i.e., the climbing of the Himalayas, Michael Thompson,
'Aesthetics of Risk: Culture or Context', in Richard C. Schwing/
Walter A. Albers, Jr, (eds), Societal Risk Assessment: How Safe
is Safe Enough}, New York 1980, pp. 2 7 3 - 8 5 .
30. Cf., the already out-of-date inquiries in Volkmar Gessner et al.,
Umweltschutz und Rechtssoziologie, Bielefeld 1978, pp. 167ff.
Since then judicial activity has certainly increased but there are no
168 Notes
new investigations. From the juristic point of view see Michael

Klopfer, 'Rechtsschutz im Umweltschutz', in Verwaltungsarchiv,
vol. 76 (1985), pp. 3 7 1 - 9 7 (Part 2 forthcoming).
31. See note 11 above.
32. Cf., Barry Boyer/Errol Meidinger, 'Privatizing Regulatory Enforce-
ment: A Preliminary Assessment of Citizen Suits Under Federal
Environmental Laws', MS Buffalo N.Y. 1985. (I owe the reference
to this work to Volkmar Gessner).
33. Cf., Gerd Winter, 'Bartering Rationality in Regulation', in Law
and Society Review, vol. 19 (1985), pp. 2 1 9 - 5 0 .
34. See dispute Klopfer ibid., (1985) p. 391 with references.
Chapter 12 Science
1. To simplify the presentation we will omit here the argument used

by Heidegger, for instance, that the original difference between
true and untrue had already been replaced by the difference
between correct and false (especially in reference to ideas) in
classical Greek philosophy and that the result was a loss of being
that has not been remedied up to the present. If this reconstruction
of philosophical semantics is correct, then a correlate of the
beginning and increasingly social differentiation of science can be
found here.
2. See for this transformation Rudolph Stichweh, Zur Entstehung
des modernen Systems wissenschaftlicher Disziplinen: Pbysik in
Deutschland 1740-1890, Frankfurt 1984.
3. See Edmund Husserl, Die Krisis der europaeiscben Wissenscbaften
und die transzendentale Phaenomenologie, Husserliana, vol. VI,
The Hague 1954.
4. Tenbruck speaks of trivialization somewhat in this sense. See
Friedrich H. Tenbruck, 'Wissenschaft als Trivialisierungsprozess',
in Nico Stehr/Volker Meja (eds), Wissenschaftssoziologie: Studien
und Materialen, special edition 18 of the Kolner Zeitschrift fiir
Soziologie und Sozialpsychologie, Opladen 1975, pp. 19-47.
5. Formulated in a more precise, systems-theoretical sense this means
that, on the one hand, such 'noise' is an indispensable stimulus of
research and continually confers on it the certainty of reality. On
the other hand, it must continually be transformed into information
and eliminated as disturbance. Accordingly, one can expect massive
irritation from the researcher's 'change of consciousness' to
'consciousness of the environment' but not automatically the
Notes 169
conferral of scientific significance. Once again, this is a phenomenon

of limited resonance.
6. Moreover, this insight was already possible in classical modern
thought. Edward Reynolds, A Treatise of the Passions and Faculties
of the Soule of Man, London 1640, reprint Gainesville, Fla. 1971,
p. 503 says, 'It is speedier to come to a Positive Conclusion by
Negative Knowledge, than a naked Ignorance.' Reynolds also
includes extensive remarks about the human causes of error.
7. See, in the context of a general theory of symbolically generalized
media of communication, Niklas Luhmann, 'Einfiihrende Bemer-
kungen zu einer Theorie symbolisch generalisierter Kommunika-
tionsmedien', in Luhmann, Soziologische Aufklarung, vol. 2,
Opladen 1975, pp. 170-92.
8. Cf., in comparison to a more modern, specifically scientific
rationality, Peter-Michael Spangenberg, Maria ist immer und
iiberall, Frankfurt 1987.
9. Cf., for this Niklas Luhmann, 'The Differentiation of Advances
in Knowledge', in Nico Stehr/Volker Meja (eds), Society and
Knowledge: Contemporary Perspectives on the Sociology of
Knowledge, New Brunswick 1984, pp. 103-48.
10. The rejection of curiositas was not meant to eliminate the striving
for knowledge as such, only knowledge that was pointless -
whether this was in reference to transcendent matters that were
accessible only through faith or things that were arcane by their
very nature and could only be destroyed by knowledge. Moreover,
as an interest for 'others', curiositas was reproached for directing
attention away from important se//-knowledge. Thus the dispute
was not 'innovation versus constancy'. This distinction was
introduced only after the raising of curiositas to the universal
striving for knowledge. Cf., for example, Thomas Wright, The
Passions of the Minde in General, rev. edn, London 1630, reprint
Urbana 111., 1971, pp. 312ff.; Reynolds, ibid., (1640), pp. 462ff.
And, of course, there is Hans Blumenberg, Der Prozess der
theoretischen Neugierde, Frankfurt 1973.
11. Historically, this classification of the concept of method begins
after the widespread effect of printing in the sixteenth century.
The most important transitional position is occupied by Petrus
Ramus (Pierre de la Rammee) who understood method as a binary
schema but still applied it directly to the decomposition of reality.
Cf., Walter J. Ong, Ramus: Method and the Decay of Dialogue:
From the Art of Discourse to the Art of Reason, Cambridge Mass.
1958, reprint New York 1979.
170 Notes
12. Cf., for example, Isaac Levi, Gambling with Truth: An Essay on
Induction and the Aims of Science, London 1967.
13. The theory of evolution, therefore, is at the very least, a significant
perspective of modern science because it comes to the rescue here
and explains how (not why) reality, without any consideration of
logic and mathematics, so simplified itself that it finally has become
what it is.
14. This has been formulated ever since Warren Weaver, 'Science and
Complexity', in American Scientist, vol. 36 (1948), pp. 5 3 6 - 4 4 .
Cf., also Todd R. LaPorte (ed.), Organized Social Complexity:
Challenge to Politics and Policy, Princeton N.J. 1975; Giovan
Francesco Lanzara/Francesco Pardi, L'interpretazione della com-
plessita: Methodo sistemico e scienze sociali, Naples 1980; Hans
W. Gottinger, Coping with Complexity, Dordrecht 1983.
15. Cf., for example, Henri Atlan, Entre le cristal et la fumee: Essai
sur /'organisation du vivant, Paris 1979, pp. 74ff. Cf., also Lars
Lofgren, 'Complexity Descriptions of Systems: A Foundational
Study', in International Journal of General Systems, vol. 3 (1977),
pp. 197-214; Robert Rosen, 'Complexity as a System Property',
in International Journal of General Systems, vol. 3 (1977), pp.
227-32.
16. Cf., Heinz von Foerster, Observing Systems, Seaside Ca. 1981,
especially pp. 288ff.
17. It is obvious that this is an offense against the rules of the theory
of science - even against the rules of the divine Popper. See, for
example, Hans Albert, 'Modell-Platonismus: der neoklassische Stil
des okonomischen Denkens in kritischer Beleuchtung', in Festschrift
Gerhard Weisser, Berlin 1963, pp. 4 5 - 7 6 . But this insight does
not lead one out of the problem but deeper into it when it raises
the question (and this is only a different version of the problem
of structured complexity) how one can protect the theory of
science against infection by paradoxes, how one can immunize
Popper himself.
18. A quite considerable investigation of the effects of distinctions of
size on social structures is summarized as follows, '(25) Other
things being equal, the above statements about the relationship
between scale and social organizations are true. (26) Other things
are never equal.' Gerald D. Berreman, 'Scale and Social Relations:
Thoughts and Three Examples', in Frederick Barth (ed.), Scale and
Social Organization, Oslo 1978, pp. 4 1 - 7 7 (77).
19. To be sure, not if one considers the relative isolation of life on
the earth. Measured by the analytic apparatus of scientific theories
Notes 171
this boundary is not an 'ecological' boundary any longer. Moreover,

