CYBERNETICS & HUMAN KNOWING

A Journal of Second Order Cybernetics & Cyber-Semiotics


Vol. 1 no. 2/3 1992

Søren Brier:
Information and Consciousness: A Critique of the Mechanistic Concept of Information

Abstract

The paper presents a discussion of the epistemological and ontological problems of attempts to found information concepts on the often implicit mechanistic idea that the physical sciences hold the key to the nature of reality and information. It is furthermore shown through an analysis of the ethological and the Batesonian understanding of cognition and behavior that it is impossible to remove the fundamental epistemological position of the observer through a definition of information as neg-entropy. Instead Maturana and Varela's concepts of autopoiesis and multiverse are invoked. But where the idea to derive information from the concept of negentropy is too physicalistic Maturana's idea of a multiverse seems to be too close to a constructivistic idealism. To develop a more fruitful non-reductionistic world view it is shown that the more pragmatic understanding of physics, where thermodynamics is understood as the basic discipline and mechanics as an idealization, opens for a non-reductionistic con-ceptualization of chaos. Attention is drawn to C.S. Peirce's conception of pure chance as living spontaneity which is to some degree regular as a realistic but non-reductionistic theory, which comprises a solution to the different world view problems of Bateson and Maturana. A fruitful connection between second order cybernetics and semiotics will then be possible and a bridge between the technical-scientific and the humanistic-social parts of cybernetics can be developed.

Introduction.

In recent years a new area of growth in the interdisciplinary sciences has created consider-able confusion about the proper basis for reflections on the possibilities and limits of human knowledge. This interdisciplinary growth area, called information science, cognitive science and artificial intelligence draws on subjects such as information theory, linguistics and philosophy of language, neuro- and computer science, anthropology. With behaviorism's fall from greatness, cognitive science is scientific rationality's newest and most powerful attempt to find a firm foundation for knowledge. Gardner (1985 p.6 and 8) writes:

"I define cognitive science as a contemporary based effort to answer long-standing epistemological questions - particularly those concerned with the nature of knowledge, its components, its sources, its development, and its deployment. (...)

Approaching these fields from a methodological point of view, I raise the questions whether philosophy will eventually come to be supplanted by an empirically based cognitive science, and whether anthropology can (or even should) ever transcend the individual case study."

The idea of purely objective (empirically based logico-mathematical) knowledge appears to be inherent to the basic philosophy of modern science as Galileo formulated it. By insisting on, that the major qualities in nature can be revealed by methods of measuring and weighing (empiricism), that mathematics is nature's language and finally that nature is basically composed of indivisible lumps of matter - atoms - the foundation was laid for an unhampered hunt for knowledge about nature. There are no limits to this mathematical, mechanical knowledge. A final insight and thereby control of nature is seen as possible in this mechanical materialism whose God is Laplace's demon.

The philosophical attitude found in extensive areas of research on artificial intelligence and the functionalistic areas relating to it appears to be a type of mechanical materialism. One believes, apparently, that information is something found in the "world outside", independent of the observer. As materialism at the present time is always joined with a fundamentally evolutionist attitude, this view's logical consequence is that information must have existed before the emergence of consciousness.

A recently published textbook "Information Science - Theory and Practice" (Vickery and Vickery 1989, p.43) states that information is just as fundamental a part of reality as matter and energy. On the one hand they seem to represent a classical mechanistic view of nature, and on the other they see nature as being full of information. There is no doubt that Vickery and Vickery are promoting a materialistic world picture where knowledge is more fundamental than the observer. A world picture, where energy, matter and information are the basic elements, as they state (Vickery and Vickery p.43). This makes life, consciousness, humans and language later by products, whose creation and function they probably expect science to explain "in the long run" through the study of evolution, genes and artificial intelligence.

In this fashion energy, matter and information become mutually joined concepts that explain the emergence, formation, structure and dynamic of mind and knowledge.

According to this world view natural objective information must have been present before the living beings and the human minds occurred on the scene in the history of the expanding Universe. In this view information is more fundamental than the observer or interpreter.

In the mechanistic scientific view of knowledge knowing is eternal and universal mathematical structures or laws which control all movement, matter and force including those determining the behavior of the living beings and language and rationality of the humans. It is believed that the sciences have and will give us access to the fundamental natural laws behind all matter. Mechanical scientific ideology would thus seem to be close to the final scientific totalitarianism, where philosophy, cosmology and epistemology are wrested from philosophy - as Gardner hints at in the above quotation - and then become a total science that can be formulated in "nature"s own language": mathematics.

Accepting information as such an objective universal law-determined thing, which humans absorb into their minds and machines from nature, change and multiply by thinking, and bring into society through language, means it must be possible to establish a new unifying science of information.

Information science then will also include cognitive science, and all epistemological problems will be solved empirically. It is this development - along with the development of the cognitive science(s) - which has promoted the idea of a unified information science for man, machines and animals. (The Vickery's even include all other natural systems).

The basic idea of cognitive science seems to be that information processing follows certain "universal syntactic, logical and mathematical laws", among other things inspired by a mechanistic view of all nature and by Chomsky's theory of a common "deep" generative grammar behind all languages. This is the Cartesian foundation for cognitive science.

Gardner (1985 p.6) sums up the basic goals and beliefs of cognitive science very aptly:

"First of all, there is the belief that, in talking about human cognitive activities, it is necessary to speak about mental representations and to posit a level of analysis wholly separate from the biological or neurological, on the one hand, and the sociological or cultural, on the other.

Second, there is the faith that central to any understanding of the human mind is the electronic computer. Not only are computers indispensable for carrying out studies of various sorts, but, more crucially, the computer also serves as the most viable model of how the human mind functions....

The third feature of cognitive science is the deliberate decision to de-emphasize certain factors which may be important for cognitive functioning but whose inclusion at this point would unnecessarily complicate the cognitive-scientific enterprise. These factors include the influence of affective factors or emotions, the contribution of historical and cultural factors, and the role of the background context in which particular actions or thoughts occur.

As a fourth feature, cognitive scientists harbor the faith that much is to be gained from interdisciplinary studies... .

A fifth and somewhat more controversial feature is the claim that a key ingredient in contemporary cognitive science is the agenda of issues, and the set of concerns, which have long exercised epistemologists in the Western philosophical tradition."

So the idea is that feelings and cultural context are not essential, the model for cognition is the algorithmic machine, and the goal is that epistemology and cognition will be understood as aspects of "Human Information Processing". This is the title of Lindsay & Normann's widely used cognitive science supportive textbook on psychology (Lindsay & Normann 1977, and later editions). Here is an example of how they state their view:

"Nevertheless, the principles of information processing are highly relevant to all systems that make use of information, including the human mind. The general principles of information processing must apply to all systems that manipulate, transform, compare, and remember information."

(Lindsay & Normann p.579)

This will eventually become a new super science: Information science. This science will restructure the whole field of knowledge. It will be a true science of sciences! And clearly mechanical physics is its ideal of science.

Stonier (1990) writes about the need to develop a general theory of information based on a physicalistic paradigm:

" To create such a theory, we need to start with the most fundamental aspect of information. And the most fundamental aspect of information is not a construct of the human mind but a basic property of the universe. Any general theory of information must begin by studying the physical properties of information as they manifest themselves in the universe. This must be done before attempting to understand the various, and much more complex forms of human information. The next step must involve an examination of the evolution of information systems beyond physical systems - first in biological, then in the human, cultural sphere".

(Stonier 1990 p.112-113)

My rejection of the above described paradigm of information science is based on views similar to Machlup's (1983) and Winograd & Flores' (1987) statement that information basically is something a person communicates to another person. And the meaning of information can only be understood considering living mindful beings in a social/cultural context and in an historical perspective. Furthermore I agree with Machlup, when he points out, that you cannot even define all sorts of information as having the ability to reduce uncertainty. But the fact is, that some kinds of information will even make the receiver more uncertain than they were before. I further agree with Searle (1986), when he states, that the common link between information processing in man and machines is not, that they both follow rules. Machines behave according to causalities, only conscious beings can choose to follow rules.

