Code-Duality and the Semiotics of Nature

by

JESPER HOFFMEYER and CLAUS EMMECHE


The final version of the paper is published pp. 117-166 in: Myrdene Anderson and Floyd Merrell (eds.):On Semiotic Modeling. Mouton de Gruyter, Berlin and New York, 1991.

The sphere of Information

Biological information is not a substance

The biological discipline morphology derives its name from the greek word 'morph'. According to the etymological dictionary the romans probably took over this word from the greek, but in a distorted way. Thus, In latin 'morph' became 'form'. From this latin word arose the verb informare : To bring something into form. And this again is the root of the now fashionable word information.

In spite of this likely etymological relationship between morpholology and information the two areas occupy nearly antagonistic positions in modern biology: While the study of the anatomical forms of organisms (morphology), is now a rather outmoded discipline, the study of 'biological information' (in DNA or RNA) is one of the most flowering specialties of present biology. Clearly, in this century 'information' has substituted 'form' as the preferred key to the fundamental problems of biology. It should be realized, however, that the old dilemma of form and substance is not at all explained away through the introduction of this new concept information. The dilemma merely assumes a new disguise.

Reading contemporary texts on biology - at whatever level it may be - one cannot escape feeling, that the very notion of biological information serves to circumvent difficulties which are in fact not solved. Thus, we are told, for instance, that in the fertilized egg of any animal is carried 'a programme' specifying how this particular animal should be formed. Although nobody claims detailed knowledge as to how this programme actually works, it is understood that its creation is no serious mystery. Natural selection acting on the endless succession of minor changes in the DNA-sequence (i.e. mutations) will do. The details remains to be worked out, of course, but the main lines are firmly established.

Essentially this explanation amounts to a claim, that the actual forms of the organisms in this world can be explained through the functionality of the genetic 'programmes' carried in the DNA of these organisms. At first sight this explanation may look sound enough. But the problem circumvented through this kind of reasoning is the following: What is the relation between DNA and the 'programme' ? How does this scheme take into account the eventual logic of the programme itself? And, above all, who is the subject to whom this programme makes sense?

Thus, the metaphor of a programme or of information serves to shift the attention from the actual life-forms to the more invisible and badly understood level of latent life-forms which, one assumes, is somehow 'coded' in the DNA of the fertilized egg. And here the whole dilemma of form and substance suddenly disppears. DNA (a substance) and programme (information/or potential form) is treated as one and the same thing. The evolutionary flexibility of DNA is seen as the basis for the evolutionary flexibility of the genetic programme. This identification of substance and form at the lowest level explains the otherwise rather astonishing easiness by which neodarwinism derives the form of biological structures from their function.

But biological information is not identical to genes or to DNA (any more than the the words on this page are identical to the printers ink visible to the eye of the reader). Information, whether biological or cultural, is not a part of the world of substance. It nevertheless depends on that world since it has to do with the pattern of substance, the way substance is organized or formed.

We suspect that the confusion of form and substance brought about through this rather ill-considered use of the concept information lies at the root of fundamental problems of present biology. And it is the aim of this paper to show, that if the concept of information is not used to conceal the distinction between form and substance, it might be possible to change the central perspective of biology in a fruitful way. The dynamics of biological information belongs to another level of analysis than the dynamics of DNA-molecules. Biological information is expressed through signs and should be studied as such, i.e. as a special case of semiotics, which we shall term semiotics of nature.

The study of living systems from a linguistic or semiotic point of view has attracted considerable attention in recent years. We discuss this development in a separate paper (Emmeche and Hoffmeyer 1988). In the present paper we are going to explore some new paths in the semiotic universe of nature.

The concept of information

Considering the enormous social, economic and scientific importance presently ascribed to the concept of information, it is a striking fact that very few people would be able to tell the meaning of this concept. And furthermore, that those who claimed to know the meaning would probably not agree at all.

The modern use of the word information clearly reflects the atomization of knowledge which has been a scientific ideal through the last hundred years. Thus, in older times information did not in general refer to the kind of detached fragments of knowledge called information today. Rather a person might recieve information in the way of being educated. And not only was information seen as a more broad concept, but also as something connected to the process of being informed. Information and person merged into each other. In other words information had not yet aquired the subject-free status, which it has in everyday talk of today.

We suspect, that this gradual process by which information was first deprived of its personal or subjective anchoring and then atomized into still more detached messages (finally ending up in units of 'bits'), reflects the alarming indifference to epistemological problems, which has characterized most of the scientific culture for the last hundred years.

It may be illuminating to remember the latin root of the word information: informare. Clearly, to bring something into form is not transmission of information. I.e. while the blacksmith obviously brings the steel into definite forms, one would never say that he informs the steel. Rather the blacksmkith, himself, has to be informed if he shall succeed in bringing the steel into the desired form. Information belongs to the mental sphere, not to the physical sphere of substance and action.

Correspondingly, forms themselves were not, in the antiquity, seen as belonging to the physical sphere. Forms were induced on, not derived from substance. They reflected human or Godly will. Thus, to bring something into form presupposed a person in whom the idea of the form must first have occurred. And this occurence, the idea, was the root of information.

In information theory this mental or subjective dimension of the concept of information has totally disappeared. Development of information theory was from the beginning connected to the technical aim of improving the existing systems for telecommunication. Central to this endeavour were concerns for increasing the capacity for transmission of signals through a channel. Thus, the question of how information was created or what should be meant by the significance of information was not adressed through the theory. By neglecting such questions a quantitative mathematical theory of information was constructed, which:

'...provides a measure for how much information is to be associated with a given state of affairs and, in turn, a measure for how much of this information is transmitted to, and thus available at, other points. The theory is purely quantitative. It deals with amounts of information - not, except indirectly and by implication, with the information that comes in those amounts' (Dretske 1981)
The basis for this quantification of information is probability theory. In fact, the 'information' of information theory simply is a quantitative measure of improbability. The higher the probabiliity of a given event, the less information does it convey. Intuitively this is in accordance with our everyday notion of information. A trustworthy oracle stating, e.g. that in 1994 the president of the United States will be a white male, would not make much of information. But if the oracle had pointed to a black woman as the president of the United States, that would have been a surprise - and, in fact, the information content of this message would be considered high.

Mathematically the concept of information may be expressed in units of 'bits':

'In general, the amount of information in bits (binary digits required to indicate the selection) conveyed by the correct transmission and reception of a given message state is equal to the number of times the initial probability of that state must be doubled to reach 100 per cent.' (Sayre 1967: 6).
The price to be payed for this quantification is a loss of semantic content. In information theory the value of information only reflects the statistical structure. But most statements in human communication are only understandable at a semantic level of analysis. Evidently, the 'information' of the mathematical theory is a far less comprehensive category, than the information exchanged between people in talking. Referring to the work of among others Shannon and Weaver (1949) and Brillouin (1968) von Foerster writes:
'...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 (this in itself is a formidable task and as we know, fabulous things have been achieved in this domain)' (Foerster 1980).
von Foerster suggests that this confusion of 'signal' and 'information' did arise because information theory was developed in the historical context of world war II. During wartime 'a particular mode of language - the imperative, or command - tends to predominate over others (the descriptive, the interogative, the exclamatory, etc.)'

Whether this historical explanation suffices or not need not concern us here. The important point in von Foersters argument is that the quantitative concept of information needs a closed possibility space. If the set of possibilities is open, one cannot ascribe precise probabilities to any single possibility and thus no information value. This requirement of a closed possibility space, of course, is perfectly fulfilled in the case of military commands. But apart from such extreme situations it would hardly ever be fulfilled in human communication:

'To recapitulate: a system in which commands function smoothly is one in which information and signal are indistinguishable from each other. This is the behaviorist ideal. The system is threatened the moment someone behaves not as he "ought" to, but as he might wish to, thereby creating a climate in which the "new" might be born' (Foerster 1980).
Actually, the mathematical theory of information collapses in the moment unpredictability enters the scene - as it necessarily does in real life. Improbability, and thus according to the theory information, presupposes the ability to foresee the possible occurences. To state that an event is improbable, one first has to know that it might occur at all. Therefore the totally unforeseen - and thus the real new - cannot be acounted for through the statistical theory of probability. (We are grateful to Peer Hull Kristensen for drawing our attention to this line of argument (Kristensen 1984)).

We are led to conclude that the objectivization of the concept of information has been obtained at the cost of depriving the concept of most of its explanatory power concerning real life situations of human communication. When this poor concept of information is imported to biology, the result is a corresponding trivialization of the dynamics of living systems.

Differences, which make a difference

'There are in the mind no objects or events - no pigs, no coconut palms, and no mothers. The mind contains only transforms, percepts, images, etc....It is nonsense to say that a man was frightened by a lion, because a lion is not an idea. The man makes an idea of the lion' (Bateson 1972: 271).
According to Gregory Bateson information is based on difference. A sensory end organ is a comparator, a device which responds to difference. While reading this, for instance, your eyes do not respond to the ink, but to the multiple differences between the ink and the paper.

The number of potential differences in our surroundings, however, is infinite. Therefore, for differences to become information they must first be selected by some kind of 'mind', the recipient system. Information, then, is difference which makes a difference (to that mind):

'Try to descibe a leaf or, still better, try to describe the difference between to leaves of the same plant, or between the second and the third walking appendages (the "leg") of a single, particular crab. You will discover that that which you must specify is everywhere in the leaf or in the crab's leg. It will be, in fact, impossible to decide upon any general statement that will be a premise to all the details, and utterly impossible to deal with the details one by one' (Bateson and Bateson 1987:164).
What enters the mind as information always depend on a selection, and this selection is mostly unconcious. In this sense one should not speak about 'getting' information, rather information is something we 'create'.

Furthermore, only a tiny fraction of the actual differences exhibited by the phenomena under study will be mapped on to our description. Thus, every time we make a description most of the differences which might be potential information, are not selected. In other words, every time we create knowledge, we also - and by necessity - create non-knowledge. (Nørretranders 1982, Hoffmeyer 1984). To make things visible, we make other things - or in a certain sense the same things - invisible. This creation of non-knowledge, which by necessity accompanies any process of investigation, is in itself a legitimate reason for the very widespread uneasiness towards the scientifc project. Might it not be, that the ecological and economic crisis now facing earth has its roots in the mud of our collective non-knowledge? Certainly, myth (Prometheus) and religion (the tree of knowledge) have known all the time the dangers inherent to the project of seeking knowledge.