the totality of events on the earth is much too complex for anyone
to be able to work with this system reference scientifically.
20. Thus, for example, Roy A. Rappaport, Ecology, Meaning, and
Religion, Richmond Ca. 1979, pp. 54ff.
21. Cf., Herbert Simon/Albert Ando, 'Aggregation of Variables in
Dynamic Systems', in Econometrica, vol. 29 (1961), pp. 1 1 1 - 3 8 ;
Herbert A. Simon, 'The Architecture of Complexity', in Proceedings
of the American Philosophical Society, vol. 106 (1962), pp. 46782
(reprinted elsewhere); Franklin M. Fisher/Albert Ando, 'Two
Theorems on Ceteris Paribus in the Analysis of Dynamic Systems',
in American Political Science Review, vol. 56 (1962), pp. 1 0 8 - 3 3 ;
Albert Ando/Franklin M. Fisher, 'Near Decomposability, Partition
and Aggregation and the Relevance of Stability Discussions', in
International Economic Review, vol. 4 (1963), pp. 5367; Albert
Ando/Franklin M. Fisher/Herbert A. Simon, Essays on the Structure
of Social Science Models, Cambridge Mass. 1963. Cf., further C.
West Churchman, The Design of Inquiring Systems: Basic Concepts
of System and Organization, New York 1971, especially pp. 64ff.;
Daniel Metlay, 'On Studying the Future Behavior of Complex
Systems', in LaPorte, ibid., (1975), pp. 2 2 0 - 5 0 ; William C.
Wimsatt, 'Complexity and Organization', in Marjorie Green/
Everett Mendelsohn (eds), Topics in the Philosophy of Biology,
Boston Studies in the Philosophy of Science, 27 (1976), pp.
17493. As the contributions quoted last indicate, the sceptical
estimation is increasing.
22. And if unattainability is a foregone conclusion then it is pointless
to look for rationality in the establishment and the adherence to
procedural rules for reaching rationality - the 'bourgeois' way of
proceduralizing difficult questions.
Chapter 13 Politics
1. 'Politique et societe', in Communications, vol. 22 (1974), pp.

119-33 (125).
2. Cf., for the widespread, earlier terminology that prevailed until
the beginning of the eighteenth century, Daniel de Priezac, Discours
politiques, 2nd edn, Paris 1666; Remond de Cours, La veritable
Politique des personnes de qualite, Paris 1692; Christian Thoma-
sius, Kurtzer Entwurff der politischen Klugheit, German trans.
Frankfurt-Leipzig 1710, reprint Frankfurt 1971; Jiirgen Habermas,
172 Notes
Kleine Politische Scbriften, Frankfurt 1981. In this context politics