In my opinion knowledge - or knowing (to underline the process) - is a far more complicate "thing" or process than expected from the above described "information processing paradigm". According to Thomas Kuhn's philosophy of science nature and human mind is not directly linguistically connected through perception. Nature does not - in the usual meaning of the word - talk to us or transfer information to us through our observation.

I think that a more fruitful way to look at it, is to say, that nature is full of differences, but - as the ethologist and the cybernetician Gregory Bateson (1973) points out - only some of them make a difference to us and thereby they become information for us. Although both theories represents fruitful alternatives to positivistic and mechanistic thinking both the ethologists and Bateson look for a physicalistic foundation for their theory of information.

In this paper I will show how the works of the ethologists and Bateson gives an alternative scientific explanation of what the nature of information and cognition is. I will then analyze the shortcomings of the search of the ethologist and Bateson for a physical foundation of a theory of cognition and information for animals and man in the relation between thermodynamics and information theory. I will show how the theory of autopoiesis solves some of Bateson's problems about who the difference makes a difference for, even though it is still unclear about the relation between mind and matter. Finally I will show how the founder of semiotics, C.S.Peirce, long ago gave a more frutful foundation of a theory of information and knowing than any form of physicalism, has ever done.

An ethological approach to cognition

Approaching the problems of the mechanical foundations of information it is relevant to discuss the fundamental problems of ethology, the biological science of behavior, as ethology in a general way deals with the cognitive and behavioral interaction between the organism and its surroundings. (See Tinbergen 1968, 1973)

Ethology observes the interaction between organisms' behavior and their milieu. Ethology is behavioral research with a theoretical foundation of biological-natural history and asks questions as: how has behavior developed phylogenetically and ontogenetically and how does it contribute to the survival of the organism as an individual and as a species. Ethology's goal is to describe and explain of the cognition and behavior of organisms within an evolutionary-ecological frame of reference. Ethology is first of all concerned with the instinctive forms of the specific innate behaviors of species. Furthermore, ethology deals with the various forms of learning that the different instinctive conditions make possible (see Bittermann 1965, and Lorenz 1976), partly where they show themselves as "constraints on learning", and partly where they occur as surprising outbreaks of "intelligent thinking" (for instance, the special abilities of rats to find their way in labyrinths). As Reventlow (1973) writes:

"As ethology demands that animal and human behavior be examined in relation to their biotopes and at all times strives to understand psychological phenomena on the basis of the relationships between stimuli, motivation and behavior, ethology becomes one of the broadest psychological theories in existence - and certainly the most ecologically oriented".

The model of reaction and releasing of behavior arrived at after 50 to 60 years of work concerning behavior's internal organization (a motivation model) is an attempt to "integrate in a simple way the effect situation, inner motivation and the outer-observable behavior" (Reventlow 1970) in a theory based upon the central concept of the "innate release response mechanism", which ethologists believe is found between perception and the primarily instinct-guided behavior.

Now, focusing on the concept of information, it is particularly relevant to analyze at the innate release mechanism. Seen from the point of view of the human observer, only small parts of the stimuli which usually occur in nature are used to activate the behavior. These parts have been characterized as "sign stimuli", i.e.,stimuli which without previous learning activate certain behaviors in an organism. Sign stimuli lead to sensations which corporate with internally motivated conditions in such a way as to release the genetically determined behavior.

The salient point is that the action-releasing perception of these sign stimuli is always connected to definite states of motivation! When von Frisch's - the discovere of the language of the bees - studies of the color sense of bees were published, von Hess denied that it was possible for bees to differentiate between colors. The experiments he himself had done on bees' escape behavior showed that they always flew toward the patch of light with greatest intensity, regardless of color. Recent studies shows that he was right, - but von Frisch was also right in claiming that bees have a sense of color which they use when seeking food; there were two different innate release response mechanisms, related to flight motivation and the search for food, respectively.

The significant result from ethology concerning the relation between reality, perception and information is that perception is not a purely mechanical process where objective information is passively taken in, - on the contrary perception requires motivation. All of ethology's explanations of the release and functionality of behavior build upon this concept of action specific motivation.

From an ethological point of view this type of process is the primitive starting point for human perception. Our cognition of information is controlled by "interests" that derive partially from the nature of the perception apparatus (Lorenz 1976) and partly from the innate release response mechanisms found in our organism. In his development of the concept of instinct and motivation, Lorenz fought for a long time with the causal role of the organism's mental experience in the release of behavior (Brier 1980), but apparently ended in 1950 by finally denying that it should have any causal role.

But in his later development of a theory of learning, based on the instinctual structure as the foundation and constrain of the organism's learning abilities, he again emphasizes experiential-emotional aspects as reinforcements of in the trial-and-error process (Lorenz 1976). Since it is basic to the theory that instinct and learning always function together a shift in attitude is indicated (Brier 1980).

Beer (1982) (unfortunately without precise indication of source) has found an apt description of this motivational-intentional drive in the following quote from McDougall's pre-ethological definition of instinct as:

"...an inherited and innate psycho-physiological disposition which determines its possessor to perceive and pay attention to objects of a certain class, to experience an emotional excitement of a particular quality upon perceiving such an object, and to act in regard to it in a particular manner, or at least to experience an impulse to such action".

Where organisms have to adapt fast to changes in their environment their instincts eventually in the course of evolution "open" different parts of their "program" towards learning; this often often takes place in so called "sensitive periods". Examples are the "imprinting" of ducklings, the song learning of many birds and children's language-learning. In the culturally based upbringing of children the basic human motivations and needs are developed and differentiated and with self-consciousness feelings are developed. From the phylogenetic perspective of evolution, the differences that observers register are heavily dictated by biologically and culturally created interest. To some extent, different species and different cultures live in different worlds.

Influenced by mechanistic thinking ethologists for many years have attempted to find physiological explanations for the different kinds of motivation and their influence on perception. Under the headline "Evidence for drive" Thorpe (1979) has written the following, which I consider to be very typical of the English school's relation to the ontological status of the concept of motivation:

"We have shown above that a very real problem is raised by the question of the motivation of particular stereotyped actions. The name used by the pioneer ethologists, following Lorenz, was "action specific energy". This was a thoroughly bad term and was later softened to "specific action potentials" etc. The increase in responsiveness is an outward sign of an internal change, and what most, if not all, ethologist want to discover is the physiological basis for such change. So the very question of the internal motivation of specific acts leads us straight into the general problems of "drives"".

But in a conclusive analysis of the role of the concept of motivation in ethology and comparative psychology Hinde (1970) shows that this is a logical category mistake. Hinde's conclusion on his critical analyses is that explanations of motivation are founded on an entirely different descriptive level from the physiological and behavioral descriptions.

One can perhaps expand the perspective in his conclusion by pointing out that the concept of motivation can be said to occupy the same position within ethology as the concept of life in biology and the law of gravity or the attraction of masses in Newton's classical mechanics. It is, in other words, an "occult" basic concept which cannot be explained within the paradigm but which one cannot ignore.

The Danish psychologist K.B Madsen (1978) documents this position of the motivation concept in ethology and other schools of thought within behavioral research in a paper summing up his many years of research into the concept of motivation (Madsen 1974). He writes:

"Under the name of "instinct", "need", "motive", "tendency" etc. the motivation variable has been introduced into psychological theories to explain behavior. Therefore one does not find the motivation variable in the psychological epochs or schools of thought where psychology's task is looked upon as the description of consciousness. One can not find motivation variables in classical experimental psychology (Fechner, Wundt, Ebbinghaus, etc.) Only with the advent of Freud's psychoanalysis and American learning theory did one attempt to explain behavior by introducing concepts such as "drive" and the like (for instance Freud's "Trieb" and Woodworth's "drive".) The concept of motivation spread from psychoanalysis to the theory of personality (for instance, Murray's "need"). After world War II, motivation psychology gradually became an independent psychological discipline..."

The concept of motivation would thus appear to be a "remnant" from the classical psychology of consciousness in mechanistic behavioral research. The desire to explain the release of observable behavior as dependant on "inner" motivation, drive, tendencies, or needs appears to be an attempt to turn the original phenomenological description level's incentives and intentions to something a bit less subjective and anthropomorphic! But this is not the proper way to make a theory of cognition more scientific. Concepts of consciousness cannot in this way be reduced to physiology. One can, at the most, claim that there is a one-to-one correlation between mental conditions and (neuro) physiological states.