Now, in a heavily overpopulated world, totally dependent for its basal needs on technology, knowledge cannot be condemned. Only, one should never forget the eliminated differences, the non-knowledge. And this precisely is the reason why epistemology needs concern us. Our theories are the instruments by which we conciously or unconsciously select the differences, which make a difference. And thus the instruments behind our creation of knowledge as well as non-knowledge. Only by understanding the processes by which we create knowledge can we hope to adjust this building of knowledge, science, so that it becomes a true servant - and not an overmighty, untouchable truth hiding an ocean of non-knowledge.

It remains a subject for metaphysical speculation, whether the world is such, that some objective number might be given for the sum total of differences it contains. At the quantum level even a raindrop would exhaust the computer capacity of the whole world if a complete enumeration of the potential differences it contains should be calculated and memorized. For differences to be information we humans have got no choice but to select. And therefore information does not belong to the sphere of matter and energy, but to the subjective and non-dimensional sphere of structure, pattern and form.

It follows from this concept of information, that information is not an exclusively human phenomenon. Rather, our world is replete with systems sufficiently sophisticated so as to respond to selected differences: Not only animals and plants, but also organs, tissues and single cells are vehicles for information processing. They are communicating systems.

At the most fundamental level the distinction between life and not-life is dependent on this ability: the response to differences. In the world of nonliving matter which is described by the laws of physics and chemistry the ability to respond to differences play no part. A stone is affected by 'forces' and 'impacts', but not by differences.

On the contrary, nothing in the world of living systems makes sense unless we include in our explanations this peculiar ability to respond to selected differences in the surroundings. Of course, living systems too are material, physical systems affected by forces and impacts. They must obey the laws of physics and chemistry. But these are not sufficient for explanation. Living systems always exhibit certain organizational characteristics, which enable them to react to differences in the surroundings, and thus to 'create' and exchange information.

The sphere of information inhabits the lifeless world of energy and matter - is dependent on it. And to the best of our knowledge it has grown out of this inanimate world. Yet, it is not itself material. Forces and impacts does not affect it, unless they somewhere occasion a difference which makes a difference to some kind of living system.

We do not want to exclude the view, however, that even the lifeless sphere of nature might be engaged in informational activity. After all, our universe seems strangely inhomogenous and irreversible. And if this was not the case creative processes and life would not have been possible. But then, the spatiotemporal scale of such eventual cosmic communication would be outside the range of interest to the study of the semiotics of living systems.

Information and the dualism of mind and body

Are we condemned then to adopt once more a fundamental dualism of the universe: The dualism of information and substance ? And this dualism, is it not just a new disguise for the old dualism af mind and body, which since the time of Descartes has tended to envelope the kinds of knowledge obtainable in western science ?

To answer this question one should note, that the relation between information and substance is quite different from the relation between mind and body. While information and substance are seperable concepts in our descriptions the two are inseparable in the 'real' world. They are different aspects of the same world. Information does not exist except as carried by (immanent in) matter and energy.

Actually, the advantage of the semiotic perspective precisely is this: To transcend the Cartesian analytic and reductionist 'either-or' (Anderson et al. 1984). Through the vision of code-duality it becomes possible, as we shall see, to integrate substance and information into a unitary relationship unfolding at different levels of complexity. '...the web of sign relations is at all times informational and energetic, spatial and temporal, objective and subjective' (Anderson et al. 1984).

Mind and body, on the other hand, are seen as separate entities in the 'real' world. In fact, the unbridgeability of the dichotomy between mind and body is necessary exactly because they are different only in substance. Since they are both thought of as kind of substances, the only way to keep them separate is to assure their fundamental (and metaphysical) foreignness towards each other.

Thus, while the mind-body dualism necessarily implies, that the human being remains alien to the universe in which he or she is placed, the distinction between information and substance only helps us to see our deep relatedness to the other living systems of this planet. And the real problem rather becomes that of the interface between the living and the nonliving sphere of nature. How could systems capable of responding to differences in their surroundings arise in the middle of that which knows no difference ?

Code-duality: Life and culture

Defining the subject: Redescription

To answer this question one has to be aware, that the notion of a system already implies some kind of distinction between what belongs and what does not belong to the system. We, of course, can make such a distinction, but we were not there at the time of the beginning of life - and neither was (per definition) any other kind of living system capable of selecting differences or distinctions in their surroundings. In a certain sense prebiotic systems did not exist - only three and a half billion years later we draw their outlines in our imagination. This of course might be said of everything pre-human. Our point is, however, that living systems did in fact exist (not only materially but as organized entities) before human minds 'created' their outlines.

Suppose that eventually a living system arose from the primordial soup - or whereever it was. Then we will have to ask: Who was the subject to whom the differences worked on by such a system should make a difference? If one admits at all, that living systems are information processing entities, then the only possible answer to this question is: The system itself is the subject. Therefore a living system must 'exist' for itself, and in this sense it is more than an imaginary invention of ours: For a system to be living, it must create itself, i.e. it must contain the distinctions necessary for its own identification as a system. Self-reference is the fundament on which life evolves, the most basal requirement. (This does not pertain to non-living systems: There is no reason for the hydrological cycle to know itself. Thus, rivers run downstream due to gravity, water evaporates due to the solar heat, nowhere does the system depend on self-recognition).

Another way to express this whole matter is to say that differences are not intelligible in the absence of a purpose. If nothing matters matter is everything.

But what is the basis of this self-reference, and thus the basis of life? We shall suggest here that the central feature of living systems allowing for self-reference, and thus the ability to select and respond to differences in their surroundings, is code-duality, i.e. the ability of a system to represent itself in two different codes, one digital and one analog (Hoffmeyer 1987). Symbolically this code-duality may be represented through the relation between the egg and the hen.

Self-reference clearly depends on some kind of redescription. The system must be able to construct a description of itself (Pattee 1972, 1977). This description furthermore must stay inactive in - or at least protected from - the life-process of the system, or else the description will change - and ultimately die with the system. In other words, the function of this description is to assure the identity of the system through time: The memory of the system. In all known living systems this description is made in the digital code of DNA (or RNA) and is eventually contributed to the germ cells.

We suggest, that it is by no means accidental that the code for memory is of a digital type. What should be specified through the memorized description is not the material details of the system, but only its structural relations in space and time. If such abstract specifications should be expressed through an analog code only very simple systems would be possible, and those would probably not survive.

(For a parallel: if human communication and memory depended solely on analog codes, e.g. the ability to mime, if in other words we did not posses the digital code of language, our cultural memory would be as short as that of chimpanzees - and the social structure accordingly simple).

Now, for the system to work the memorized description in the digital code must be translated to the physical 'reality' of the actual living system. For this translation (the developmental process) to take place the fertilized egg cell, or some equivalent cell, must be able to decipher the DNA-code as well as to follow its instructions in a given way. This need for the participation of cellular structure shows us that a sort of 'tacit knowledge' is present in the egg cell {Polanyi 1958, Pattee 1977). And the existence of this tacit knowledge hidden in the cellular organization must be presupposed in the DNA-description. Thus, the digital redescription is far from a total description.

The realization in space and time of the structural relations specified in the digital code defines what kind of differences in the surroundings the system will actually select and respond to. Through this realization a new phase is initiated, the phase of active life. One might say that in this phase, the 'analog phase', the message of the memory is expressed.

Eventually the system will survive long enough to pass on its own copy of the digitalized memory (or part of it) to a 'new generation'. Essentially this corresponds to a back-translation of the message to the digital form. But this is a process which takes on its significance only when seen at the level of the population. The population (rather than the single organism) passes on a message about conditions of life to the memory (the gene-pool). The population should be considered a code expressing a message. This code, however, necessarily is analog, since it has to interfere with the physical surroundings and thus must share with these surroundings the properties of physical extension and continuity.

The chain of events, which sets life apart from non-life, i.e. the unending chain of responses to selected differences, thus needs at least two codes: one code for action (behaviour) and one code for memory - the first of these codes necessarily must be analog, and the second very probably must be digital.

There is a certain similarity, of course, between the concept of code-duality and the 'replicator'/'interactor' distinction of David Hull (Hull 1980). The terminology of Hull, however, tends to repeat the 'thingish' conception of information whereas we, by using the term 'code', want to stress the sign-character of living systems.

Returning to the question of how living systems, i.e. systems capable of responding to differences in their surroundings, have originated in the lifeless world of matter and energy, we can now reformulate it as the question of how code-duality could arise. And this essentially is the question of how form became sign..

Language: The digital code of human culture

When the question of prebiotic evolution is posed in this form, it can hardly escape notice that it contains a striking similarity to the question of human evolution, which is often identified as the question of how language evolved. To the extent language depends on the exchange of discontinous signs, words, with an arbitrary relation between expression and content, language is, in fact, based on a digital code. Thus, culture may be seen as built on the code-duality of digital language and analog 'reality'. Through language the actual life of human society - its 'reality' - is translated onto a code for common memory, language, protected from the sphere of activity, i.e. the sphere of social life. Even a nuclear explosion will not directly change the english grammar or the meaning of any sentence expressed by human beings. It will change such things only because it changes - or kills - the minds of human beings.

On the other hand, language does not itself directly interfere with social life, it does so only through the intermediate link of human actors. If language did interfere directly with social life - as is actually supposed in certain forms of witchcraft - it would, in fact, be very embarrassing in most life situations. If for instance telling that president Reagan is a pig meant that he actually became a pig, then one had to be very careful about telling things.

These characteristics of language nicely corresponds to the role of the genetic information of DNA. The rejection of socalled 'lamarckism', precisely amounts to the claim, that no events in the life of an individual will change the content of the genetic information carried by that individual. Aquired characters are not inherited. On the other hand, genetic information does not directly interfere with the game of survival, but does so only through the intermediate link of active individuals.

Needless to say, the concept of a cultural code-duality depends on the acceptance of human 'reality' as just another message, i.e. as a kind of information. This view does not imply that human beings are not of blood and flesh. But it implies, that human bodily action - from the simple rhythm of breathing to complicated affairs like that of playing the Goldberg variations of Johann Sebastan Bach or climbing Mount Everest are always of significance. Life processes are embedded in meaning. They always communicate at least the deep message of the human condition and most of the time they communicate the slightly less deep meaning of the social condition as well.

Besides, blood and flesh after all are words. As such they do of course - or so we assume - refer to real distinctions (differences) in the world. Only, those distinctions were selected by human minds among billions of other (abstractly) possible distinctions. To the extent human 'reality' is understandable it necessarily is information, for as Bateson told us 'there are in the mind no coconuts ...'.

Human life is translated forth and back In an endless stream between the digital code of words and the analog code of behaviour. To name one of those codes 'reality' and the other 'description' or 'explanation' seems pointless or even untimely. Rather 'reality' should be reserved for this process of coding and recoding itself, thus surpassing the unbearable dichotomy of materialism and idealism (Hoffmeyer 1987).