means not much more than public behavior.
3. For example, Ciro Spontone, Dodici libri del governo di Stato,
Verona 1599.
4. We are talking here in the context of the systems theory of social
systems, i.e., about communication. That the situation is different
for psychical systems is obvious.
5. Cf., Niklas Luhmann, 'Der politische Code: "Konservativ' und
"progressiv" in systemtheoretischer Sicht', in Luhmann, Soziologis-
che Aufklarung, vol. 3, Opladen 1981, pp. 2 6 7 - 8 6 ; Luhmann,
Politische Theorie im Wohlfahrtsstaat, Munich 1981, pp. 118ff.
6. This argument can be supported by numerous other comparisons.
To simplify the presentation we have restricted ourselves to the
differentiation of coding and programming that is relevant for the
particular theme of resonance capacity.
7. Cf., Manfred Schmitz, Theorie und Praxis des politischen Skandals,
Frankfurt 1981; Francesco M. Battisti, Sociologia dello scandalo,
Bari 1982. An empirical investigation of historical scandals would
supposedly show quite readily that ecological interests also increase
in this form, i.e., have become capable of being scandalous - either
because the total number of scandals increases or because the
distribution within this total shifts from morality to ecology. The
moralization of ecological themes may then no longer have the
function of making them capable of becoming scandalous.
8. For more on this cf. Niklas Luhmann, Macht, Stuttgart 1975, pp.
60ff.
9. We will come back to this later, pp. 283ff.
10. This case is exactly parallel, in a systems-theoretical sense, to the
question that we have been pursuing with respect to society. In
both cases we are concerned with whether and how a system can
find its rationality through calculating the effects of its own
operations on its environment in reference to the reactions on
itself.
11. Cf., for this also Walter Buhl, 'Okologische Knappheit', in ibid.,
pp. 141ff.
12. See especially Peter Graf Kielmansegg, 'Politik in der Sackgasse?
Umweltschutz in der Wettbewerbsdemokratie', in Heiner Geissler
(ed.), Optionen auf eine lebenswerte Zukunft: Analysen und
Beitrage zur Umwelt und Wachstum, Munich 1979, pp. 3 7 - 5 6 .
Notes 173
Chapter 14 Religion
1. In Horst Westmiiller, 'Die Umweltkrise - eine Anfrage an Theologie

und Christen', in Hans Dietrich Engelhardt (ed.), Umweltstrategie:
Materialien und Analysen zu einer Umweltethik der Industriege-
sellschaft, Giitersloh 1975, pp. 3 1 4 - 4 8 , 331). The quotation is
chosen arbitrarily and ought not to slander its author, but it is
representative for what I have found everywhere in the literature
about the religious position towards the environmental crisis. Cf.,
also Martin Rock, 'Theologie der Natur und ihre anthropologischen
Konsequenzen', in Dieter Birnbacher (ed.), Okologie und Ethik
Stuttgart 1980, pp. 7 2 - 1 0 2 ; and Gerhard Liedke, Im Baucb des
Fiscbes: Okologiscbe Theologie, Stuttgart 1979, with a much more
direct biblical orientation.
2. Formulated in terms of complexity one could also say its ultimate
difference resides in the indeterminability (transcendence) of the
determinate (immanence). This is the formulation that I use in Die
Funktion der Religion, Frankfurt 1977.
3. See also the factor of 'surprise' in Matthew 25, 3Iff. as an
expression of the difference of immanent and transcendent
valuation.
4. Or as Shaftesbury had already dared, ibid., vol. Ill, p. 316 and
repeatedly dared to say that faith is irreproachable, 'as by Law
Established'.
5. Johann Heinrich Lambert, Cosmologische Briefe iiber die Einrich-
tung des Weltbaues, Augsburg 1761, p. 116.
6. Anthropological formulas for this - like the self-created inscruta-
bility of God (fear of humanity), the cunning of God, the screening
of humanity from unbearable knowledge - are certainly old. Cf.,
some material in Stephen D. Benin, 'The "Cunning of God" and
Divine Accommodation', in Journal of the History of Ideas, vol.
45 (1984), pp. 179-91.
7. With this in mind I decided to use the Montaigne quote as a
motto. It ought to be considered in that context.
8. Michel Serres, he Parasite, Paris 1980.
Chapter 15 Education
1. Cf., Norman J. Faramelli, 'Ecological Responsibility and Economic

Justice', in Ian G. Barbour (ed.), Western Man and Environmental
174 Notes
Ethics: Attitudes Toward Nature and Technology, Reading Mass.

1973, pp. 188-203.
2. For this concept of career cf., Niklas Luhmann/Karl Eberhard
Schorr, Reflexionsprobleme im Erziehungssystem, Stuttgart 1979,
pp. 277ff.
3. This estimation is also found in Heinz von Foerster, Observing
Systems, Seaside Ca. 1981, pp. 209ff. - concluding with the
proposal, 'Would it not be fascinating to think of an education
system that detrivializes its students by teaching them to ask
"legitimate questions," that is, questions for which the answers
are unknown?'
4. Just to mention others: the problem of observing and understanding
in a system with structured complexity; the problem of actor/
observer differences of attribution; the problem of the 'hidden
agenda' and the socialization for survival in school.
5. So suppose William J. Baumol/Wallace E. Oates, Economics,
Environmental Policy, and the Quality of life, Englewood Cliffs
N.J. 1979, pp. 282ff. (although their own research results,
concerning recycling, seem to contradict this).
Chapter 16 Functional Differentiation
1. Cf., among others Jeffrey L. Pressman/Aaron Wildavsky, Implemen-