But as Jensen (1973) has shown such a correlation theory can not be maintained. The human brain and the brain of the higher mammals are so complex that it is highly unlikely that any nerve impulse would ever run the same way, even in simple reflexes. Here we face an immense complexity!

The complexity of neurophysiological states makes it necessary to classify them based on some kind of interest or point of view. Jensen points out that it is culture through language that performs this classification for the referrals of the concepts via the conventions determined by the surroundings during a child's upbringing.

Karl Popper (1972 A) argues along the same lines in his clash with the idea that scientific hypotheses via induction emerge spontaneously from the researcher's observations. No matter how many times we have predicted and observed an empirical phenomenon we can never prove the existence of a natural law because, among other things, the experience of something repeating itself is not a value-free, absolutely objective observation but one based on a judgement of similarity.

Hume's argument, which Popper develops further, is that repetition is based on similarity. But repetition-for-us is based on similarity-for-us, that is, we identify situations as identical. But it is highly unlikely that the situations we observe are completely identical (they are in reality too numerous and varied in composition and appearance). It can only be a matter of similarity. The danish psychologist A.F. Petersen, who is developing on Popper's theories within ethological perspective, writes (1972 p.46):

"The problem of repetition arises the moment an organism's viewpoint is introduced. The world of phenomena can not be shown to be repetitive in its own right; we do not know whether the laws of nature that have been revealed to us will apply in the future. The world of phenomena has an objectively historic course, but no one can guarantee that the historic laws we find apply for all time and in every milieu which might arise. This does not mean that the knowledge an organism can possess must be subjective. Figuratively speaking, the organism finds itself in the objective stream of phenomena and by assuming certain attitudes towards this stream of phenomena the organism will perhaps be able to discover phenomena which, seen from its particular perspective, repeat themselves. This will perhaps reveal itself at the first step to understanding certain connections between the phenomena and to explain the respective connections within the given spatio-temporal region".

Summing up, we can add that things can only be classified and become identical or different in this way, i.e. by being placed in relationship to needs or interests. This rule applies not only to animals, but also to human beings - even to scientists.

So the concept of information - although not purely subjective - cannot be defined without the observer and his/its motivational and intentional selective attention. How then can this problem be solved? I believe that Gregory Bateson's cybernetic theory of mind is an interesting attempt to give a new fruitful non-mechanical understanding of the relation between the concept of motivation-intentionality and information.

Gregory Bateson's concept of information.

One of the major projects in Bateson's books (1973, 1980) is an attempt to solve the problem of explaining the nature of mind on a modern scientific basis, avoiding a metaphysical dualism à la Descartes and the mechanicism of Laplace. Bateson is able to provide a new delimitation of the concept of information, which attempts in a more consistent way to unite a scientific and materialistic world view with the concept of mind in agreement with a non-mechanistic view of ethology.

Ethology then - in my view - provides a scientific basis for understanding Bateson's redefinition of the concept of information:

"In fact, what we mean by information - the elementary unit of information - is a difference which makes a difference,...." (Bateson 1973, p. 428)

This is a wonderful foundation for the understanding of cognition in a world of complexity. The world view is not subjectively idealistic or solipsistic. There is a world of differences or potential information, but it is the observer who has to "bring them forth" (to use the expression of Maturana and Varela 1986).

Bateson's world view is scientific and materialistic but not mechanicistic. His "working hy-pothesis" of the world is that its basic constituents are space, time, elementary particles (matter) and energy. He thinks that science will come to an end if we have to endow the elementary particles with mind qualities (Bateson 1980 p.103). His project is to explain mind as a function of complexity and cybernetic organization. Bateson believes that the strength of cybernetics lies in its ability to provide a more profound understanding of what the mental is, incorporating his concept of information in a universal cybernetic philosophy. Bateson thought that his cybernetics could provide an understanding of mind which was neither subjectively idealistic or mechanically materialistic. In his work to unfold this vision he contributes to the development of the classical cybernetics into second-order-cybernetics. There are two major reasons why he did not quite succeed:

1. He did not manage to develop the concept of autopoiesis.

2. He was not able to liberate his concept of information from Norbert Wiener's. Al-though Bateson's definition of information seems well suited for second-order-cyberne-tics he had - as I will show - tied it to the concept of neg-entropy, which gives his theory a physicalistic flavour.

Mind, information and entropy.

To Bateson mind is a cybernetic phenomenon, a sort of mental ecology. The mental relates to the ability to register differences. It is an intrinsic system property. The elementary, cybernetic system with its messages in circuits is in fact the simplest mental unit, even when the total system does not include living organisms. Every living system has the following characteristics which we are accustomed to call mental:

"1.The system shall operate with and upon differences.

2.The system shall consist of closed loops or networks of pathways along which differences and transforms of differences shall be transmitted. (What is transmitted on a neuron is not an impulse, it is news of a difference).

3. Many events within the system shall be energized by the respondent part rather than by impact from the triggering part.

4. The system shall show self-correctiveness in the direction of homeostasis and/or in the direction of runaway. Self-correctiveness implies trial and error." (Bateson 1973 p. 458)

Mind is synonymous with the cybernetic system which is comprised of the total, self-correcting unit that prepares information. Mind is immanent in this wholeness. When Bateson says that mind is immanent he means that the mental is immanent in the entire system, in the complete message circuit.

One can therefore say that mind is immanent in the circuits that are complete inside the brain. But mind is also immanent in the greater circuits which are complete inside the system "brain + body". Finally mind is immanent in the even greater system "man + environment" or - more general - "organism + environment" which is identical with the elementary unit of evolution, i.e. the thinking, acting and deciding agent. This is the mind-system:

"The individual mind is immanent, but not only in the body. It is immanent also in pathways and messages outside the body; and there is a larger Mind, of which the individual is only a subsystem. This larger Mind is comparable to God and is perhaps what some people mean by "God", but it is still immanent in the total inter-connected social system and planetary ecology. Freudian psychology expanded the concept of mind inward to include the whole communication system within the body - the autonomic, the habitual and the vast range of unconscious processes. What I am saying expand mind outward. And both of these changes reduce the scope of the conscious self. A certain humility becomes appropriate, tempered by the dignity or joy of being part of something bigger. A part - if you will - of God". (Bateson, 1973, pp. 436-37).

Bateson is also clearly in line with the ethologists when he resists the positivistic split between the rational and the emotional in language and thinking which is so important for cognitive science. He writes ( Ibid p. 438):

"It is the attempt to separate intellect from emotion that is monstrous, and I suggest that it is equally monstrous - and dangerous - to attempt to separate the external mind from the internal. Or to separate mind from body. Blake noted that "A tear is an intellectual thing", and Pascal asserted that "The heart has its reasons of which reason knows nothing". We need not be put off by the fact that the reasoning of the heart (or of the hypothalamus) are accompanied by sensations of joy or grief. These computations are concerned with matter which are vital to mammals, namely matters of relationship, by which I mean love, hate, respect, dependency, spectatorship, performance, dominance and so on. These are central to the life of any animal, and I see no objection to calling these computations "thought", though certainly the units of relational computation are different from the units which we use to compute about isolable things".

It thus seems obvious that Bateson's "pattern that connects" includes the phenomenological-emotional dimension in its concept of mind. In my opinion this cybernetic viewpoint tells a great deal about what motivational and emotional functionality is, seen in an ecological and evolutionary framework. It seems to avoid physicalistic explanations. But it has its limits.

Autopoiesis, mind and information.

The first major obstacle for Bateson's theory in its attempt to become a general theory of information and knowing is the difficulty of determining to whom or what the difference makes a difference. His concept of mind-system is too general to be the agent to whom the difference makes a difference as it extends from the smallest feedback loop in machines to the working cybernetic system of Gaia.

In my view Maturana has developed this analysis an important step further. Maturana's and Varela's (1980) definition of the autopoietic system answers the question as to whom the difference makes a difference. It is the autopoietic system that is able to react to differences (perturbations).