Speculating on the origin of life, theorists even today very often invoke miracle - or its modern representative in the form of 'extraterrestrial' germs. Jacques Monod, while of course rejecting the idea of miracle, considered the event of creation of life so utterly improbable that its actual occurence might be taken as proof that human beings are unique and literally lonesome creations in a foreign and unparticipating univers (Monod 1971).

Concerning the existence of the human being itself, arguments might very well point to a similar improbability, as certainly they did in the past. In our century, however, belief in organic evolution apparently has convinced most people that human evolution must have taken place through processes not dissimilar to those invoked for the explanation of the creation of the rest of the creatures of this world.

Seen from the point of view presented here, the origin of life and the origin of the human being, are two unique events sharing the same essential step: The formation of code-duality through the 'invention' of a digital redescription allowing for identity through memory. These two steps, nevertheless, belong to different levels of complexity, and their parallellism should be understood only in a formal sense.

Supposing that human evolution took place through natural processes, we are thus led to believe, that the origin of life is likewise explainable through such processes. And we shall suggest, furthermore, that the problem of the interface between life and non-life be studied through inspiration from the much better known interface between the human and the non-human, the cultural and the natural.

Origin of human digitality: The aesthetic mind

Not being linguists, we shall make no attempt to enter the longstanding debate concerning the eventual possesion of latent linguistic abilities by our remote relatives among primates. We shall be content, rather, to observe that whatever language chimpanzees might posses, they apparently don't (mis)use it to discus the eventual possesion of language among humans. Human language clearly belongs to a qualitatively more sofisticated linguistic universe than that of the other primates. And, on the other hand, if one does not adhere to belief in miracle, human language must have arisen through evolution, and it would be very surprising therefore if nothing like language was to be found among our primate relatives.

From the point of view advanced above the important point is the degree of digitalization involved in primate communication. The distinction between analog and digital codes is certainly not a simple one, as is illustrated for instance by the hieroglyphs. While a single hieroglyph may be taken as an analog representation, it becomes digital when presented in written text. Likewise a painting can be seen as an analog representation of some internal reality in the painter. If the painting is part of an exhibition, another message appears - that of the exhibition. In the context of this message the single painting suddenly becomes a detatched and arbitrary sign. It has become part of a formalism - a digital code.

Thus, the distinction between analog and digital codes depends on context. Using the terminology of Gregory Bateson the single picture and the picture exhibition are of different 'logical type' (Bateson 1972, 1979). In a way it is only in the context of the higher logical type, the exhibition, that the painting becomes a painting at all. Or, if exhibitions didn't exist paintings would be something very different. The exhibition digitalizes the work of the artist and in the same time conserves it in the form of paintings. Through this operation the painting obtains the freedom of disengagement: it can now be freely combined in collections, histories of art, or in furnitures. But this gain of freedom is payed through a loss of meaning - which is why, we guess, artist often don't like to sell their paintings. As Anthony Wilden puts it:

'The analog is pregnant with meaning whereas the digital domain of signification is, relatively speaking, somewhat barren. It is almost impossible to translate the rich semantics of the analog into any digital form for communication to another organism' (Wilden 1980: 163).
But:

'...what the analog gains in semantics it loses in syntactics, and what the digital gains in syntactics it loses in semantics. Thus, it is that because the analog does not posses the syntax necessary to say "No" or to say something involving "not" one can refuse or reject in the analog, but one cannot deny or negate' (Wilden 1980: 163).
The trick of digitalization consists in introducing gaps into continuums, thereby creating boundaries. These boundaries, however, do not themselves belong to the continuum, neither are they part of the gap. The boundary is the locus of an external intervention and thus necessarily defines a goalseeking system that drew that boundary. Therefore a system of a higher logical type - defining the goal - is necessarily established in the process of digitalization.

That chimpanzees do in fact master processes of logical typing seems indisputable. The story of the chimpanzee Washoe moving around in a forbidden area of the garden while trying not to be seen is impressive: Believing she was alone Washoe made the sign 'quiet' (in the american sign language), apparently in order to remind herself of the need not to be discovered. Obviously she was able to conceptualize the context of her hiding as a 'quiet-situation' (and maybe this creation of a higher logical type was the real fun of it).

Whether chimpanzee use of words (in American Sign Language or the like) should be considered a kind of linguistic capacity or not, the experiments do at least prove the ability of chimpanzees to use a digital code when trained to. And yet, neither chimpanzees nor any other animal make use of digital communication in the wild condition (with the song of the whales as a possible exception).

In human development apparently the child's first semantic system is a holophrastic 'sentence dictionary' in which words correspond to complete sentences. Interestingly Bronowski (1967: 385) points out that 'the normal unit of animal communication, even among primates, is a whole message'. The ultimate solution for the memory load that such a dictionary entails is of course a word dictionary, a dictionary in which the discrete units are 'subtracted' from the continuum of the various sentences (Wilden 1980): 'The original dictionary is a dictionary of messages; it is replaced by a dictionary of the code'

Speculating on how this ability for digital communication arose in the hominid line one should remember that the essence of the digitalization is the drawing of boundaries. But this, as we said, requires an intervention, or a decision as to how the boundary should be drawn. 'A gestalt' write Wilden 'is formed by the decision to digitalize a specific difference, so as to form a distinction between figure and ground. This is in effect a decision - which may be neural, or conscious, or unconscious, or habitual, or learned, or novel - to introduce a particular boundary into an analog continuum' (Wilden 1980: 174).

Now, for the digitalization to be functional in communication this decision must be part of a shared pattern among the communicating persons. Therefore for digital communication to arise in a population of hominids they must necessarily construct the image of their surroundings, their 'umwelt' (J. v. Uexküll 1931), in a formalized way allowing for a shared patterning. Digital communication requires the establishment of a highly organized social frame capable of sustaining a formalized, or maybe ritualized, mental sphere, common to all members of the social group.

If, in fact, a decisive prerequisite for the evolution of language is the construction of a common digital representation of the 'umwelt', then the real stumbleblock would seem to be the creation of a shared mental sphere of sufficient sophistication. While chimpanzees apparently did have the brains for simple digital communication, they probably did not have any need for this shared mental sphere, and so they did not develop a language.

These considerations strongly support the hypothesis originally put forward by Glynn Isaac and further developed by Richard Leakey and Roger Lewin (1978), that the creation of a 'food-sharing economy' was the essential step, which marked off the line of our hominid ancestors from the apes. The essential element in this sharing-economy is that both sexes participates in the procuring of food, each contributing a different component to the common meal (respectively meat and vegetables).

According to Leakey and Lewin the mental requirements for the successful acomplishement of such a pact would be far more of a challenge than the mental requirements for the construction of advanced stone tools. Thus, in modern hunter-collecter cultures the intellectual level is only little reflected in the technical equipment. By far the most important skill consists in the ability to make in the brains an effective mental map of the surroundings, and a map one should remind in time. Hunter collecters have to know where and when to search for a whole range of food items. It would seem therefore that from the moment the hominids began sharing their meals, there would have been an enormous survival-premium on the ability to communicate such mental maps, and thus on the construction of a socially shared mental pattern.

For several million years our ancestors have created simple stone tools. Although a gradual refinement of these stone tools can be documented,

'it is not until perhaps one and a half million years ago, that a distinctly new dimension of technology arrives. This comes in the form of the beautifully tear-drop shaped handaxe that is the hallmark of the Acheulian culture. The conceptualization involved in manufacturing these impressive symmetrical objects is of a different order from that involved in making simple core tools with sharp edges' (Leakey and Lewin 1979)
But: 'The point is that there is virtually no job that can be done with the implements in the well-developed Acheulian tool-kit that cannot equally well be performed using the essential random but effective products of enthusiastically crashing two stones together' (ibid )

This, of course, raises the question of why formalized and ordered technology did emerge at all. The answer that Leakey and Lewin offer to this question turns upside down our traditional understanding of the relation between the development of technology and intelligence. Such a technology, they claim:

'reflected an increasingly formalized and ordered social structure, a structure that demanded the facility for a sophisticated language. As rules and customs developed for controlling the social and practical organizational problems of operating a successful gathering and hunting economy, those rules and customs impinged on the elaboration of material technology too' (ibid ).
Thus, the stone tool manufacture became more formalized: 'not because economics demanded it, but because that was the way our ancesters' minds were working: structure was stamped on social dynamics and on material technology ' (ibid , our italics).

At what time our ancestors actually aquried the possession of real language remains speculation of course. But it seems reasonable to assume, that the engagement in the production of aesthetic products would appear at a time when a formalized mental pattern had settled itself in the population as an autonomous ressource shared by all. And at this time all preconditions for digital communication would seem to be present.

The Acheulian period might very well mark the beginning of this important phase of pre-human history: The phase of digital communication and thus of cultural code-duality. As shall be further discussed in the next section the establishment of code-duality creates an enormous dynamics in a system. In fact, code-duality is probably a necessary condition for evolution (cultural as well as biological). And following the creation of code-duality the acquisition of real language would henceforth seem an inevitable next step.

Code-duality in evolving systems: Life and Culture

Before returning to the question of the origin of code-duality in prebiotic systems let us now compare the way code-duality works in living systems and in cultural systems. What is it that makes the possesion of digital coded versions of the world such a powerful instrument?

The simplest way to approach an answer to this question might be to look at the most clearcut example of the cultural application of digital codes: The written text (Hoffmeyer 1987). In other words:

What is the advantage of books?

One possible answer would be: The freedom from the constraints of the physical. No natural law restricts the possibility-space of a written (or spoken) text. Anything can happen. Napoleon could have lunch with Meryl Streep, or Socrates could drive a Rolls Royce. Salamanders could rule the world, or babies could be born with green hair instead of a mouth so as to nourish by photosynthesis. But nothing of this could possibly happen in the real world.

Thus, the trick of digital codes in the human sphere is to translate 'reality' to a form in which its content can be freely manipulated so as to test the universe of possibilities.

Another answer would be: Books are objects. They may be cut into pieces, fragments can freely be circulated (copied) or combined with fragments from other texts (e.g. in education). Thus, the digitalization of human experience in the form of written text makes selections of such experience accessible to all members of culture (although, of course, something is necessarily lost in the process of digitalization). Also, books being objects (and not subjects) don't die. People may forget them or burn them, but eventually they are kept for centuries at libraries. Thus nobody would have known anything about Sokrates, had it not been for Platon writing his dialogues. Digital codes work as cultural glue through space and time.