tation: How Great Expectations in Washington are Dashed in
Oakland, Berkeley Ca. 1973.
3. According to E. T. A. Hoffmann, 'Des Vetters Eckfenster', Werke,
Berlin-Leipzig, no date, vol. 12, pp. 14264.
4. Cf., for a historico-semantic context Niklas Luhmann, 'Temporalisi-
erung von Komplexitaet: Zur Semantik neuzeitlicher Zeitbegriffe',
in Luhmann, Gesellschaftsstruktur und Semantik, vol. 1, Frankfurt
1980, pp. 235-300.
5. Cf., Andre Bejin, 'Differenciation, complexification, evolution des
societes', in Communications, vol. 22 (1974), pp. 109-18 (114)
in connection with Henri Atlan, L'Organisation biologique et la
theorie de I'information, Paris 1972, pp. 270ff.
6. The still unclear semantic career of the concept of value (especially
prior to the middle of the nineteenth century) might have one of
its sources here. To be sure, it is incorrect to say that the concept
of value was appropriated by morality, literature, aesthetics and
philosophy from economics only in the middle of the nineteenth
Notes 175
century. (The Abbe Morellet, Prospectus d'un nouveau dictionnaire
de commerce, Paris 1769, reprint Munich 1980, pp. 98ff., observes
a restriction to economic profit. But the entire eighteenth century
used it in a much more general sense). It is equally clear, however,
that the concept of value has been used as an ultimate guarantee
for meaning and therefore non-contradictably in the last hundred
years.
7. This happens in any event. But it is also required in many respects
and viewed as the precondition for the solutions of problems.
Cf., Karl-Heinz Hillmann, Umweltkrise und Wertwandel: Die
Umwertung der Werte als Strategie des Uberlebens, Frankfurt-
Bern 1981.
8. Cf., Talcott Parsons, 'On the Concept of Value-Commitments', in
Sociological Inquiry, vol. 38 (1968), pp. 135-60 (153ff.).
9. For a comparison: the self-description of stratified societies had
always used a moral schematism whether in the direct moral
criticism of typical behavior in the individual strata or in the
formulation of types of perfection from which everyone could
measure their distance.
10. Cf., for example, Simon-Nicolas-Henri Linquet, Le Fanatisme des
pbilosopbes, London-Abbeville 1764; Peter Villaume, Uber das
Verhaltnis der Religion zur Moral und zum Staate, Libau 1791,
and of course, the widespread critique of the French Revolution
as the outbreak of a naive faith in principles.
11. This led many to the conclusion of 'revolution' - with very little
support for possibilities and consequences. One finds typically that
the manifest/latent schema is introduced without further reflection
as a description of facts and forms the basis for analyses. This
has been the case especially since Robert K. Merton, 'The
Unanticipated Consequences of Purposive Social Action', in
American Sociological Review, vol. 1, (1936), pp. 894-904.
12. Jiirgen Habermas judges much more sharply and leaves more
room for hope. He views this as the theory-immanent problem of
the Enlightenment's erroneous semantic guidance by the theory of
the subject and its object and therefore sees the solution of the
problem in the transition to a new paradigm of intersubjective
agreement. Cf., Der philosopbische Diskurs der Moderne: Zwolf
Vorlesungen, Frankfurt 1985. To make this useful sociologically,
one must still clarify how this erroneous guidance and the
possibility of correcting it are connected with the structure of
modern society.
176 Notes
Chapter 17 Restriction and Amplification
1. Thus according to Simon-Nicolas-Henri Linguet, 'Lettre sur La

Theorie des loix civiles\ Amsterdam 1770, p. 96 (in connection
with a critique of Montesquieu). To earlier authors who began
from the aristotelian theory of motion equilibrium was seen as
corruption, i.e., as indecisiveness in the direction of motion - 'as
it were in aequilibrio, that it cannot tell which way to encline', as
Reynolds says, ibid., (1640), p. 463.
Chapter 18 Representation and Self-observation
1. We will leave aside the special case of Egypt in which neither the
one nor the other is true but, instead, where religion assumed the
representation of unity. It found no successors.
2. Marcel Gauchet, 'L'Experience totalitaire et la pensee de la
politique', in Esprit, July/August 1976, pp. 3 - 2 8 (26).
3. For potentializing through inhibition cf., Yves Barel, Le Paradoxe
et le systeme: Essai sur la fantastique social, Grenoble 1979, pp.
185ff.
4. The recursivity in the formulation is intended.
5. The impetus that the theory of attribution experienced within this
context was itself an interesting theme of research. It is still
visible, for example, in Felix Kaufmann, Methodenlehre der
Sozialwissenschaften, Wien 1936, especially pp. 18Iff.
6. For a topical survey cf., Ortwin Renn, 'Die alternative Bewegung:
Eine historisch-soziologische Analyse des Protestes gegen die
Industriegesellschaft', in Zeitschrift fur Politik, vol. 32 (1985), pp.
153-94; Karl-Werner Brand (ed.), Neue soziale Bewegungen in
Westeuropa und den USA: Ein internationaler Vergleich, Frankfurt
1985.
7. In the sense of Michel Serres, Le Parasite, Paris 1980.
Chapter 19 Anxiety, Morality and Theory
1. Since we are concerned here only with communication we will