The point is that the autopoietic system does not receive information in an objective physical sense and responds to the stimuli or information as it would be understood in a mechanistic paradigm. No, it is a closed organization whose main concern is to stay organized, to survive. It does so by changing its organization when its coherence is threatened by perturbations from "outside" or by spontaneous behavior from inner parts. This we can observe as physiological reactions and behavior. Through this autopoietic process the organism brings forth a world. Maturana (1988 p.26) writes:

"... a composite unity whose organization can be described as a closed network of productions of components that through their interactions constitute the network of productions that produce them and specify its extension by constituting its boundaries in their domain of existence, is an autopoietic system; and second, that a living system is an autopoietic system whose components are molecules. Or in other words, we propose that living systems are molecular autopoietic systems and as such they exist in the molecular space as closed networks of molecular productions that specify their own limits..."

In the next step of developing the theory Maturana tries to show that this is not only a scientific theory of knowledge - it is a general science of knowing. He now uses this theory of knowledge obtained by thinking within a biological frame in a reflexive manner, to make a general philosophy: a self-organizing and self-sufficient epistemology with no need to refer to an objective existing world of ideas and theories: the autopoietic epistemology.

The argumentation is as follows. When we realize that there is no objective reality for any autopoietic (biological) system, then there is no objective reality for us - or science - either. There are no autopoietic systems without an observer who "brings them forth". This means that there is no objective reality that can legitimize our claims to truth. The universe is a multiverse (Maturana and Varela l987, Maturana 1988). Together, through languaging and social practice we bring forth our worlds.

I find that his theory gives a better foundation for Bateson's theory of information through the important point that the autopoietic system does not receive information but only perturbations of its organization. The so called reaction is internal adjustments to preserve the internal organization of the system. It underlines the distinction between the world and its being of differences - potential information if you like - and information processing of the autopoietic systems. In fact neither Maturana or Varela like to use the concept of information.

The limits of "bring-forth-ism".

Krippendorff (1991) mentions the many different kinds of constructivism flourishing within second-order-cybernetics, and mentions Maturana and Varela's "bring-forth-ism" as one of them. Without choosing any of them he sums up the essential goal of second order cybernetics and constructivism:

"The task of constructivism, as I see it, is to describe a system's operation within its own domain of description and account for the constitution of its identity and the conditions of its continued persistence in its own terms. Said differently, constructivists need to find a way of putting the knower into a known that is constructed so as to keep the knower viable in practice."

To accomplish that would - as Nagel (1986, s.74) points out - be an important step forward in objectivity:

"We tend to use our rational capacities to construct theories, without at the same time constructing epistemological accounts of how those capacities work. Nevertheless, this is an important part of objectivity."

It is clear that Maturana and many cybernetic constructivists have good arguments against physicalism. But when Maturana writes and talks about the autopoietic system and the domain its organizational closure creates - and underlines the general point that there are no autopoietic systems without an observer to distinguish them - then he seems to be in a pure phenomenalistic and solipsistic position.

Behind this position lies, however, the basic problem, that one can't talk about reality in itself. It makes no sense to talk about reality without being able to have access to the operations which bring forth this result (Maturana 1990). In other words it is illusory to talk about reality because each of us will assume we know what the other means. But it is an assumption only. We are safer not to talk about reality at all. It is more scientific not to talk about autopoietic systems when there is no observer. This understanding gives a much softer version of Maturana's basic epistemology .

Talking of languaging as a kind of co-evolution between two or more autopoietic systems stabilizing a common domain of distinction of distinctions, he seems to adhere to some kind of social constructivism, which is the general trend among the second order cyberneticians because most of them deal with social systems, without being primarily interested in natural phenomena. But that will not explain the very rigid and narrow constrains physical things put on our constructions, and Maturana (1988) is aware of the thermodynamic conditions for existence of living beings. That again leaves us with Maturana's fundamental question of the relation between the observer and the observed.

The problem is that where "bring-forth-ism" and constructivism seems to reveal something important about the relativity of social concepts, belief-systems and institutions it is very difficult to reconcile with the experience of material things. They seem to force themselves upon us with independent necessity no matter how we choose to perceive them or ignore them. It was well founded that the concept of 'things' was so central to the philosophy of Aristotle. Only in mystical (and magical-occult) traditions we have theories on how to change things through mental operations alone and claims that it has been done. This is clearly not what Maturana has in mind.

On the other hand there seems to be a great problem in establishing the same kind of independent reality in the sub-atomic worlds of quantum mechanics (uncertainty principle, Bohr's theory of complementarity, EPR-paradox etc.). Although we have gained knowledge about some of the organizing patterns in the natural macro world, we seem forced to recognize that the knowledge we get in experiments -especially in the realm where the quantum effects are of significant importance - is not a picture of material mechanisms working independently of our doing. The concepts of particle, wave and field seem to have intermingled in such a way, that they obscure our previous vision of the elementary particles as the material building block of nature.

So it seems to be fruitful to accept a great deal of independent reality to "things" including living "things" and living languaging "things" as a part of the basic distinctions which are prerequisites of the human knowing on which science is built. The other prerequisite for science is languaging and thereby the existence of conscious beings as ourselves. So to accept that the scientific endeavour in fact gives us intersubjective knowledge of survival value, we have to accept that the languaging living "things" are conscious minds existing relatively independent of our own. As von Foerster (1980) points out this is a basic epistemological choice, but it is fundamental for the game called science. Maturana (1988) seems to recognize this.

But the question still remains with us: how can we catch anything "out there" of some general value if there are no mechanisms, no structures and no things to be uncovered, and if our concepts are purely social constructs? It is important to remember that this kind of question takes precedence of scientific theory of evolution and natural selection. There seems to be an unavoidable philosophical obligation to say at least something more of the world than to characterize it as a multiverse, if one wants to do a fair job in putting forth a view in philosophy of science.

This bring us to the second shortcoming of Bateson's attempt to build a theory of information and knowing. As with Maturana he has problems with his world view. But they are of another kind. The problem is his attempt to raise his theory on the foundation of a unification between thermodynamics and information theory.

Information and negative entropy

In "Mind and Nature" (1980 p.103) Bateson further develops his criteria for a cybernetic definition of mind:

"1.A mind is an aggregate of interacting parts or components.

2.The interaction between parts of mind is triggered by difference, and difference is a nonsubstantial phenomenon not loca-ted in space or time; difference is related to negentropy and entropy rather than to energy.

3.Mental process requires collateral ener-gy.

4.Mental process requires circular (or more complex) chains of determination.

5.In mental process, the effects of difference are to be regarded as transforms (i.e. coded versions) of events preceding them. The rules of such transformation must be comparatively stable (i.e. more stable than the content) but are themselves subject to transformation.

6.The description and classification of these processes of transformation disclose a hierarchy of logical types immanent in the phenomena."

These criteria are all famous within cybernetic understanding of mind, and I will not discuss them further here. My critique is concentrated on the foundation of the second criteria: "difference is related to negentropy and entropy..."

It is problematic, that Bateson is following Norbert Wiener's idea of a basic community between thermodynamics and Shannon and Weaver's theory of information based upon a shared starting point in the concept of entropy.

Regarding the problem of the relation between the concept "information" and the concept "negative entropy". Bateson writes (Ruesch & Bateson 1968, p. 177):

"Wiener argued that these two concepts are synonymous; and this statement, in the opinion of the writers, marks the greatest single shift in human thinking since the days of Plato and Aristotle, because it unites the natural and the social sciences and finally resolves the problems of teleology and the body-mind dichotomy which Occidental thought has inherited from classical Athens".

However, Shannon's theory of information has never had anything to do with the semantic content of messages. Shannon and Weaver (1969 p.31-32 ) write:

"The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. Frequently the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem. The significant aspect is that they are selected from a set of possible messages."

So, what people and animals conceive as information is something quite different from what Shannon and Weaver's theory of information is about. As von Foerster (1980 p. 20-21) concludes:

"However, when we look more closely at these theories, it becomes transparently clear that they are not really concerned with information but rather with signals and the reliable transmission of signals over unreliable channels ..."