Finally, and most important perhaps, language (verbal or written) can speak about itself, it can meta-communicate (e.g. 'this sentence is in English' is a sentence about a sentence). Therefore messages in digital codes can belong to levels of different logical typing. For instance, animals may or may not be able to pretend, but 'only humans pretend to pretend' (Wilden 1980: 173).

In order for this to be fruitful in normal life-situations the insights gained by these mental manipulations must be translatable back to the analog code of active life, that is change in the 'real' world. This in fact is the way of evolution in the cultural sphere: an unending chain of coding and recoding, translation, between digital and analog form, language and 'reality'.

Now, turning to the sphere of organic life, we find that the evolution of the living systems is based on a very similar chain of coding.

Thus, the parallel in organic life to the writing of books would be sexual reproduction. Through this process the actual experiences of a population is transmitted to the next generation in the form of a changed gene pool. By converting the experiences to the digital form of DNA (in sperm and egg), it becomes possible to combine the information in new ways, thus testing the universe of possible relations between individual and environment. Again, this requires a digital phase (one can not combine half a hen and half a cock and thereby obtain a chicken!). Without the combinatorial power of the digital DNA-code, biological evolution would not have been possible.

An essential element in this process also is the exchange of DNA fragments due to crossing over or to the work of transposable elements. Because of these processes the gene pool becomes a kind of populational glue forcing the population to develop as a unit. Through positional mutations or chromosomal rearrangements new signification is created. For instance, among the chimpanzee and the human about 1% of the DNA-sequence differs (King and Wilson 1975). Clearly what differs is not primarily the text, but the context.

This brings us to the final characteristic of the digital code, that of meta-communication. Context corresponds to messages about messages, i.e. about how to read or understand other messages. (Being a message, however, context itself needs context, which in biology brings us to the deeper problems of 'goal'. But this question needs a separate treatment) In biology context (at the proximal level) probably denotes relations in time and/or space between different parts of the embryo or between the embryo and the maternal organism in which it develops. How such messages are organized in the DNA is for the moment totally unknown. And we suspect, it will remain so, until the present confusion of information and substance has been transcended in biology. Genes of course are important for understanding the text of DNA, as words are for literature. But to understand, what genes (or words) mean, semantics and the hierarchy of context must be consulted. To day not even the syntax of DNA is proberly understood - if the existence of such a phenomenon is at all recognized.

While the translation from analog to digital form, i.e. from population to gene-pool (through sexual reproduction), is a process which should be understood at the level of the population, the translation back to analog form (epigenesis), concerns the realization in time and space of individual life. Through 'reading the DNA-text' the fertilized egg (or an equivalent cell) learns how to construct itself in a manner, which served the survival of its parents.

Morphogenesis takes us back to the world of natural constraints and physical form. One should notice, however, thatform has no location, it is, like concepts such as 'frequency', 'contrast', or 'correspondence', a non-dimensional phenomenon. Form concerns relations but it is nowhere. As Bateson explains:

'The contrast between this white paper and that black coffee is not somewhere between the paper and the coffe...Nor is that contrast located between the two objects and my eye. It is not even in my head; or, if it be, then it must be also in your head. But you, the reader have not seen the the paper and the coffee to which I was referring...' (Bateson 1972: 408).
This problem of localization is not a trivial one. To solve it, the cellular (analog) structure of the fertilized egg must necessarily participate. (Løvtrup 1974, 1987, Ho and Saunders 1979). As we noted above, the 'tacit knowledge' of the cellular apparatus is not itself a part of the DNA-redescription. This, in fact, parallels the problem of using techical knowledge for the construction of actual machinery. Although the knowledge is present at libraries nobody would expect the library itself to be able to construct a machine. It takes people to convert the knowledge of books to actual constructions. Again, digital codes do not interfere directly with the game of life.

The problem of localization involves a kind of regulatory anticipation of intricate relations between parts of the embryo (e.g. morphogenetic fields) and, in mammals, of relations between the embryo and the maternal organism. The metaphor of DNA as a 'computer-program' directing this process, has been criticized because the egg-cell must be understood as an active and autonomous participant in the establishment of morphogenetic fields (Goodwin 1983, 1984, Emmeche and Hoffmeyer 1988). However, the computer-program metaphor might approach reality, when the next generation of computers are launched, since - as we are told - the programmes of these computers are no longer deterministic. They allow for a real interaction between program and process. In any case, the translation of the discrete sequential in-formation in DNA to analog form, continuous in space and time, is certainly an extraordinary complex affair. The simplistic notion of genes as carriers of 'objective information' detached from contextual bindings, seems utterly insufficient.

The semiotics of nature

Paradigm

Through centuries biological theories have been molded to conform to the view of nature established in classical physics. An apparently infinite succession of deep-rooted controversies bear witness to the fact, that this was not at all an easy fit. Vitalism, teleology or finalism have perpetually been called upon to account for living systems. But the authority of physics was such, that in the end those deviations from the ideal was always defeated - to reappear, nevertheless, in new disguise in the next generation.

With the birth of molecular biology and especially molecular genetics in the fifties and sixties a strange thing happened. Suddenly a new and very foreign vocabulary was introduced into biology, that of cybernetics or information theory. Terms like 'program', 'genetic code', 'information', 'messenger-RNA', 'feedback' and the like became respectable or even indispensable notions. Such terms, however, clearly played no role in the world view of classical physics.

This contradiction disappears when it is recognized that these new terms did not mean the same thing in biology as they did in general language. Facilitated by a widespread indifference to epistemological problems among biologists the concept of genetic information became for all practical purposes identified as the sum of the genes which carried it. Thus, the equation:

INFORMATION = GENES

imperceptibly slid into the unquestioned set of shared preconceptions forming the paradigm of biological theory. Since genes were considered solid physical entities, 'pieces of DNA', i.e. kind of particles, no contradiction to the physical foundation of biology was thought to follow from the new cybernetic vocabulary.

We highly suspect the fruitfulness of this paradigm. If the complexities of dynamic (evolutionary) biological systems should be grasped, we will have to abandon the naive confusion of information and substance. Genes must be understood as signs, not as particles or 'pieces of DNA'. And as such they must not be treated as physical units in the sense of calories or kilogrammes. Furthermore, genes are not the only kind of biological information, since, as we noted, the whole of life is information exchange, communication.

What we propose then, is that the traditional paradigm of biology be substituted by a semiotic paradigm the core of which is that biological form is understood primarily as sign (whether analog or digital). Signs, of course, always require a physical medium such as ink or voices. But after all, we don't purchase a book because of the printers ink in it. Maybe it is time we don't ask for genes or DNA, when we want information.

Before adressing the question of this new paradigm in more detail, we wil have to discuss more deeply the meaning we ascribe to the word 'sign', and how a world of matter and energy could give birth to such alien things as signs. How could a 'world of physics' create a 'world of biology'?

A molecular sharing economy

Clearly, if the Newtonian model of nature was the whole truth of our universe, then a biology based on 'signs' would be inconsistent with 'truth' and, in fact, only miracle could explain the existence of organic life. As has shown Ilya Prigogine, the Newtonian model of classical physics is build to understand the world as being , it is unable to understand the world as becoming. (Prigogine 1980). In spite of the darwinian effort, organic evolution remains a contradiction inside a world constituted among other things by a concept of reversible time . For all we know as biologists evolution definitely is irreversible, and how could, logically seen, a reversible universe create irreversibility?

Working on physical systems far from equilibrium Prigogine showed that in such systems, which he terms dissipative structures, irreversible change does actually take place in non-deterministic ways. An infinitisimal disturbance of the system may amplify itself and cause the macroscopic system to change and this change virtually takes the form of a 'choice' between several possible states. By such processes order is persistently created out of chaos.

The claim of Prigogine is, that most parts of our universe, including living systems, do not fulfil the conditions for which the idealized model af classical physics is applicable. In these parts of the universe quantitative laws do not apply. Rather, since time must be counted as a real asymmetry of our universe deterministic prediction is impossible. The creation of real newness implies that only historical explanation can be offered.

We must leave it to phycisist to decide on the fate of this new view of physical nature. Seen from the biological perspective it has the advantage of offering an alternative to miracle as the explanation for the creation of organic life forms. If even the non-living universe needs historical explanation, then it becomes much less of a mystery how organic evolution, the history of nature, could take place. Eventually, then, the contradiction between biology and physics has come to an end. And the question of prebiotic evolution would seem to challenge only our informed imagination, not our general world view.

While the creation of order out of chaos is certainly a prerequisite for prebiotic evolution, the possibility of this process does not itself explain how the essential step for the creation of living systems, the establishment of code-duality, actually took place. We of course can not hope to solve such a problem. We would like to suggest, however, that the solution be searched for through considerations parallel to those which in our view account for the creation of code-duality in the evolutionary line of hominids.

In this event, as we saw, the decisive step was not the development of brain complexity itself, since chimpanzees are able to pass the test for complexity and yet do not use digital communication in the wild condition. Rather the decisive step was the introduction of a social need (due to the creation of a sharing-economy), which ultimately was best satisfied through the creation of an autonomous supra-individual mental sphere shared in by all individuals. Thereby was guaranteed a common formalized 'map of reality' on which digital communication could be based.

Turning to prebiotic evolution, what would be a comparable step in the prebiotic world? Evidently, we should not expect mere complexity to suffice. Rather one should figure out some prarallel to 'social need', which might account for the establishment of a new level of cooperation between subsystems. Such cooperation should depend on some kind of analog proto-communication, which might gain in effectivity through a process of gradual formalization. The decisive step would be, that the formalization reached a level, such that it became an autonomous factor in further development - when, in other words, no region of the system could escape the dictate to share in it or to perish.

To illustrate these abstract considerations let us offer just one sketchy and very speculative scenario.

Consider that a loose aggregate of peptide complexes was present at some specific locality. Some regions of such an aggregate might well be able to collect small pieces of RNA on their surfaces, and these RNAs perhaps might stabilize other parts of the aggregate. Suppose, furthermore, that several such aggregates shared the same locality. Inevitably small sequences of RNA would diffuse all over the locality spreading a kind of analog latent information referring to other regions of the system.

Now, the first decisive step would be the creation of an interdependence between the aggregates such that their individual survival/persistence depended on each other. If for instance the accidental difusion of RNA fragments between the aggregates in some way contributed to the stability of the whole system (thus defining a system). Considering the now well established catalytic properties of small RNA molecules (North 1987) this creation of a higher level organization seems plausible. For instance an interaction between different aggregates might be brought about by a reciprocal emission of RNA molecules with catalytic effect on certain regions of other aggregates. If only some systematic difference between the base-frequencies of RNAs produced at different regions is allowed, then this would correspond to a kind of 'molecular sharing economy'. If to this scheme we add the creation of selective sites or channels for selective diffusion, then a simple kind of analog communication is in fact established.