omit the components of emotional agitation in what follows and
deal only with (the expression of) worry. For the distinction of
both these components of the (psychological) concept of anxiety
cf., Ralf Schwarzer, Stress, Angst und Hilflosigkeit: Die Bedeutung
Notes 177
von Kognitionen und Emotionen bei der Regulation von Belastungs-
situationen, Stuttgart 1981, pp. 87ff.; Schwarzer, 'Worry and
Emotionality as Separate Components of Test Anxiety', in Inter-
national Review of Applied Psychology, vol. 33 (1984), pp.
2 0 5 - 2 0 . The distinction has been worked out in the so-called Test
Anxiety Research. Further contributions to this can be found in
the annals Advances in Test Anxiety Research (from 1982).
2. Cf., Anthony, Earl of Shaftesbury, Characteristicks of Men,
Manners, Opinions, Times, 2nd edn, ibid.; no date 1714, reprint
Farnborough, Hants UK 1968, vol. 1, p. 16. The collapse of a
theoretical construction of a Hobbesian type can easily be seen in
this.
3. See, for someone with this intention, for example, William W.
Lowrance, Science and the Determination of Safety, Los Altos Ca.
1976.
4. Cf., as case-studies Dorothy Nelkin, 'The Role of Experts on a
Nuclear Siting Controversy', in Bulletin of the Atomic Scientists,
vol. 30 (1974), pp. 2 9 - 3 6 ; Helga Nowotny, Kernenergie: Gefahr
oder Notwendigkeit: Anatomie eines Konflikts, Frankfurt 1979;
furthermore, writings from the extensive literature Dorothy Nelkin/
Michael Pollak, 'The Politics of Participation and the Nuclear
Debate in Sweden, the Netherlands and Austria', in Public
Policy, vol. 25 (1977), pp. 3 3 3 - 5 7 ; Edgar Michael Wenz (ed.),
Wissenscbaftsgerichtshofe: Mittler zwischen Wissenschaft, Politik
und Gesellschaft, Frankfurt 1983.
5. A good indicator is that high grades in school among those with
a positive academic attitude can accompany greater uncertainty
about self-value and greater fears over performance than average
grades. Cf., Helmut Fend, 'Selbstbezogene Kognition und institu-
tionelle Bewertungsprozesse im Bildungswesen: Verschonen schuli-
sche Bewertungsprozesse den "Kern der Personlichkeit?"', in
Zeitschrift fiir Sozialisationsforschung und Erziehungssoziologie,
vol. 4 (1984), pp. 2 5 1 - 7 0 . In view of contrary research results,
above all in regard to the inhibiting effects of anxiety on
performance (cf., Schwarzer, ibid., 1981, pp. lOOff.), and the
discovery that this is stronger among those who are more intelligent
(cf., Henk M. van der Ploeg, 'Worry, Emotionality, Intelligence,
and Academic Performance in Male and Female Dutch Secondary
School Children', in Advances in Test Anxiety Research, vol. 3
1984, pp. 201-10), the research has to view this question as still
open.
6. Cf., Heinz von Foerster, Observing Systems, Seaside Ca. 1981,
w
178 Notes
especially the contribution 'Objects: Tokens for (Eigen-) Behavior',

pp. 274ff.
7. In any event, empirical investigations that have pursued this
question have led to inconsistent results. Cf., Kenneth L. Higbeen,
'Fifteen Years of Fear Arousal: Research on Threat Appeals
1953-1968', in Psychological Bulletin, vol. 72 (1969), pp.
42644; Werner D. Froehlich, 'Perspektiven der Angstforschung', in
Enzyklopadie der Psychologie, C IV, vol. 2: 'Psychologie der
Motivation', ed. Hans Thomas, Gottingen 1983, pp. 1 1 0 - 3 2 0
(178ff.).
8. Cf., Franz L. Neumann, Angst und Politik, Tubingen 1954.
9. Werner Froehlich, Angst: Gefahrensignale und ihre psychologische
Bedeutung, Munich 1982, p. 27.
10. In any event, this is the way William C. Clark sees it in 'Witches,
Floods, and Wonder-Drugs: Historical Perspectives on Risk
Management', in Richard C. Schwing/Walter A. Albers, Jr (eds),
Societal Risk Assessment: How Safe is Safe Enough}, New York
1980, ppp. 2 8 7 - 3 1 3 . The same thing seems to be the case
for other domains of anxiety, for example, for anxiety over
examinations. See the results in D. Gertmann et al., 'Erste
Ergebnisse einer Fragebogenuntersuchung zur Priifungsvorberei-
tung im Fach Psychologie', in Brigitte Eckstein (ed.), Hochschulprii-
fung: Riickmeldung oder Repression, Hamburg 1971, pp. 5 4 - 9 .
11. Cf., for a survey William D. Rowe, An Anatomy of Risk, New
York 1977, pp. 119ff., 300ff.; for a (contested) quantitative
estimation Chauncey Starr, 'Social Benefit versus Technological
Risk: What is Our Society Willing to Pay For', in Science, vol.
168 (1969), pp. 12328. This 'double-standard' hypothesis seems
to hold empirically even if one considers the inclusion of other
factors. Cf., Paul Slovic/Baruch Fischhoff/Sarah Lichtenstein, 'Facts
and Fears: Understanding Perceived Risk', in Schwing/Albers, ibid.,
pp. 181-214 (196, 205ff.).
12. For change and adequacy in the perception of risk from the political
perspective cf., Meinholf Dierkes, 'Perzeption und Akzeptanz
technologischer Risiken und die Entwicklung neuer Konsensstrate-
gien', in Jurgen von Kreudener/Klaus von Schubert (eds), Tech-
nikfolgen und sozialer Wandel: Zur politischen Steuerbarkeit der
Technik, Koln 1981, pp. 1 2 5 - 4 1 ; also, with a detailed literature
and empirical studies of its own, Ortwin Renn, Wahrnehmung
und Akzeptanz technischer Risiken, 6 vols., Jiilich 1981.
13. For the fear of atomic disasters it is remarkable that this is
consciously estimated counter-inductively, i.e., that the calculation
Notes 179
of the risk is not obtained from statistics of past accidents but is,
so to say, projected freely and unrestrictedly. In contrast to the
perception of other risks this is quite clear in Slovic et al., ibid.,
(1980), p. 193.
14. Moreover, it is remarkable here that the fears increase while the
dangers clearly decrease. This is a case of the self-inducement of
anxiety-related communication treated above. In this field of
anxiety the 'double-standard' of involuntary/voluntary can be
observed easily. One is more afraid of the chemistry of the food
industry than one's own poor eating habits when in reality there
is more reason for worrying about the latter.
15. In the sense of the Allport school. Cf., especially Richard L.
Schanck, A Study of a Community and its Groups and Institutions
Conceived of as Behaviors of Individuals, Princeton N.J. 1932;
Ragnar Rommetveit, Social Norms and Roles: Explorations in
the Psychology of Enduring Social Pressures with Empirical
Contributions from Inquiries into Religious Attitudes and Sex
Roles of Adolescents from Some Districts in Western Norway,
Oslo 1955.
16. For the morality and logic of warning, cf., Lars Clausen/Wolf R.
Dombrowsky, 'Warnpraxis und Warnlogik', in Zeitschrift fiir
Soziologie, vol. 13 (1984), pp. 2 9 3 - 3 0 7 .
18. See also Niklas Luhmann, 'The Self-Description of Society: Crisis
Fashion and Sociological Theory', in International Journal of
Comparative Sociology, vol. 25 (1984), pp. 5 9 - 7 2 .
Chapter 20 Toward a Rationality of Ecological