In a conclusive analysis summing up many years of work with the concept of information in the physical sciences and information theory Voetmann Christiansen (1984) points out, that it is in fact a materialistic reductionism to claim that one's theory of information is based upon the physical concept of entropy:

"....in as much as the intentional aspect of entropy is its meaninglessness and uselessness. The measure for information which was introduced by C. Shannon and N. Wiener, among others, is also in the theory of information designated "entropy", as it formally is identical with the measure of entropy in statistical mechanics. One attempts in the theory of information to get out of the oddity of entropy in physics being a measure for missing information about the distribution of energy among the degrees of freedom by placing a minus sign in front of the entropy measure. "Information" is de-fined as "negative entropy" (negentropy): i.e. information theory's message to us can be summarized in the following manner: "You must not at first be interested in meaning, but you will learn to measure the meaningless in a precise way. This way one can always learn to understand meaning afterward by changing signs for meaninglessness".

(Voetmann Christiansen 1984)

According to Voetmann Christiansen, Bateson's theory would appear to end in a materialistic short-circuit. It is also well known that to determine the entropy in a system it is necessary in advance to determine what will count as macrostates. Further it is necessary to determine the probability of every state in advance. There is no room for the complete unexpected. But as Bateson's original definition goes, it is the observing system that determines which differences make a difference as it goes along in its historical drift. We are then back to the problem about which basic assumption we have to make about the world/reality and the relation between the observer and the world - including the theory of evolution - to "...find a way of putting the knower into a known that is constructed so as to keep the knower viable in practice" as Krippendorff has stated.

How can we create a viewpoint which is neither physicalistic nor an idealistic form of con-structivism? How can we make a - non-re-ductionistic - foundation for second-order cybernetics which is compatible with our knowledge in the sciences as well as in the humanities and in the social sciences, and do not fall into any of the known traps of our tradition of philosophy? I do think that Maturana is well under way. But I think it is possible to create a bit more clarity by analyzing the foundation of physics and especially its use of the concepts of entropy and chaos.

The problem of order and chance in physics.

The concept of entropy was introduced by Claudius around 1865 as an expression of the dissipation of energy as heat into the many degrees of freedom of the movements of the molecules in the surrounding space in such a way that it cannot be recovered again to perform the work from which the concept of energy was originally defined. In other words, a degradation of energy had occurred. Only the so-called free energy could be used to do productive work. Based on his reflections, Claudius then advanced the two famous dicta of thermodynamics:

1. The total amount of energy in the world is constant.

2. Entropy seeks a maximum.

In this way - in opposition to the mechanical idea of particles, whose movements are determined by deterministic, mathematically describ-able and time- reversible laws - a time direction and thus evolution in physics was introduced.

Soon, however, it turned out that there are divergent views on the epistemological status of thermodynamics and, consequently, the physical concept of entropy. Can thermodynamics be reduced, in principle, to classical mechanics? or is it a science - irreducible in principle - so fundamental that the inclusion of mechanics is an intrinsic part of it? The cardinal point in a century-long discussion is the question about the ontological status of the concept of randomness in physics. Thermodynamics seems to break fundamentally with the determinism and time reversibility of mechanics by being a basically statistical theory. Indeed in classical mechanics the direction of time is not a significant parameter, whereas "the arrow of time" is introduced in and with thermodynamics - if one accepts its fundamental status.

At precisely this point Prigogine and Stengers (1984) intervene with a reflective argument. They indicate that evolution, life, and therefore the conscious of mankind is not possible without the increasing entropy of irreversibility and thereby an increase of objective randomness. If one now insists that this randomness is not objective but only an expression of our limited knowledge - well, one is in reality insisting that our physical bodies are merely products of our own ignorance!

Notice how closely the mechanical paradigm is connected to the idea of objective materialistic information. Laplace's demon succinctly expres-ses the idea that the world consists of a mathematically-describable, completely exhaustible information. If it had knowledge of the position, speed, direction, mass and acceleration of all the "particles", it could compute all the events of the past and the course of the entire future. If this mechanistic world view were true, then reality would be full of objective information waiting "out there" independent of any observer. As Stonier (1990) for example thinks:

"In the present work, and it is crucial to the entire analysis, information is consi-dered to be distinct from the system which interprets, or some other way processes, such information." (p.18)

"Information exists. It does not need to be perceived to exists. It requires no intelligence to interpret it. It does not have to have meaning to exist. It exists." (p.21)

But in the further development of thermodynamics Boltzmann had to give up the idea of a basic simple objective order between the movements of particles to make comprehensible the thermodynamic evolution towards states of equilibrium. The idea is that it is impossible to initiate the description by noting every single position, speed, etc. when dealing with large populations (ensembles) of particles. It is therefore unrealistic to construct a classical mechanical calculation on this basis. Out of this molecular chaos one is on the other hand able to define a mathematical density function in phase space. The mathematical model which emerged from this work represented a considerable scientifically advance.

Boltzmann introduced probability as a basic concept in order to explain how systems consisting of large populations of particles eventually "settle" in stationary conditions in which the mixture of the elements is most random. This state is called thermodynamic equilibrium. It functions as a general attractor. In other words, if a liquid containing white and blue particles is mixed, the most probable state of rest is not blue particles in one side of the glass and white particles in the other, but a more random mixture with a light blue color. The system has "forgotten" its original state. Boltzmann's principle of order asserts that the system seeks the state in which a large number of events occur simultaneously - statistically speaking - cancelling each other. Entropy increases because probability increases!

The important point is, that Boltzmann's principle of order is based upon the assumption that molecules behave independently of each other before they mix and collide. This state is called "molecular chaos" and it was in fact a manifestation of the abandonment of Laplace's deter-minism and Leibniz' "prestabilized harmony", both based upon an idealistic assumption that the attainment of total knowledge of every single "particle" is possible.

The more general modern pragmatic view of physics sees classical deterministic mechanicism and the use of its paradigm in quantum mechanics as an idealistic view of the world which has given us some very useful models for calculation of certain types of systems. But it is not the basic model of "reality". As Prigogine points out we must accept randomness as a basic property in the physical world, and not just as a result of simple ignorance.

From this point of view Voetmann Christiansen (1970, 1975) shows the impossibility of establishing a materialistic concept of information by pointing out the impossibility of registering all the connections in a given field of measurements. In the thermodynamic chaos of the micro world there is actually an infinite and unattainable mass of potential information because the microscopic variables are so numerous (magnitude about 10 to the power of 23) that it must be supposed an impossibility in principle for man to achieve more than an infinitesimal fraction of the information necessary to plot the values which determine the Brownian movements of the molecules.

Voetmann Christiansen shows that the starting point for the concept of information must be phenomenological, but still within a realistic - but not mechanistic - world view. And on this basis Bateson's definition of information as a difference which makes a difference is still valid. Information is what you get in reply to a question. One defines the question's limits and area of measurement, and from these one can connect hypothetical probability values. One must then from this point work towards the sole form of objectivity that is given ordinary people: intersubjectivity. The chaos of thermodynamic microcosms can be characterized as objective ignorance in as much as the amount of energy and time which must be expended in the attempt to attain complete knowledge about it would create such massive amounts of entropy that the world would be considerably altered before the project was well under way.

But is this not a reflective trap? Won't our argument for a logical connection between chaos, irreversibility, time and evolution lead us to the claim that man as an evolutionary product is the result of this physics-determined chaos? This is still the most common way to comprehend the neo-Darwinist theory of evolution, which is basic to the science of biology. Somehow, scientists and philosophers imagine, life and consciousness emerge out of the physical universe. Non-equilibrium thermodynamics, dissipative structures and self-organization are evoked as explanations - also among many second order cyberneticians.

Von Foerster is well aware of the problem. He writes:

"... I propose to continue the use of the term "self-organizing system," whilst being aware of the fact that this term becomes meaningless, unless the system is in close contact with an environment, which possesses available energy and order, and with which our system is in a state of perpetual interaction, such that it somehow manages to "live" on the expenses of this environment."(von Foerster 1984 p.4)

Later in the same paper he goes on:

"... to show that there is some structure in our environment..., by pointing out that we are obviously not in the dreadful state of Boltzmann's "Heat Death". Hence presently still the entropy increases, which means that there must be some order - at least now - otherwise we could not loose it.