In that case a premium (effectivity in stabilization) would be obtained through an increasing formalization of the RNA-production, i.e. through a gradual restriction on the range of sequences produced at the surfaces of different regions as well as restrictions on the channels of diffusion or the specificity of recognition sites.

The important second step (se scheme 1) In this process would be the separation between functional molecules (e.g. peptide-complexes) and informational molecules (RNA). This specialization is essential to the creation of a digital code which, in order to be protected from the dynamic processes of the system, should preferably be non-functional in a physical sense. (This is why the voice was a good candidate for evolution of digital communication among hominids, and also why it is very unlikely that the neanderthals spoke with their hands as suggested in a recent bestseller (Auel 1980)). These developments of organization in the system, we suppose, would all be consistent with the dynamic properties of dissipative structures as suggested by Prigogine.

It seems likely that such a system might reach a level of formalization such that every region of the system would have to share the formalization. A very simple digital code would then be present in the form of an autonomous RNA-sphere. Signs in this code, would probably in the beginning be whole RNA-molecules - but effectivity would be likely to increase through gradual subtraction of shorter sequences from the continuum of the whole molecule.

At this stage of RNA-communication one might say that something parallel to a simple spoken language is present in the system. The real code-duality would not be established, we think, until this language of RNA had been supplemented by a code for memory through the introduction of a separate sphere of DNA. This later step formally corresponds to the important step taken in cultural evolution, when spoken language was translated to the more independent and timeless code of written language.

The biological sign relations

Now, the question remains: In what sense would the formalized RNA-molecules or sequences of our scenario have become signs and not only organized forms of nature? To answer this question we must first develop the fundamental sign-relations of biological systems.

According to Charles Sanders Peirce a sign is a triadic relation, i.e. it is a relation between three and only three parts (Buchler 1955). Thus, for something to be a sign (the primary sign) it must refer to something else (the signified object). And underlying this reference must be some 'convention' or 'key' (the interpretant). Graphically this triadic relation may be depictet as in figure 1.

Applying this understanding of the sign to biological entities we have to confront the problem of how to reconcile the triadic relation of the sign to the dyadic relation of code-cuality. Code-duality, as we saw, involves the interaction of two translational processes. One from digital to analog, i.e. epigenesis, and one from analog to digital, i.e. sexual reproduction. But, as was also noted, the two processes do not belong to the same organizational level. Epigenesis takes place at the level of each single organism while sexual reproduction takes on its semiotic significance only at the level of the evolving population, or the lineage.

Actually, both of these processes may be understood as interpretative acts, and, as we shall see, their combination makes clear the semiotic character of organic evolution.

In the epigenetic process DNA, i.e. the genome, may be seen as just one fragment of an evolutionary stream of signs passed down through the generations. The interpretant selecting such signs among the myriad of internal cytoplasmic differences is the fertilized egg, the zygote. The fertilized egg, the real 'person' of biology, can make sense of the elaborate message contained in the DNA, using it to master the epigenetic process, i.e. the construction of the phenotype, the actual organism. This gives us the sign-relation shown in figure 2.

Actually, the competence to 'read the DNA' is conserved through several cell divisions, but for some reason the daughter-cells become less and less 'prophetic'. Gradually they lose the ability to 'understand' the full message of the genome, each daughter-cell considering only its own selected sections. Still, it is the zygote which is the subject in the process: It initiates the deciphering of the DNA-message and becomes gradually changed to the embryo in response to the interpretation.

Even in species without sexual reproduction the same fundamental semiotic relation between the DNA and the organism holds true. Only, in such species, the competence for interpretation of the DNA-message is apparently not irreversibly lost through differentiation.

In nature - not to be confused with culture - the individual is the tool of the fertilized egg. Expressed in the metaphorics of language, the zygote 'reads' the 'book' in its DNA, 'interprets' its meaning as a kind of 'manual' for the construction of a tool for survival, the individual organism. With the help from this tool, the egg cell can hope to continue its cell-line for yet another generation on the condition, of course, that the tool is sufficiently well-made to survive and reproduce in its ecological niche.

Implied in this view is a very important but widely overlooked fact: The DNA does not specify the zygote, the zygote must be there beforehand. The digital description contained in the DNA does not specify the material details of the system to be constructed (i.e. the organism) but only the structural relations in space and time. Analog information built into the cellular organization of the zygote is necessary for the interpretation in terms of continuous extension in space and time of the sequential DNA message to take place. Nothing digital in the world can function by itself. (But then: may be nothing analogical in the world can evolve by itself).

While the organism must necessarily die, the zygote, itself, may eventually survive provided that sexual reproduction takes place. An unbroken line of cells connect the zygote to the germ cells and thus to the continued chain of life. To support this chain of life seems to be the main purpose of the analog coded version of the message, the organism.

By constructing an organism the fertilized egg also constructs a specific 'umwelt' (J.v.Uexküll 1931), which constrains the interaction of the organism with its environment. For instance, the night moth can only perceive sounds corresponding to a very limited range of frequencies. In practice, night moths hear nothing but their specific enemy, the bat. No other sounds are part of their 'umwelt': The rest of the world is silent for them (J. v. Uexküll 1982 [1940]) Or, in the terminology of this paper: The rest is differences, which make no difference to the night moth.

Now, the manifold organisms belonging to a shared locality may appear in the 'umwelts' of each other and thereby is created a sphere of purely analog communication. This may be seen as a horizontal semiotic process interphering with the vertical semiotic process connected to the digital code of the genetic system. Because of this horizontal communication the organism - and thus the zygote - becomes part of ecological entities. Likewise, because of the vertical communication the organism, and the zygote, becomes part of genealogical entities. Thus, the distinction of horizontal and vertical semiosis correspond nicely to the distinction of Eldredge and Salthe between ecological and genealogical hierarchies (Eldredge and Salthe 1984). The interaction of horizontal and vertical semiosis connect the two hierarchies at the level of the zygote-organism.

At the ecological level new and unpredictable signification may emerge, i.e. the 'umwelt' of organisms may be tuned to each other in new ways. An illustrative example is the case of barn owls imported into the rubber plantations of Malaysia. In a few generations these barn owls completely changed their social life. In Europe barn owls have one clutch per year, rigidly defend their territories, frequent open ground, and quarter that ground in search for prey. In the Malaysian plantations the barn owls have two or three clutches per year, do not rigidly defend their territories, and perch to wait for the prey - the abundant rats in the plantations. Juveniles unable to find nesting sites congregate at the end of the day's hunting - social behavior not seen elsewhere. (Lenton 1983, cit. from Taylor 1987).

Also in this connection should be mentioned the phenomenon for which J. v. Uexküll coined the term contrapuntal relationship, e.g. the mutual correspondence of a birds wing and the air, of a fin and the water, and also of the sun and the eye (J. v. Uexküll 1982 [1940]). From the point of view of this paper these 'contrapuntal relationships' are especially illustrative cases of the analog coded messages, which makes up organisms.

Most important, males and females of the same species may appear as sexual cues in the 'umwelt' of organisms. And through sexual reproduction certain patterns of horizontal communication may be differentially strengthened in succeding generations. This is the way in which the evolving unit, the lineage, integrates horizontal semiosis into its own vertical semiotic process.

Thus, the lineage may be seen as an interpretant, incorporating into its digital repertoire the learnings of its analog phase. The ecological niche, i.e. the horizontal communication among the individual organisms, does contain a message about the present conditions of life. This message has no meaning at the level of the single 'umwelt', but for the evolving unit, the lineage, it refers to the genetic composition, the DNA, which should eventually be changed. The pattern of sexual reproduction may be seen, then, as an interpretative act, brought about as a response to a sign, the ecological niche (figure 3).

Peirce was a realist, i.e. signs ultimately reflected activity in real life. According to T. L. Short the mature Peirce denied his earlier nominalist belief that interpretants of signs were themselves signs in an unending regression. Thus, he admitted that habits or habitchanges can be the 'ultimate' interpretants of some signs (Short 1986). The change of genetic organization in response to conditions of life may be seen as a change of habit in the lineage.

In this view the lineage is a 'semi-transcendental' autonomous historical subject. It interprets the conditions of life, it does not only reflect them. Biologically speaking, the phylogenetic history of the lineage, as conserved in the dynamics of its actual 'ontogenetic trajectory' (Alberch et al. 1979; Wiley 1981, Salthe 1988 a), establishes a frame inside which any translation process necessarily must be contained. The lineage may be perturbated (by a changing milieu) but it responds to such perturbations according to its own (epigenetic or historically approbriated) rules.

This is where the neodarwinistic perspective seems to be inadequately narrow. In the traditional view natural selection is seen as a kind of rational mirroring of changing environments on to the gene-pool (if it was not rational, no adaptation would follow). Neodarwinism has no need for interpretation. But why should living creatures be those complicated informational systems and yet exhibit such a lack of independence? In our opinion neodarwinism gravely understimate the autonomy and inventivenes inherent to semiotic systems exhibiting code-duality.

Now, combining the two triads we get figure 4.

Here, it must be admitted, DNA is an ambiguous term since it denotes both genome and gene-pool (better 'genomorph', see below). Actually, the lower triad should be seen as just one representative of a multitude of triads corresponding to the total set of genomes.

While every genome is a sign to the zygote of what should be done to survive one cycle more, the total genomic organization is needed to signify to the 'population', how survival in the ecological niche is assured. To be precise, however, the word population is wrong. Populations do not interpret DNA-signs, only zygotes do that. What we need is a term denoting the sum total of developing zygotes inside a population. In danish we might call it a kimflok, i.e. a flock of germs (Hoffmeyer 1987). This kimflok rather than single fertilized eggs should be considered the real subject of the living world.

According to Peirce, the interpretant need not be there for something to be a sign: 'It is not necessary that the interpretant should actually exist. A being in futuro will suffice' (cit. from Short 1986). Thus, a mutation, even if 'mute', might be considered a new sign, if only it will make sense in later cycles. But wether it will make sense or not, does not depend on the environment as such, but on what the lineage can make out of it, given the environment. As interpretant, the lineage should be considered a truly inventive actor on the scene of organic evolution.