Communication
1. Jiirgen Habermas, Der philosophische Diskurs der Moderne: Zwolf

Vorlesungen, Frankfurt 1985, p. 432.
2. One can object that then the question is no longer the same, but
this would only change the discussion to the problem of the criteria
for the rationality of a problematic.
3. Cf., Douglas R. Hofstadter, Godel, Escher, Bach: An Eternal
Golden Braid, Hassocks, Sussex 1979. Furthermore see above
chapter 5.
4. This is in accordance with Habermas in Der philosophische
Diskurs der Moderne, Frankfurt 1985, pp. 426ff.
180 Notes
5. Cf., also Niklas Luhmann, Soziale Systeme, Frankfurt 1984, pp.

638ff.
6. According to Heinz von Foerster, Observing Systems, Seaside Ca.
1981.
Chapter 21 Environmental Ethics
1. For more detail see Niklas Luhmann, 'Soziologie der Moral', in

Niklas Luhmann/Stephan H. Pfiirtner (eds), Theorietechnik und
Moral, Frankfurt 1978, pp. 8 - 1 1 6 ; Luhmann, 'I fondamenti sociali
della morale', in Niklas Luhmann et al., Etica e Politica: Riflessioni
sulla crisi del rapporto fra societa e morale, Milan 1984, pp. 9 - 2 0 .
2. In the seventeenth century this was quite explicit (and moreover
a requirement projected even on God: God hates the sinner!). Cf.,
for example, Edward Reynoldes, A Treatise of the Passions and
Faculties of the Soule of Man, London 1640, reprint Gainesville
Fla. 1971, pp. 11 Iff., 137ff. For such a theory, one that no longer
reflects the paradox, it is then a mere appearance (the world after
the fall from grace), that both love as well as hate can have good
as well as evil consequences. One could even talk of a parallel
coding of morality and the passions that merely has to reckon
with the fact that this can run askew.
3. Robert Musil, Der Mann ohne Eigenschaften, Hamburg 1952,
p. 1204.
4. Only the supposition is noted that a sufficiently radical reflection
has to lead consistently to the reinclusion of the paradox, perhaps
in the form of justification through unjustifiability which as a
generalized unjustifiability affects every critic simply because he
or she participates in ethical discussions and thus, at least,
recognizes the desirability of ethical justifications. Karl-Otto Apel
seems to argue in the same way. Surely it is insufficient to delegate
the question of justification simply to the ethical discourse alone
and then to wait and see whether it comes to any results and
which ones.
Index
anxiety Baechler, Jean 84

and morality 12732 Bender, Dieter 58
and politics 8 9 - 9 0 Benjamin, Walter 10
arbitrariness in environmentally Bertalanffy, Ludwig vof 4
related legal decisions 6 8 - 9 binary codes 3 6 - 4 3 , 44-5
archaic societies and ecological and the economy 52
knowledge 3 2 - 3 in education 103
Aristotle ix, 84 and functional differ n t i a t l o n
attribution 109
in religion 97 and the law 64
and observation 25, 124 and morality 139, 14
authority, political 86, 88 and politics 3 6 - 7 , 45> 8 8
autopoiesis 7 and rationality 135, ^ 6
autopoietic self-reference 23 and religion 96, 99
autopoietic self-reproduction in science 80
84 and self-observation 1^2, 123,
closed autopoietic systems 61 125
complex autopoietic systems
13-14 categorical imperative
in the economy 56, 91 and the law 65, 7 3 - ^
and functional differentiation and morality 141
107-8 causal factors and the l a w ^6
and the law 64, 65 causes and ecological pfobleins
and politics 88, 89 ^ ' 8-10
and resonance 1 7 - 1 8 , 19, 117, change, environmental, social
119 communication 32, 33
and second-order observation Christianity 5, 9 6 - 7
25 civilization and nature 2 - 3
182 Index
closed systems safe products 61