Let me briefly summarize the points I have made until now:

(1)By a self-organizing system I mean that part of a system that eats energy and order from its environment.

(2)There is a reality of the environment in a sense suggested by the acceptance of the principle of relativity.

(3)The environment has structure.

(von Foerster 1984 p.8)

We must accept some kind of structure of the world/reality. But physical structure alone won't do it.

It is not difficult to point out the inconsistence in attempting to account for human consciousness in terms of physical determinism (which we have already dismissed). One cannot be a determinist because it is a true philosophy, but only because one is predestined by the physical chain of causes and effects. But in this way these views loose their logical and truth dimensions which are a pre-requisite in the analysis of the views. Not only do they claim to be in accordance with the physical sciences, but they also claim to be logically valid and true. And that is something fundamentally different, which requires consciousness and knowing - and those were the things we set out to explain!

The questions of origin, the nature of reality and how to know it are logically necessary in their form but impossible to give an explicit and final symbolic answer. It seems always to end in self-circularity and paradox. This insight is basic to second-order cybernetics (see for instance von Foerster 1990). But by asking these questions we can delimit some of the premises for the possibility of raising these questions at all. A total paradox-free self-reflection does not seem to be possible in language without incorporating some kind of falsehood as is shown for example in Gödel's incompleteness theorem for mathematics. There will always be indecidables. In von Foerster's (1992) view this is where ethics and aesthetics come in.

We have some knowledge, and we know that we can obtain more. But we must admit that we do not have universal knowledge neither in the physical sciences nor in philosophy. Human knowledge is the meeting point between the subjective (autopoietic) and the objective (partly independent reality) and so it is relational and prone to mistakes. It is an ongoing process. It is human knowing. Information always has a historical index.

We strive for universal knowledge. At least it is part of the ideal of philosofy and science. But the fact is that our knowledge is always contextual, and therefore limited to a part of reality. We are not even able to give a simple description of the limits of the truth-content in our knowledge (models, theories) in any absolute theoretical way before the practical testing and the attempts of falsification. To use a modern image: the border between the areas in which a given model makes true and untrue statements is not a smooth curve but seems rather to be a fractal one. Science probes, it does not prove, says Bateson.

So when we try to generalize knowledge it is always prone to failure. This is intrinsic to what we call human knowing. But if we not able to recognize when we are mistaken, our knowledge could not grow. It is through our original ability to make distinctions in particular matters that we are able - by way of logic - to falsify our general models.

It is not only the so-called "outside" world that persists in surprising us with its complexity and spontaneity, it is also our so-called "inner" world, the "subconscious" complexity and spontaneity behind our acting in the world including languaging (as Maturana calls it to underline a process we are in rather than the conscious instrumental use of language). This basic incompleteness in our knowledge of our-selves, the reasons for our acting and lack of absolute conscious control over speech are a the same time the prerequisite for our ability to say something new and to cognize something new; to be in our basic flow of knowing and languaging. To realize that is not the end but the beginning of science, or rather of second-order-science.

But in relation to the indeterministic view one could furthermore ask how this world can exist at all and how scientific gathering of invariances is possible if the fundamental nature of reality is objective chance. If the world is governed by chance, how can any form of structure and law exist? The solution can only be found in a world view on a higher level, above determinism and indeterminism. One that does not contradict determinism in terms of the world of experience, but which turns it into a partial feature of a more extensive total point of view.

In the last part of this paper I will analyze such a candidate.

Living chaos as a basic concept in a theory of information.

A possible solution to this paradox is given by C. S. Peirce (1891,1892). Peirce points out that randomness or chaos must necessarily come before lawfulness and determination in an evolutionary philosophy! In accordance with the above-mentioned analysis by Voetmann Christiansen regarding information defined as negentropy, Peirce makes it clear that one cannot in a positive way base chance on the physical concept of laws - but only on a purely negative definition such as absence of law or absence of knowledge.

But Peirce (1892B) - in opposition to most modern scientists, among them Prigogine - then very clearly sees that his concept of chaos then cannot be limited to the mechanistic concepts of a dead and mechanical world. You cannot in advance - so to say - remove life and mind from the basic undifferentiated chaos from which worlds are initiated. The fundamental chaos is not the absence of law, it is the mother of law. It is not only empty, it is also full of possibilities. It is not only "dead", it is also full of life and mind, and then Peirce coins a concept to explain at the same time evolution and regularities of nature. He writes, that it has a tendency to be habit forming. Peirce therefore boldly calls it pure spontaneity! It is the spontaneity of living feeling.

One can determine law from chance, i.e., as 'habits'. "The laws of nature are the habits of the cosmos", says Peirce. The laws of nature we find are exact only in their mathematical descriptions, whereas the measurements on which they are based are always vitiated by uncertainties. The laws are only approximate model descriptions of a far richer and more varied, spontaneous and living reality. Peirce writes (1892):

"To undertake to account for anything by saying boldly that it is due to pure chance would indeed be futile. But this I do not do. I make use of chance chiefly to make room for a principle of generalization, or tendency to form habits, which I hold has produced all regularities. The mechanical philosopher leaves the whole specification of the world unaccounted for, which is pretty near as bad as boldly attribute it to chance. I attribute it altogether to chance it is true, but to chance in a form of spontaneity which is to some degree regular".

In order to impart meaning to this philosophy we must view chaos as spontaneously dynamic with a tendency to form habits. Symmetry breaking is a more modern scientific concept for the same phenomenon which is used in both quantum field physics and in thermodynamics. If we are going to accept chaos as a concept as fundamental as the concept of natural law, then we should not conceive of chaos as the absence of regularity or the absence of the ability to create structures. It should be viewed, rather, like all possibilities, as a hyper-complexity of potential structure and potential information in a living, infinite dynamics. It could thus be possible to transcend the dilemma between determinism and indeterminism.

Peirce's argument now is that if chaos is the fundamental concept then law is the unusual, unexpected thing, and thereby the thing to explain. It is not the other way around. From the point of view of the statistical information of Shannon and Weawer there is maximal information in chaotic random behavior. But not so with Peirce. Here it is the departures from the random which are interesting, which gives us information on structures and lawlike behavior (the habits of matter as Peirce calls them).

Talking of law coming from the random, Cosmos coming from Chaos, the habits of the universe coming into being, then the creation of the World and our own world melts together in a way; which the new cybernetics and constructivism has for a long time been on the track, for instance with Maturana and Varela's concept of autopoiesis.

As Peirce has noted, it is necessary to transcend the fruitless antagonism between idealism and materialism. Mechanism is what exorcises the spirit from nature and converts it into a machine. As soon as one perceives that deterministic mechanics has no ontological status, one sees that this presupposition about reality is not justifiable, and the way is paved for a more comprehensive view. About this, Peirce writes (1892 a, p. 335):

"On the other hand, by supposing the rigid exactitude of causation to yield, I care not how little it be but an strictly infinitesimal amount - we gain room to insert mind into our scheme, and put it in the place where it is needed, into the position which, as the sole self-intelligible thing, it is entitled to occupy, that of the fountain of existence; and in so doing we resolve the problem of the connection of soul and body".

The latter statement is perhaps a bit extravagant but it is true that we long ago had to accept that a stimulus-response theory of learning within the framework of a mechanistic ontology would not aid us significantly in understanding what knowledge is and how it is at all possible. Since Descartes, the dualistic philosophy of mechanism has driven the spirit from the world and the world from the spirit and, as a result, encountered the most fearful problems in attempting to reunite them through knowledge.

I think that Peirce provides us with a basic world view which can unite a form of realism with Maturana and Varela's biologically inspired concept of autopoiesis and the theory of the social construction of knowledge. Creating what I would call a conscious myth in that we explicate the prerequisites necessary to explain the abilities and knowledge we actually have. Peirce realizes, that such an ontology must include a view of the "creation of the world", which is not in conflict with our present scientific knowledge.