Now, this informational view of evolution should not,of course, be used to hide the fact that evolution is actually a process in space and time. To stress the processual character of evolution it might be preferable to use arrows to represent the physical relations between the semiotic elements. Furthermore, to make clear the fundamental role of code-duality in this kind of semiosis, the two triads might be contracted to a tetrade as in figure 5:

In figure 5 the arrows do not belong to the same organizational level. The figure presumes a shift from the level of the individual (left part) to the level of the population (right part). The reason for this is buried in the dynamics of code-duality. The digital coded period of life is the 'objective' or shared phase of life. The function of this period exactly is to sustain a higher level organization, that of the population. The analog coded period of life is the subjective or individual phase. The function of this phase is expansive activity, i.e. the actual job of life, survival and reproduction.

Also, as pointed out by Stanley Salthe, through these relations a symmetry appears between the ontogenetic trajectory (left part) and the lineage (right part). Thus, in a certain sense the ontogenetic trajectory constructs phenotypes in accordance with rational matching to the umwelt just as the lineage constructs gene pools that represent niche meanings (Salthe 1988 b).

The individual and the zygote

The biological sign-relations developed above implies a view of the individual which may seem to violate deeply rooted humanistic values. To avoid unfruitful misunderstanding on this important aspect of the semiotics of nature it may be worthwhile, therefore, to compare the view of the individual as a tool for the fertilized egg to other commonly held conceptions of the biological role of the individual.

One very fashionable conception among biologists today is that of the individual as a survival machine for the selfish genes (Dawkins 1976). In our opinion this utterly reductionist view grants to the digital code far too much independence. The opposite may be said of the common sense conception according to which the individual is unique and should not therefore be considered a tool for anything but himself or herself. Everyone is the architect of his own fortune, as it goes. From a biological point of view this individualism would seem rather ungrounded. Individuals after all are mortal and without sexual reproduction they would not exist. Thus, individualism - biologically seen - tends to overestimate the independence of the analog phase of active life and underestimate the significance of the species history as reflected in the digital code of the common gene-pool. Both 'DNAism' and 'individualism' tend to blind us for the importance of the code-duality, i.e. the subtleties of the translation processes between analog and digital code.

Actually, the view of the individual as tool for the germ cells is not foreign at all. Through most of human history the generation and not the individual was concieved as the important unit. The feeling of kinship was as powerful as is individualism in modern times, and this feeling persists in non-industrialized countries even today. Biologically seen the fertilized egg is the representative of the generation. Even more: in it is carried an unbroken line of cells going back to the beginning of life. For millions or even billions of years germ cells have gone on dividing themselves and reuniting in an unbroken line of cells, through time modifying or improving their internal library, the DNA, so as to construct still more fantastic tools, organisms, for their own survival. That some such tool eventually came to know its own mortality is the special gift or burden left over to us through human evolution.

A simple return to the old feudal feeling of a higher relationship with the family may not be desirable in a world looking so different. But a deep feeling of relationship with the rest of the life-forms on earth clearly is. Considering the power of present holistic conceptions which tend to reduce people to the humble and anonymous bearers of a higher meaning, Gaia or whatever that is called, one should be anxious to refine the scientific thinking on the role of the individual. The holistic conceptions may be understood as a protest against a scientific image of the world rendering too much importance to individuals. Rather than oposing this scientific image with the image of Gaia, we would prefer to confront individualism with the image of the fertilized egg as the primary creation on earth.

However, we do certainly not concieve human culture as a mere bi-product of germ cells. The semiotics of nature should not be confused with the semiotics of human culture. But every human being is deeply involved in both kinds of semiosis. Let us forget none of them.

From RNA to Sign

We can now consider the answer to the question of the sign-character of small RNA molecules in prebiotic evolution. How would these RNA molecules fit into the sign-relation?

The simple situation of reciprocal exchange of catalytical RNA may be illustrated in figure 6:

Clearly as depicted in figure 6, the cycle might be understood as purely physical. Nevertheless, the effect of RNAs on aggregates would be that of triggering an internal molecular change of structure, i.e. induction of specific order (orform ) not transmission of energy. In a way, the whole cycle concerns transmission of information. Only, what is lacking, is a subject to care for the 'information'.

Now, for a subject to be constituted some kind of boundary must be drawn to distinguish the subject from the surroundings. The molecular sharing economy suggested above, requires a cooperative specialization between distinct entities. The establishement of the molecular sharing economy automatically defines boundaries across which the cooperation proceeds and thereby also a system of entities. For the subject to arise, the sharing economy is the first condition.

But also for the subject to process information it must persist through time. Or else differences would be nonsense to it. It must 'exist' for itself, as was noted above. The development of a distinct code for memory, however, in the form of DNA was probably a rather late event. In the primitive prebiotic stages simple maintanance of structure through system-dependent catalytic activity would be all there was. May be this should not at first be called 'existence' but it might suffice for the gradual development of a separation between an informational sphere based on RNA and a functional sphere based primarily on peptides.

On the presumption that such a system could gradually develop sequence-conservation in distinct units of peptide aggregates a primitive kind of analog redescription of these aggregates might be said to take place. And certainly the single aggregate would then be engaged in what - from its own point of view - was purposeful behaviour. It would select differences, which made a difference. A semiotic relation would have established itself.

Figure 7 would describe a range of situations of more or less complexity. Even at the most complex stage however, real code-duality would hardly be established. The informational sphere of RNA would serve the internal communication between subunits, and thus indirectly rediscription of these subunits through time. And this internal communication would probably itself become digitalized But there would be no need for digital rediscription of the subunits. For this to happen one more event, would have to take place: reproduction.

In our view, however, the development of reproduction would not be the difficult step. Topological expansion of the system followed by splitting might be a first step. From a certain level of complexity such splitting would be dangerous due to the internal integration of the system. This would place a strong premium on copying mechanisms. Surviving splits capable of copying central components of the system would automatically spread through processes analogous to natural selection. From a modest start a better and better mechanism for match between peptides and RNAs would thus be developed. The chain of code-duality and evolution would have been set free.

The particulars of this scheme for the prebiotic development of information-processing entities may of course be far from reality, i.e. far from the route actually taken in this historical process. In our opinion, however, the succession of events leading from non-life to life would have to follow the pattern outlined in scheme 1.

The essence of this scheme is that an autonomous informational sphere would be developed only after the development of some kind of higher level organization was created. And, furthermore, that the development of a matching mechanism between sequences of RNA and protein was a late event in prebiotic evolution occuring only after the autonomous informational sphere was created. The 'cell' would have to be there for other reasons than the later creation of an internal library of DNA.

Nature's two digital systems

To the logician the use of metaphors is seldomly if ever justified. And clearly the import to the natural sciences of concepts developed inside humanistic disciplines like linguistics and semiotics is bound to provoke criticism (for some reason import the other way round are generally much more easily accepted, e.g. 'psychological energy', 'social homeostasis').

It has become increasingly clear in recent years, however, that the use of metaphors is at the very center of scientific development (Hesse 1966, Barnes 1974, Ricoeur 1977). There simply are no ways to escape the fact that scientific theories are linguistic expressions and therefore are bound up in the intricacies of language. A general condemnation of the use of metaphors would have a paralyzing effect on scientific thought. This, of course, does not mean that the use of metaphors is unproblematic. Most metaphors probably create more noise than message. Using metaphors one should at least be careful to make clear the theoretical implications of this new usage of a concept outside its established frame (Emmeche and Hoffmeyer 1988).

For centuries the guiding metaphor in the mechanistic tradition of biology has been that of nature as a machine to be controlled by the human population. The key-machine supporting this view of nature has changed, however, in correspondence to technological development. Thus, the old cartesian image of nature as a clockwork was later substituted with the image of nature as a steam-engine, and finally in our century the metaphor of nature as a self-governing cybernetic machine took over.

Although important traditions of biology did not strictly adhere to the machine metaphor, preferring maybe to see nature as a creation of God or as a supra-individual organism, the mechanistic tradition has had by far the greatest impact on the modern view of nature. Thus, the image of nature as a cybernetic machine which predominated biology in the 1970ties was also the rational basis for the greatest collective effort ever launched in biology, that of the International Biological Programme from1968 to 1974 (Kwa 1987).

Our proposal to view nature as a semiotic system is a proposal to abandon the belief in nature as a machine of any kind. Since in our understanding, semiotic systems presupposes code-duality, any metaphor reducing the essence of life to either digital or analog representation would be insufficient. To the extent computers are operated and controlled by humans they have no independent analog coded sphere. Unless connected to functional machines even the most sophisticated computers remain bound to the passive role of digital codes. Computers are allopoietic, not autopoietic (Maturana and Varela 1980)

The image of nature as an organism, on the other hand, is defective on the opposite criterium. The organism, after all, is only half of life, that of the analog sphere. As such the historical dimension of life is lost, since that dimension requires the digital phase of the organized gene-pool. And accordingly, excluding a few species such as primates or whales, organisms posses nothing like the semiotic freedom of the evolutionary double coded system.

At the bottom, what we suggest is a nature divided in a new way. Pointing to the semiotic freedom of evolutionary systems we imply, that such systems of nature are not so different from cultural systems. Or at least they would seem to be more like cultural systems than like systems of inanimate nature.

The old dichotomy of nature and culture does seem impropriate. Rather our place in nature should be seen as in figure 8. If the horizontal line seperating quadrants 1 and 2 from quadrants 3 and 4 in this figure is considered a watertight boundary, then we have the old culture/nature dichotomy of the mind/body dualistic worldview. Our claim, that the life sphere should be seen as a semiotic system based on the evolving code-duality between the digital DNA-code and the analog code of actual organisms, requires that much more stress should be laid on the vertical line seperating quadrants 1 and 3 from quadrants 2 and 4. The most amazing thing in our universe according to this view is the existence of code-duality at all. Due to the possesion of code-duality the life sphere and the cultural sphere should be seen as deeply related forming together the sphere which Bateson termed, through inspiration from C. G. Jung, the creatura to be distinguished from the world of no distinctions, the pleroma (Bateson 1972, Bateson and Bateson1987, Jung1961)

The construction in recent time of machines exhibiting socalled artificial intelligence (A.I.) means that entities now exist which posses the secondary digital system of human origin without possesing the original digital system of life. This might be taken as proof that an empty place (quadrant 3) in the logical space of possibilities has been filled. While the invention of artificial intelligence marks the transcendence of an important threshold in human tool-making, we do not consider it an important step in evolution. This is because even sophistcated computers, as we said above, are dependent on human beings for the function of real code-duality. A true member of quadrant 3 will have to await the construction of independent and self-reproducing computers - which we believe is not going to happen in the forseeable future, if ever.

'La langue' in nature: The genomorph ?