and binary codes 40, 45 costs of environmental
and the economy 61 improvement 61
and functional differentiation criteria
108 and binary codes 3 9 - 4 0
legal system 6 4 - 5 codes and programs 4 4 - 5 0
and science 79 cybernetics, ix
co-operation and competition 54 second-order 2 3 - 6 , 37
codes and coding xiv, xv
criteria and programs 4 4 - 5 0 Darwin, Charles 3
differentiation of 127 de-differentiation
in education 101-2, 103-4 and functional differentiation
and function systems 116, 117 109-10
and the law 73 in religion 97
and morality 140 decisions
and politics 85-8, 91, 120 centralized decision-making
and rationality 135 106-7
religious 9 4 - 9 and functional differentiation
scientific, of truth and falsity 111-12
76, 7 7 - 8 , 79, 81 legal 7 0 - 1
see also binary codes democracy and environmental
communication, social, 7, 2 8 - 3 1 , politics 9 2 - 3
36 differentiation
and ecological knowledge 3 2 - 5 and codes 44, 4 5 - 5 0
and resonance 16-17, 116 in education 101, 102, 104
competition and co-operation 54 of function systems 3 4 - 5 ,
complexity, viii-x 4 1 - 2 , 76, 77
and binary codes 41 of politics 85, 88
and evolution 11-14 and resonance 116
and functional differentiation of social systems 18, 19-20
110, 114 and social unity 121, 122
and the law 6 6 - 7 , 70 stratified 85
and morality 142 see also functional
in religion 95, 99 differentiation; system
in science 7 9 - 8 0 differentiation
and self-observation 122 dual-valued codes see binary
conflict and morality 140 codes
consciousness Durkheim, Emile 2
and communication 28, 2 9 - 3 1
of selection, in science 83 ecological balances and evolutioji
consensus and the law 69, 74 14 '
consumers and environmentally ecological dangers
Index 183
communicating 2 8 - 3 1 analysis of 84
and the languages of prices and anxiety 127-8
and norms 63 and binary codes 3 6 - 4 3
and the law 68 differentiation of 3 4 - 5 , 4 1 - 2 ,
economy 5 1 - 6 2 , 84 76, 77
binary code, 36, 52 and education 100-1, 104
definition of 51 and environmental dangers 115
and ecological politics 92, 93 and functional differentiation
and functional differentiation 106-8, 109, 110, 114; see
106-7 also functional
and the law 63 differentiation
limitation of 5 1 - 2 and observation 26
and morality 46 and rationality 135, 136
and politics 63, 90, 91 and resonance 115-20
and resonance 117, 118, in society 127
119-20 functional differentiation 4 2 - 3 ,
and responsibilities 10 106-14, 121
and self-observation 123-4 of codes 4 6 - 5 0
ecosystems 81 and the economy 52
education 42, 100-5 and the law 6 3 - 4
environment, Bender's definition and politics 85, 89
of 58 and rationality 137
environmental compatibility, and resonance 115, 116
testing of 67 and self-observation 125
environmental complexity 11-14 in society 127
environmental dangers and
religion 98 goods, balancing of, and ethical
Erasmus 34 responsibility 70
ethics Greece, ancient
and the balancing of goods 70 and binary codes 43
environmental 5, 55, 139-42 view of environment 6
evolution Green political movement 86, 89,
and binary codes 38, 42, 43 90, 126
complexity of 11-14
Habermas, J. 82, 133, 134, 136
falsity and the scientific code 76, Harnack, Adolf von 126
77-78, 79, 81 Hegel, G. W. F. 108
Frankfurt School 133 Hofstadter, Douglas 60, 111
freedom and the law 65, 66, Husserl, Edmund x, 17, 29, 77,
67-8 82
French Revolution 3, 46, 88
function systems xiii-xiv, xv identity, collective 133-5
184 Index
immanence in religion 95, 96, methods, scientific, and theory

97, 98 79
investments in the capitalist money and the economy 51, 52
economy 55 morality
invisible hand and anxiety 127-32
and the economy 46, 60 and causes 9
in religion 98 and ethics 139-42
and prices 53
just price, theory of 53 and religion 95
justice and science 78
codes and programs 48 and the self-destruction of
and the law 74 society 5
Miinchhausen, Baron 80
knowledge
ecological, and social natural law 69
communication 3 2 - 5 and binary codes 45
scientific 78, 81 and politics 89
and prices 53
language and resonance 16-17 and theory of law 47, 48
law 6 3 - 7 5 , 84 nature
binary code 36 and Christianity 5
codes and programs in 478 and civilization 2 - 3
and ecological problems 8 domination of 14
ind economic restrictions 53 and natural law 69
enforcement of 75 and religion 96, 97
environmental 67, 74 New Guinea, pig cycle in 3 2 - 3
and politics 9 0 - 1 norms
and resonance 117-118, language of 63
119-20 and the law 63, 64, 74, 75
see also natural law and morality 127
Luther, Martin 34
observation
market theory 54, 55, 60 and binary codes 37
Marx, Karl 125, 126 observation of 2 2 - 7
Maturana, Humberto x, xi, 23 in science 7 9 - 8 0
medieval society/Middle Ages second-order 23, 141
and functional differentiation self-observation 9, 20, 79-80,
121 113, 121-6, 131
and law 47 obstruction, politics of 90
and politics 87 open systems
and religion 95 and binary codes 40
view of environment 6 conditions of 84
Index 185
and functional differentiation, and risks 71
108 printing and social
legal system 6 4 - 5 communication 34
and science 79 production organizations and
environmental protection 61
paradoxes profits
and anxiety 129 and environmental
and binary codes 37, 38, 39 opportunities 57, 59
in the economy 60 and risks 69, 71
and functional differentiation programs and programming
108, 113 xiv-xv, 4 4 - 5 0
and morality 140, 141, 142 differentiation of 127
and rationality 134, 137 and the economy 52, 53
in religion 97 in education 101, 102, 103,
in science 81 104
Parsons, Talcott viii, ix, x, xii and the law 6 4 - 5
payments and the economy 52, and politics 86, 87, 88
53, 5 5 - 8 , 61, 91, 117 and rationality 135
Plato 84 religious 94, 96, 99
politics 8 4 - 9 3 and resonance 116, 117
and anxiety 128 and science 76, 77
and binary codes 3 6 - 7 , 45, 88 progress, belief in 46
and the 'causer-principle' 8 - 9 property
and the economy 51, 54, attitudes to 3
59-60 and the economy 52
and education 1 0 4 - 5 public opinion and politics 89
and functional differentiation
107 rationality
and the law 63, 64, 66, 73, of ecological communication
74-5 133-8
and resonance 117, 118, in science 8 2 - 3
119-20 of system and environment
and responsibilities 10 131
and science 76 reciprocity, generalized, and the
power and politics 8 5 - 6 , 8 9 - 9 0 , law 7 3 - 4
91 redundancy, social 115, 116
prices rejection of 1 1 0 - 1 1
and the economy 5 3 - 4 , 56, reflection, theories of 1 3 6 - 7
57, 62 religion 9 4 - 9
and environmental and anxiety 128
opportunities 58 and binary codes 42, 4 5 - 6
language of 63 and differentiation 1 2 1 - 2
186 Index
andêcological problems 3 3 - 4 and causal interconnections 9