Peirce (1891, p. 170) describes a possible model of the creation of world and order in the following way:

"It would suppose that in the beginning, - infinitely remote, - there was a chaos of unpersonalized feeling, which being without connection or regularity would properly be without existence. This feeling, sporting here and there in pure arbitrariness, would have started the germ of a generalizing tendency. Its other sportings would be evanescent, but this would have a growing virtue. Thus, the tendency to habit would be started; and from this with the other principles of evolution all the regularities of the universe would be evolved. At any time, however, an element of pure chance survives and will remain until the world becomes an absolutely perfect, rational, and symmetrical system, in which mind is at last crystallized in the infinitely distant future".

This statement is very much in agrement with Einstein's theory of general relativity, where the Universe is seen as the unfolding of space-time. In modern physical cosmology - the Big Bang theory and the superstring theory - the universe is seen as arising from a random sporting from the vacuum field. It is very small in the beginning but rapidly it expands and thereby unfolds space-time. The difference is, that most modern physicists believe that chaos is non-living and non-mental. They have a physicalistic world view.

Peirce solution to Maturana's problem of the world's existence before and partly independent of the observer is a unique variation of the objective idealistic position, which is - as far as I can see - the only position that can support Maturana's philosophy that there is no system without an observer. Peirce (1891, see Buchler 1955 p.322) writes:

"The only intelligible theory of the universe is that of objective idealism, that matter is effete mind, inveterate habits becoming physical laws".

On p.321 he explains a little more:

"The law of habit exhibits a striking contrast to all physical laws in the character of its commands. A physical law is absolute. What it requires is an exact relation. Thus, a physical force introduces into a motion a component motion to be combined with the rest by the parallelogram of forces; but the component motion must actually take place exactly as required by the law of force. On the other hand, no exact conformity is required by the mental law. Nay, exact conformity would be in downright conflict with the law; since it would instantly crystallize thought and prevent all further formation of habit. The law of mind only makes a given feeling more likely to arise. It thus resembles the "non-conservative" forces of physics, such as viscosity and the like, which are due to statistical uniformities in the chance encounters of trillions of molecules.

The old dualistic notion of mind and matter, so prominent in Cartesianism, as two radically different kinds of substance, will hardly find defenders to-day. Rejecting this, we are driven to some form of hylopathy, otherwise called monism."

From this position he goes on to develop the theory into the realm of semiotics and knowing.

Voetmann Christiansen (1970, pp. 1-15) is on the same line of thought in his discussion of how one can meaningfully conceptualize the connection between the information-theoretical entropy concept H and the thermodynamic entropy concept S avoiding to base them upon the negatively determined concept of chaos of mechanics:

"H and S represent in reality exactly the same thing - that is, "missing" information, i.e., information not received but awaited when we have undertaken measurements. To avoid confusion over concepts such as negative information and negative entropy, it would be preferable to say the "missing information" means subconscious or preconscious information. Information is always positive, be it conscious or unconscious. The concept of "negentropy" is science's defense against the inclusion of consciousness in the description of nature..."

(Christiansen 1970, p. 1-15)

I believe that in this way it is possible to outline a consistent and meaningful foundation for Bateson's concept of information. It will not only make possible the inclusion of the entire range of Peirce's semiotic reflections, but it could also be reconciled with the important further developments of Bateson's modes of thought which have been made by Maturana and Varela.

Conclusions

I agree with Bateson (1973) that we have to start our understanding of information with the process of knowing. I think that Bateson's definition of information as a difference which makes a difference is very fruitful. His problem is that he nearly makes every cybernetic system a communicator and a knower.

Seen from this way of formulating the problem, the main achievement of the work of Maturana (1983, 1988) and Maturana and Varela (1987) is that they have conceptualized the basic requirements for a system to be living and knowing, in short to posses mind: it has to be autopoietic.

They underline an important aspect of Bateson's theory of information when they stress that the autopoietic system does not receive information but only perturbations of its organization. The so-called reaction is internal adjustments to preserve the internal organization of the system. It underlines the distinction between the world and its being of differences - potential information if you like - and information processing of the autopoietic systems.

Maturana's theory of autopoiesis and knowing is clearly not physicalistic. But on the other hand he says so little of the world and the problems of independent structures in it, that it almost seems as if the observers create the physical world totally by themselves, although this is not his view. There still seems to be a philosophical gab in the theoretical foundations of second-order-cybernetics, or at least its for-mulations, with which cyberneticians has to deal.

In his creation of a basic semiotic Peirce did some very fundamental thinking about the necessary relationship between the interpreter, the sign and the object and the minimum qualities he had to ascribe to them to make a realistic model of the process of knowing and sign making. I have not focused on his doctrine of sign in itself in this paper. My main focus has been on its epistemological and ontological foundations.

In accordance with modern thermodynamics - but not built on it - Peirce sees the basic quality of reality as randomness or chaos. He then clearly sees, that if chaos is basic then you cannot explain it as the absence of law, because chance or randomness is before law. So you have to explain law from randomness, not the other way around. Chaos, chance, randomness must then be seen not only as emptiness but also as fullness, i.e. as a hypercomplex dynamic process, which includes the characteristics of mind, matter and life. He calls it pure spontaneity, and living feeling, somewhat mystically inspired.

To explain how law and structure comes from randomness Peirce finds it necessary to endow chaos with one more quality, namely the tendency to form habits. Evolution of order - emergence - demands a projection of the quality of the tendency to form habits into the world substratum. In this minimum statement he avoids saying too much about a virtual order in the transcendental and on the other hand he avoids denying such an order. Furthermore he does not start with an elimination of mind, living and knowing from reality. Thereby he avoids to have to reinvent them later on by "changing signs for meaninglessness". In this way he creates a fruitful non-reductionistic philosophical foundation (world view) for the theories of Bateson and Maturana, and more generally speaking for second-order cybernetics.

We cannot say that the world we live in has no structures, on the other hand we cannot say that our process of knowing has no influence on these structures. We cannot say that the world is basically logical and determinate or that it is absolute irrational and chaotic. We cannot say that reality is basically simple (and logic) or that it is too complex to be even partly understood. We cannot say that reality is basically "dead" material or that it is basically "pure spirit". Our theory of knowledge/of knowing must be based somewhere between these positions. I therefore think that Peirce's theory of basic reality as a hypercomplexity of living feeling with a tendency to form habits, is a good and fruitful supplement (or vice versa) to Bateson's and Maturana's theories of information, communication, knowing and languaging. I see here a fruitful theoretical connection between second order cybernetics and semiotics which will give both theories more strength. How that is to be done in specific areas and practices must be dealt with in further development within these areas in an exchange between the general theory and the specific area or practice.

References

Bateson, G. (1973): "Steps to an ecology of Mind", Paladin. Frogmore, St. Albans, USA.

Bateson, G. (1980): Mind and Nature: A necessary Unit, Bantam Books, New york.

Beer, C. G. (1982):" The Study of Vertebrate Communication - Its Cognitive Implications" in Griffin, D.R. (ed.): Animal Mind - Human Mind, Dahlem Konferencen 1982, Berlin, pp. 251-267.

Bittermann, M.E. (1965): "The Evolution of Intelligence", Scientific American january 1965.

Brier, S. (1980): Der ønskes analyseret (evt. v.h.a. egne undersøgelser) om heirarki- og sandsynslighedsbetragtninger i beskrivelsen af adfærd kan anvendes i - og udbygge - én eller flere motivationspsykologiske modeller" Prizewinning essay (gold medal) on concepts and models of motivation in behavioral science, Psychology department, University of Copenhagen.

Buckler, J. (ed.)(1955): Philosophical Writings of Peirce, Dover Publication,New York.

Foerster, H. von (1979): "The Cybernetics of Cybernetics" i Krippendorff, K. (ed.): Communication and Control in Society, Gordon and Breach Science Publishers, New York, pp. 5-8.

Foerster, H. von (1980): "Epistemology of Communication" in Woodward, K. (1980): "The myths of Information", Madison, Wisconsin.

Foerster, H. von (1984): "Observing Systems", The Systems Inquiry Series. Intersystems Publications. California, USA.