Needless to say, our metaphoric redescription of nature as a semiotic system will be fruitful only if central concepts and distinctions of semiotics do not lose their content by this transport from the human area to the area of living systems. Let us consider therefore in more detail the meaning to be ascribed to such concepts when refering to phenomena of life.

One severe objection to our scheme would be, that the concept of sign loses its content if the postulated interpreters of the sign are not intentional subjects. Are fertilized eggs intentional subjects?

It may seem ridiculous to answer this question in the affirmative. In a certain sense, however, we would like to do that. The problem is, that intentionality is not at all a well defined concept. For one thing, the existence of 'mental states' 'behind' or 'beneath' our linguistic expressions is highly disputable. Thus, according to Peirce (in 1868) 'the word or sign which man uses is the man himself' (cit.from Seboek 1986). And for Wittgenstein the subjective feelings, images and states of conciousness that attend the meaningful use of words and gestures are mere bi-products. They are not the causes of our ability to use words or signs, they are the effects of that ability. '...the mental experiences which accompany the use of a sign undoubtedly are caused by our use of the sign in a particular system of language' (Wittgenstein 1960: 78)

For Wittgenstein the real source of 'life' in a word or sentence is provided not by the individual mind but by society. Words are animated with meaning because of the social practices of which they are an integral part (Bloor 1983).

This does not mean, of course, that people are not intentional beings, only that their intentionality is inseparable from the broader context of the social practice of the involved group.

Now, zygotes do - we believe - contain nothing like subjective feelings or other mental states. They are not using signs of human origin and thus are not burdened either with the 'bi-products' of this use. They nevertheless are intentional creatures in a way comparable to that of human persons. Their intentionality is inseparable from the broader context of the project of the kimflok as defined above. Together they participate in the construction of the next generation of the population, reading for this purpose the message (in DNA-code) left over from the surviving part of the former kimflok. The whole cytoplasmatic organization of the fertilized egg is an analog message to the cell requesting the purposeful interpretation of the DNA-text it recieved. The meaning of this strange communication is not in the DNA, but in the perpetual process of translation between the DNA and the behavior of the organism in its ecological niche.

When using the concept 'communication' for the exchange of information through the life-cycles of the kimflok (one turn in figure 5), one should be aware that this is a very slow process compared to that of human communication. For the concept to be meaningful many generations (or cycles) would have to be considered, each generation contributing only one message to the 'talk'. Organic evolution is a slow process seen from the point of view of human life, but it has had to its disposal billions of years. Through such spans of time conversation, even if slow, can create many ideas.

Thus, the genome of one single zygote parallels, what in the structural linguistics of Saussure was termed parole (Saussure 1916), i.e. the actual use of language when persons speak or write. Guido Forti proposed that the phenotype should be considered the biological counterpart of 'parole' (Forti 1977). We would reject this analogy because the phenotype belongs to the analog part of code-duality, whereas 'parole' clearly resides in the digital sphere of life. (Emmeche and Hoffmeyer 1988) The analogy to 'parole' must be sought for in the messages written in DNA. And, obviously, single genomes are the vehicles for information exhange (language use) among zygotes. To be concrete, when the flower is pollinated a kind of 'talk' is going on. Two distinct plants make a common 'statement' as to the conditions of life, each contributing with half of their 'knowledge'. This statement, 'parole', is contained in the genome of the newly fertilized egg.

Saussure made a distinction between 'parole' and 'langue' , langue being the deep hidden structure of language, unconsciously shared by communicating persons. We shall suggest that the corresponding pair of concepts in biology is that of 'the single genome' (parole) as distinct from 'the deep structure of the gene pool' (langue). Since this deep structure may be regarded as a kind of 'informational morphology' at the genetic level, we shall term it the genomorph.

This concept of 'a genomorph' is, it must be admitted, somewhat speculative. But that might be said of the concept 'langue' too. In fact, 'langue' is a kind of abstraction thought necessary to explain the fact of language: that people talk to and understand each other. For this to happen something outside the individuals, a set of rules underlying the actual practice of talk, must exist.

In our view the existence of a deep structure in the gene-pool of a species as a necessary prerequisite for 'evolutionary communication' to take place, is one of the most fruitful suggestions to be gained from the semiotic perspective on living systems.

That genetic barriers exist, preventing sexual reproduction between species, implies that the freedom of the gene pool is in fact restricted by its own internal logic. And the phenomenon of genetic homeostasis (Lerner 1954) shows us that structural bindings inside the gene pool are important for understanding the evolutionary process. Until now, however, the 'mechanisms' of genetic homeostasy remain obscure. We suggest that this may be because there are no 'mechanisms' in the proper sense of this word, and that one should rather search for a set of 'linguistic rules', a genomorph.

It seems highly improbable that the genome would be capable of specifying the enormously complicated process of epigenesis without the application of directives belonging to different levels of logical typing. In fact, as we noted above, the ability of handling hierarchically ordered instructions, or 'meta-instructions', exactly is one of the main advantages of digital codes above analog codes. But if the fertilized egg - and its less and less prophetic daughtercells - should be able to distinguish between instructions of different logical typing, then this ability must be governed through the organization of the instructions on the genome. However, for sexual reproduction to be possible, this organization must itself be consistent to rules of still higher logical types, to be found in the organization of the species gene-pool. Thus, a many-levelled hierarchy of such rules should be considered. This postulated hierarchy of rules then would constitute the genomorph.

According to this view organic evolution concerns the change through time of the genomorph. While chance mutations are of course the ultimate source for such change, only a minor subclass of such mutations would be evolutionary interesting: those mutations affecting genome-coded directives of a higher logical type, i.e. mutations altering the frame through which the zygote (or better: the kimflok) interprets the messages of the genome. For evident reasons the actual genomorph characteristic to a certain species would itself restrict the space of possible alterations of this kind. Most mutations belonging to this subclass would cause the zygote to construct nonviable phenotypes, not even able to pass the filter of epigenesis. By prohibiting certain decisive mutations and endorsing others the structure of the genomorph would itself play a main role in further evolution. This, of course, would account for the predominance of evolutionary trends and support the model of puntuated equilibrium put forward by Eldredge and Gould (Eldredge and Gould 1972, Gould and Eldredge 1977). Since most mutations would be totally deprived of evolutionary significance the mere change of gene frequencies through time do not model evolution. These changes, which have been the subject of population genetics through most of this century, would only cause minor fluctuations around some average state, which itself would be stable. Measured over spans of time characteristic to paleontology this would look like stasis (Williamson 1981). Decisive events only would follow the successful introduction of mutants altering directives of a higher logical type. Whether such events deserve the name of macromutations (Goldschmidt 1940, Løvtrup 1987) depends on taste. In our view such events are comparable to what in the human sphere might be called epistemological breaks, i.e. the sudden restructuring of fundamental paradigms guiding our understanding. We should suggest then, that such alterations were termed epistemic mutations.

Another way to express this is, that the 'programmmes' contained in the DNA-code do have a logic of their own. Evolution is not just gradual change of DNA, but change in the deep structure or logic of the 'programme': Informational reframing. The neodarwinistic confusion of DNA and programme (substance and information) would therefore seem to be far too simplistic.

This claim, that the genomic programme obeys an evolutionary logic of its own, a logic which is not reducible to the aimless logic of DNA, does of course depend on the recognition of an intentional subject at play in the center of the evolutionary process. If sexual reproduction should be seen as a kind of language game whereby germ cells exchange information, then necessarily the question of purpose is posed. In the terms of structural lingustics there must be a signifié connected to sexual reproduction. What is this signifié ?

Again the intuitive conception of inner mental states behind our 'language games' would fool us to deny that a thing like 'signifié' might exist in the general life-sphere. And again let us cite Wittgenstein:

'If we had to name anything which is the life of a sign, we should have to say that it was its use' (Wittgenstein 1960: 4).

Now the 'use' of sexual reproduction can only mean one thing: The surprising fact that fertilized eggcells do in fact care to read the messages of the genomic programmes and furthermore that they do care to follow those instructions. The fact in other words that redescription and code-duality does exist. Certainly the cellular organization of the zygotes should be seen as a kind of analog information, telling the zygote to take the trouble to do exactly this job of translation. Gregory Bateson expressed a similar view:

'It seemed to me that we might think of the state of the egg immediately before fertilization as a state of a question, a state of readiness to recieve a certain piece of information, information that is then provided by the entry of the spermatozoon.' (Bateson and Bateson 1987)
And the question why this is so takes us back to the question of prebiotic evolution, the creation of the first system ready to select differences. The 'signifié' of structural linguistics would correspond in biology to nothing less than this 'readiness' of the egg?

Also, there is a striking parallel here to Piagets thesis of the cirularity of the reflex which actually, as has demonstrated Thure von Uexküll, is a semiotic relation. According to Piaget every stimulus presupposes a 'readiness to react' and this readiness therefore 'selects' as a stimulus a phenomenon of the environment which had been neutral up to that point. As the stimulus must realize the reaction, however, the reflex can only be described as a circular event, in which a neutral phenomenon recieves a property which it does not have independently from the reacting organ, and which it loses again after the completion of the reflex, i.e. with the cessation of the readiness to react (T.v.Uexküll 1986).

The hen and the egg

For centuries the central story guiding the imagination of science has been that of the falling apple of Newton. This was a story about how truth should be obtained through analysis of physical causality connecting phenomena of nature. Admirable amounts of knowledge have been obtained through this adventure. Even living creatures gradually became less and less of a mystery due to the continued application of this kind of scientific curiousity.

It cannot be denied, however, that many students of living nature, kept feeling uneasy towards the project of reducing the life-sphere to a purposeless result of 'blind' causality. Thus, for each generation of biologists a new wave of vitalistic or finalistic escape has been revived, just up to the present time.

The application of a semiotic paradigm to the study of nature has nothing to do with vitalism or finalism. But it does suggest another story to guide the scientific imagination concerning the life-sphere: That of the hen and the egg . This story directs our curiousity away from causality since, after all, in the case of the hen and the egg one would never be able to tell, which was the cause of which. Rather, according to this story, the key to understanding the phenomena of life is the processes of coding and recoding, i.e. translation between analog and digital codes (symbolically hen and egg).

Now, had the famous apple hit an egg instead of Newtons head, then of course no hen would be born from that egg. The story of the hen and the egg does not imply that causality need not be obeyed. It only implies that something more than causality should be accounted for, when living systems are explained.

As we noted, informational concepts have a long time ago been included in the set of accepted biological categories. Coding and self-description, however, are not just additions to biochemical and biophysical causality. Rather, the semiotic character of life has staged the physical and chemical processes of the world in ways, which are simply unbelievable from the point of view of mere causality. Causality, then, should be studied in the light of semiosis, not vice versa.. Due to the uninformed concept of information in present biology, however, the power of nature's semiosis is still largely hidden. A total reframing of biological theory is needed.