and the economy 51 and resonance 117
and education 101 self-observation 127
and nature 2 - 3 self-reference xi, xiv, 2 3 - 4
and science 78 and binary codes 37
and self-observation 123 and the economy 62
representation and resonance 20 and rationality 136
resonance ix, xv, 1 5 - 2 1 , 22 in religion 95
and anxiety 129, 131 in science 79
codes and programs 50 self-regulation in science 81
in the economy 5 4 - 5 , 57, 59, Serres, Michel 38, 99
61-2 Shaftesbury, Anthony, 3rd Earl
in education 101, 104 of 127
and function systems 49, 84 Smith, Adam 46
and the law 68 Social Darwinism 3
and observation 22, 24, 25 social differentiation and politics
and politics 85, 8 8 - 9 0 , 9 1 - 2 85
and religion 94, 9 8 - 9 social functions of the law 65
restriction and amplification social movements 105, 1 2 1 - 6
115-20 social problems 2
and science 76, 8 0 - 1 , 82 society
and social communication 32, and communication 2 8 - 3 1
35, 36 and politics 8 4 - 5
and social movements 124 and resonance 117
responsibilities and ecological unity of 1 0 6 - 7 , 1 1 3 - 1 4
problems 8 - 1 0 sociological abstinence 17
rights and the law 6 7 - 8 sociology and functional
risk differentiation 106, 113
and anxiety 128, 129 sovereignty, territorial, and
and the law 68, 6 9 - 7 2 , 73, 74 political resonance 91
and morality 1 4 1 - 2 state government 87
Rousseau, Jean-Jacques 46 stratification
Russell, Bertrand xiv, 23 and functional differentiation
108-9
scarcity of resources political 88
effects on different function structural contingency 1 1 1 - 1 2
systems 49 structured complexity in science
and functional differentiation 79-80
49, 109 subsystems
paradox of 60 and functional differentiation
science 7 6 - 8 3 107, 111
binary code 36 and self-observation 123
Index 187
synthesis and scientific transcendence in religion 95, 96,
knowledge 79 97, 98
system differentiation 18-19, 20 truth and the scientific code 76,
codes and programs 48 7 7 - 8 , 79, 81
and self-observation 123
system rationality 138 values
systems theory 4, 5, 6 - 7 and binary codes 4 0 - 1 , 45,
and ethics 142 103, 125
and functional differentiation and functional differentiation
107-8 111-12
and rationality 134 and the law 73
and resonance 16 and science 79
and the theory of evolution 13 Varela, Francisco x, xi, 29
Tarski, Alfred xiv, 23 Wealth of Nations (Smith) 46

tautologies and binary codes 37 Weber, Max 2, 87, 138
technological change and risks 72 Whitehead, A. N. xiv, 23
theodicy 98 writing and social
theories, scientific, and method communication 34
79
time-loss and the economy 57 Index by Isobel K. McLean

Ecological Communication, Niklas Luhmann PDF

Uploaded by

Copyright:

Available Formats

Ecological Communication, Niklas Luhmann PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ecological Communication, Niklas Luhmann PDF

Uploaded by

Copyright:

Available Formats

Ecological

The University of Chicago Press

The University of Chicago Press, Chicago 60637

Library of Congress Cataloging-in-Publication Data

This book is printed on acid-free paper.

Translator's Introduction vii

18 Representation and Self-observation

Ecological Communication enjoys a unique position in the

powerful synthesis of several quite diverse intellectual traditions.

of expectations in the process of human interaction'. 4

psychical systems are unique because this connection is

themselves are so related that they constantly (re)produce each

theory of the social domain.

in accordance with its own mode of operation, the mode of

consequential communication does.

the unity (of self-reference, J.B.) that would be unacceptable in

Function systems structure their communication through binary

for society. They screen it off from its environment organization-

a new environmental morality^iAccordingly, right is on the side

John Bednarz Jr.

On 15 May 1985, at the invitation of the Rhenish-Westfallian

In many respects the analyses presented here diverge from the

Compared to the history of reflection on humanity and society

cerned with the internal aspects of society. It entangled itself in

was retained) as the counter-concept to nature.4 Thereby nature

the self-critique of society vis-a-vis determinate ideals, not

has left little time for theoretical consideration. Initially, therefore, !

of unity and difference, i.e., in terms of the unity of the ecological

i.e., from unity to difference as the theoretical point of

2 The idea of system elements must be changed from substances

Causes and Responsibilities?

Once one acknowledges the phenomenon of evolved complexity

significance of the causer-principle, then, is not a causal statement

have consequences themselves. Political coalitions, for instance,

this cannot be extolled as responsibility. The difference must be

Complexity and Evolution

reduce environmental complexity - primarily by restricting the

structures can meet such demands?

macromolecules within cells) but eventually creates systems from

Concepts like complexity, reduction, self-reference, autopoiesis

Physics can also be called on to help us. A differentiated system

would establish boundaries simply through its coming into being

processing is continued. We can designate the structural technique

of its own operations. Instead, this equips it with the capacity

2 The system's unity can, if necessary, be represented within the

3 Since every operation is only one among many, every

Further inspection reveals that the theory of self-referentially

Instead, the observation accompanying all political, economic

The Observation of Observation

System resonance, then, is always in effect when the system is

self-reference while it distinguishes several levels as a hierarchy

the assumption of let us say, a supermodal, observationally

with the necessary exactness, second-order cybernetics must be

is only a particular kind of observing of its own environment.10

In the following I will not consider the very limited, socially

not the subject of communication it has no social effect. Society

Viewed realistically, we have to reverse the idea that a 'subject'

proposals than to present it in communication as positive

Ecological Knowledge and