Foerster, H. von (1990): "Ethics and second-order cybernetics". Cybernetics & Human Knowing, vol.1, no.1, 1992, Aalborg, Denmark.

Gardner, H. (1985) : The Minds new Science, Basic Books, New York

Hinde, R.A. (1970): Animal Behaviour: A Synthesis of Ethology and Comparative Psychology, Second edition, International Student Edition, McGrawhill Kogakusha, Ltd. Tokyo, London.

Kuhn, T.S. (1970): "The structure of scientific revolutions", 2nd enlarged ed. Univ. of Chicago press.

Glasersfeld, E. von (1991): "Distinguishing the Observer: An Attempt at Interpreting Maturana", Methodologia vol. V, no.8, p. 57-68.

Glasersfeld, E. von (1992): "Why I consider myself a Cybernetician", Cybernetics & Hu-man Knowing, Vol.1, No.1.

Heims, S.J. (1991): The Cybernetic Group, The MIT Press, Cambridge, Boston, USA

Hofstadter, D.R. (1983): "Artificial Intelligence: Subcognition as Computation" in Machlup, F. & Mansfield, U. (eds.): The Study of Information, Wiley, New York, s. 263-285.

Jensen, U. J. (1973): Videnskabsteori I & II, Berlingske Leksikon Bibliotek, København.

Lindsay, P. & Norman, D.A. (1977): Human Information Processing: An Introduction to Psychology", 2.end edition, Hartcourt Brace Jovanovich, San Diego.

Lorenz, K. (1976): Bagsiden af spejlet, J.H.Schultz, København (orig. Die Rückseite des Speigels. R.Pieper & co. Verlag, München 1973)

Machlup, F.(1983): "Semantic Quirks in Stu-dies of information" in Machlup,F. & Mansfield,U.:"The study of information" p. 641-671. John Wiley & Sons, New York.

Madsen, K.B (1974): "Modern theories of motivation" Copenhagen: Munksgaard

Madsen, K.B. (1978): "Motivationsbegrebets udvikling i psykologien", Pædagogik nr.2.

Maturana, H. (1983): "What is it to see?" Arch. Biol. Med. Exp. 16, pp.255-269.

Maturana, H.R. (1988): "Ontology of Observing: The Biological Foundation of Self Consciousness and the Physical Domain of Existence." in Donaldson, R.E. (ed.)(1988): Conference workbook for "Texts in Cybernetic Theory". An In-Depth Exploration of the Thought of Humberto R. Maturana, William T. Powers, Ernst von Glasersfeld. A Conference of The American Society for Cybernetics, October 18-23, Felton, California, pp.4-52.

Maturana, H. (1990): Personal Communication at the Conference of The American Society for Cybernetics in Oslo.

Maturana, H & Varela, F. (1980): "Autopoiesis and Cognition: The realization of the Living", Reidel, London.

Maturana, H & Varela, F. (1986): "Tree of knowledge: Biological Roots of Human Understanding". Shambhala Publishers., London.(Danish version from 1987 used).

Peirce, C.S. (1891): "The Architectures of Theories", The Monist, Vol. 1, No. 2, jan. 1891.

Peirce, C.S. (1892A): "The Doctrine of Necessity Examined", The Monist, Vol. II, No. 3, april 1892.

Peirce, C.S. (1892B): "The Law of Mind", The Monist, Vol. II, No. 4, july 1892.

Peirce, C.S. (1893): "Evolutionary Love", The Monist, Vol. II, No. 4, p. 533.

Petersen, A. Friemuth (1972): "Personlighed og tænkning. En teoretisk biologisk analyse". Psykologisk Skriftserie nr. 2, Psykologisk Laboratorium, University of Copenhagen.

Popper, K.R. (1972): "Objective Knowledge: An evolutionary Approach", Oxford At The Claredon Press, Glasgow, Great Britain.

Popper, K.R. & Eccles, J.C. (1977): "The Self and Its Brain", Springer International, New York.

Prigogine, I. (1980): "From Being to Becoming", W.H. Freemann and Co., San Francisco, USA.

Prigogine, I & Stengers, I. (1984): Order out of Chaos, Bantam Books, New York

Reventlow, I. (1970): "Studier af komplicerede psykobiologiske fænomener"; Munksgaard, Copenhagen. Doctoral thesis. University of Copenhagen.

Ruesch, J. & Bateson, G. (1987): "Communication", New York, W.W. Norton & Company, pp.314,(original 1967).

Searle, J. (1989): Minds, Brains and Science, Penguin.

Shannon, C.E. & Weawer, W. (1969): "The Mathematical Theory of Communication", The University of Illinois Press, Urbana, Chicago, London, pp.125.

Stonier, T. (1990): Information and the Internal Structure of the Universe, Springer Verlag, Berlin, London, pp. 155.

Thorpe, V.H. (1979): The origin and Rise of Ethology: The Science of the Natural Behaviour of Animals, Heinemann, London.

Tinbergen, N. (1968): "On War and Piece in Animals and Man: An Ethological Approach to the Biology of Aggression", Science, vol.160, pp. 1411-1418.

Tinbergen, N. (1973): "Ethology" in Tinbergen, N.: The Animal in Its World, London, pp. 136-196.

Varela, F.J. (1984): "The Ages of Heinz von Foerster" in Foerster, H. von (1984).

Vickery, A. & Vickery, B.(1988): Information Science - Theory and Practice. Bowker-Saur, London.

Winograd, T. & Flores, F. (1987): "Understanding Computers and Cognition", Alex Publishing Corporation, Norwood, New Jersey.

Voetmann Christiansen, P. (1970): Information, entropi og udvikling, kompendium, H.C.Ørsteds institut, Københavns universitet.

Voetmann Christiansen, P. (1984): Informationens Elendighed, synopsis to a workshop on the Information society, IMFUFA, Roskilde University Center.

Voetmann Christiansen, P. (1988): "Charles S. Peirce: Mursten og mørtel til en metafysik". (Translation of and introduction to five articles from the Monist, 1891-93) Texts from IMFUFA, Roskilde University Center, Roskilde, Denmark).

Wiener, N. (1961): Cybernetics or control and communication in the animal and the machine, The M.I.T. Press and John Wiley & Sons, New York, sec. ed. (original 1948).

Notes

  1. I want to thank Fred Steier, Old Dominion University, Virginia, Ib Ravn, Munksgaard Publishers, Lars Qvortrup, Odense University, Stig Andur Pedersen, Roskilde University and Jesper Hoffmeyer, Copenhagen University for fruitful comments on earlier versions of this paper.
  2. Maturana and Varela seldom use the concept philosophy. Maturana (1988) insist that what he does is science. But in the usual way of using the word what they do is philosophy. There is both an epistemology (and within that a philosophy of science), a world view and an ethics in their writings.
  3. There is no logical necessity here between biology and philosophy. The broadening of the theory is the result of what Maturana - when asked - calls historical drift (Maturana 1990). It is with thought as with language, that we have no final causal control over where it will lead us. But still in their writings their is a lot of arguing from a biological scientific level of thought to a more general philosophical level (see Maturana 1988).
  4. Thanks to Hugh Gash for a helpful critique of my viewpoints on Maturana's work.
  5. "The notion of the amount of information attaches itself very naturally to a classical notion in statistical mechanics. That of entropy. Just as the amount of information in a system is a measure of its degree of organization, so the entropy of a system is a measure of its degree of disorganization; and the one is simply the negative of the other." (Wiener 1961 p.11)
  6. This might be true on a spiritual level. But the present claim is in the context of objective science.
  7. I am using the concept "particle" as a general theoretical concept encompassing molecules, atoms and sub-atomic "particles" (and which in reality is used also "upward" in the scientific hierarchy when one limits the units as passive and independent in stimulus-response psychology and in sociology, among others). The essential point is that the discussion is relevant on all levels, especially physics and chemistry.
  8. We do not deal with Prigogine's controversial theory of micro-entropy in this context.


Return to the content of this issue

Return to the Cybernetics and Human Knowing Homepage


The Web edition of Cybernetics and Human Knowing is edited by Jan B. Steffensen and Søren Brier
Rev. 26.03.1997