The eventual merit of the semiotic paradigm would be that old riddles of biology would disappear while a whole new kind of questions would have to be asked. In particular evolutionary and developmental biology might be fruitfully reformulated through such a perspective. In scheme 2 are shown some fundamental conceptions which would seem to undergo radical change if the traditional view of life is substituted by a semiotic view.

One consequence of such a paradigmatic change in biology would be that competences peculiar to humanistic disciplines might show up as very important to the future study of information-biology. Future policy for research and education should take this into account.

And, most important maybe, the connexion between biology and several humanistic disciplines operating in the neighborhood of the No Mans Land between human nature and human culture might take advantage of such a biology. The traditional perspective of biology have tended to make human nature look very alien to the human being. And humanists have often had very good reasons not to incorporate this view of human nature into their theories. It remains however a serious problem, that no convincing relation exist between these two areas. Since, according to the semiotic paradigm the processes of animate nature would seem to be much more like processes hitherto thought characteristic to human culture, many humanistic disciplines might in the future be able to incorporate biological insights into their thinking.

The broken bridge between nature and culture must be reconstructed. And to accomplish this would seem to be an urgent scientific challenge (Hoffmeyer 1988). To see living systems as fundamentally structured according to the rules of semiotic systems would be just one step towards this goal. Wittgenstein wrote: 'Only in the stream of thought and life do words have meaning' (cit. from Bloor 1983). We should like to add: Only in the stream of natural history and epigenesis do genes have meaning.

* We thank Peder Voetmann Christiansen, Frans Gregersen, Simo Køppe, Søren Løvtrup, Viggo Mortensen and Tor Nørretranders for comments on the manuscript and Stanley Salthe for a stimulating correspondence.

References

Alberch, P, S. J. Gould, G. F. Oster and D.B. Wake (1979). Size and Shape in Ontogeny and Phylogeny, Paleobiology 5, 396-317.

Anderson, Myrdene, John Deely, Martin Krampen, Joseph Ransdell, Thomas A. Seboek and Thure von Uexküll (1984). A Semiotic Perspective on the Sciences: Steps toward a new Paradigm, Semiotica 52, 7-47.

Auel, Jean M. (1980). The Clan of the Cave Bear, New York.

Barnes, Barry (1974). Scientific Knowledge and Sociological Theory. London: Routledge.

Bateson, Gregory (1972). Steps to an ecology of mind.. New York: Chandler.

Bateson, Gregory (1979). Mind and nature. New York: Bentam Books.

Bateson, Gregory and Mary Catherine Bateson (1987). Angels Fear. Towards an Epistemology of the Sacred. New York: Macmillan.

Bloor, David (1983).Wittgenstein: A Social Theory of Knowledge. London: Macmillan.

Brillouin, L. (1968). Life, Thermodynamics and Cybernetics. In Modern Systems Research for the Behavioural Scientist, Walther Buckley (ed.). Chicago: Aldine.

Bronowski, J. (1967). Human and Animal Language. In To Honor Roman Jakobsen. Essays on the Occasion of his Seventieth Birthday, Thomas A. Seboek (ed.), 1, 375-394. The Hague: Mouton.

Buchler, J. (1955) Charles Sanders Peirce: Philosophical Writings. New York: Dover.

Dawkins, Richard (1976). The Selfish Gene. Oxford: Oxford University Press.

Dretske, Fred I. (1981). Knowledge and the Flow of Information. Oxford: Blackwell.

Eldredge, Niles and Stephen J. Gould (1972). Punctuated Equilibria: An Alternative to Phyletic Gradualism. In Models in Paleobiology, J. M. Shopf (Ed.), 82-115. San Fransisco: Freeman.

Eldredge, Niles and Stanley N. Salthe (1984). Hierarchy and Evolution. In Oxford Surveys in Evolutionary Biology, R.Dawkins and M.Ridley (eds.), 1,184-208. Oxford: Oxford University Press.

Emmeche, Claus and J. Hoffmeyer (1988). The semiotics of nature. MS. To be published elsewhere. [here it is!]

Foerster, H.v. (1980). Epistemology of communication. In The Myths of Information.: Technology and Postindustrial Culture, Kathleen Woodward (ed.), Madison, Wisconsin: Coda Press.

Forti, Guido (1977). Structure and evolution in language and in living beings.Scientia 112, 69-79.

Goldschmidt, R. (1940).The Material Basis of Evolution. New Haven: Yale University Press.

Goodwin, Brian C. (1983). History and Structure in Biology and the Organism. Cahiers de la Fondation Archives Jean Piaget no.4, 275-319.

Goodwin, Brian C. (1984). Changing from an Evolutionary to a Generative Paradigm in Biology. In Evolutionary Theory: Paths into the Future. J. W. Pollard (ed.). New York: Wiley.

Gould, Stephen J. and Niles Eldredge (1977). Punctuated Equilibria: The Tempo and Mode of Evolution Reconsidered. Paleobiology. 3, 115-151.

Hesse, Mary (1966). Models and Analogies in Science. Notre Dame: University of Notre Dame Press.

Ho, M.W. and P.T.Saunders (1979). Beyond Neo-Darwinism - an Epigenetic Approach to Evolution. J. Theor. Biol. 78, 573-591.

Hoffmeyer, Jesper (1984). Naturen i hovedet. Om biologisk videnskab. Copenhagen: Rosinante.

--(1987). The Constraints of Nature on Free Will. In Free Will and Determinism. Viggo Mortensen and R. C. Sorenson (eds.). Aarhus: Aarhus University Press.

--(1988). The Possibility of a Historical Naturalism. MS, submitted for publication in Zygon.

Hull, David (1980). Individuality and Selection. Ann. Rev. Ecol. Syst. 11, 311-332.

Jung, Carl Gustav (1961). Septem Sermones ad Mortuos. In Memoires, Dreams and Reflections. New York: Random House.

King, M.-C. and A.C. Wilson (1975). Evolution at two levels in humans and chimpanzees. Science 188, 107-116.

Kristensen, Peer Hull (1985). Informationens og forandringens dobbeltspiral. Roskilde: Roskilde University Center.

Kwa, Chunglin (1987). Representation of Nature Mediating Between Ecology and Science Policy. Social Stu. Sci. 17, 413-442.

Leakey, Richard and Roger Lewin (1978). People of the lake. London: Collins.

--(1979). The origins of human language. New Scientist, 20.9.1979, 894-897.

Lenton, G. (1983). Wise Owls Flourish among the Oil Palms. New Scientist 97, 436-437.

Lerner, I. M. (1954).Genetic Homeostasis. Edinburgh: Oliver & Boyd.

Løvtrup, Søren (1974). Epigenetics. London: Wiley.

Løvtrup, Søren (1987): Darwinism - the Refutation of a Myth. London: Croom Helm.

Maturana, Humberto and Fransisco Varela (1980). Autopoiesis and Cognition. The Realization of the Living. Dordrecht: Reidel.

Monod, J. (1971). Chance and Necessity. New York: A. A. Knopf.

North, Geoffrey (1987). Back to the RNA world - and beyond.Nature 328, 18-19.

Nørretranders, Tor (1982). En videnskabsteoretisk analyse af beslutningsgrundlaget ved miljøhensyn i produktionsplanlægning. Roskilde: Roskilde University Center.

Pattee, Howard (1972). Laws and Constraints, Symbols and Languages. In Towards a Theoretical Biology, C.H.Waddington (ed.),4, 248-258. Edinburgh: University of Edinburgh Press.

--(1977). Dynamic and Linguistic Modes of Complex Systems, Int. J. General Systems 3, 259-266.

Prigogine, Ilya (1980): From Being to Becoming. San Fransisco: Freeman.

Ricoeur, Paul (1977). The Rule of Metaphor. Toronto: University of Toronto Press.

Salthe, Stanley N. (1988a). Possible Discourses on Neoteny in Biology. MS.

-- (1988b) Personal communication

Saussure, Ferdinand de (1966 [1916]). Cours de linguistique générale, (Payot, Paris 1965). English edition (1966): Course in General Linguistics. New York: McGraw-Hill.

Sayre, K.M. (1967). Philosophy and cybernetics. In Philosophy and cybernetics. J. Crosson (Ed.). Notre Dame: University of Notre Dame Press.

Seboek, Thomas A. (1986): The Doctrine of Signs, J. Social Biol. Struct., 9, 345-352.

Shannon, Claude E. and Warren Weaver (1949).The mathematical theory of communication. Urbana: University of Illinois Press.

Short, T. L. (1986). What they said in Amsterdam: Peirce's semiotic today. Semiotica 60 (1/2), 103-128.

Taylor, Peter J. (1987). Historical versus Selectionist Explanations in Evolutionary Biology, Cladistics 3 (1), 1-13.

Uexküll, Jacob von (1931). Der Organismus und die Umwelt. In Das Lebensproblem im Lichte der modernen Forschung, H. Driesch and H. Woltereck (eds.), 189-224. Leipzig: Quelle und Meyer.

--(1982 [1940]). The Theory of Meaning, Semiotica 42 (1), 25-82.

Uexküll, Thure von (1986). From Index to Icon. A Semiotic Attempt at Interpreting Piaget's Developmental Theory. in Iconicity. Essays on the Nature of Culture. Festchrift for Thomas A.Seboek on his 65th birthday. Paul Bouissac, Michael Herzfeld and Roland Posner (eds.), Stauffenburg Verlag.

Wilden, Anthony (1980).System and Structure. New York: Tavistock.

Wiley, I. W. (1981). Social Structure and Individual Ontogenies: Problems of Description, Mechanism and Evolution. Perspect. Ethol. 4, 105-133.

Williamson, P.G. (1981). Palaeontological Documentation of Speciation in Cenozoic Molluscs from Turkana Basin. Nature 293, 437-443.

Wittgenstein, Ludwig (1960). The Blue and Brown Books. Oxford: Blackwell.


[from the 1991 text:]

Jesper Hoffmeyer (b. 1942) is associate professor at Institute of Biological Chemistry B, University of Copenhagen. His principal research interests concerns the nature-culture relation in its historical development, the changing view of nature and its relation to biological theory.

Claus Emmeche (b. 1956) is a research worker in theoretical biology and theory of science at the Institute of Biological Chemistry B, University of Copenhagen. Ph.D. thesis on the concept of biological information in preparation. He is engaged in a technology assesment project concerning the meaning of biotechnology, the view of nature and the constraints on technological control over nature.