Physics is a science whose objective is to study
the components of matter and their mutual interactions.
In terms of these interactions the scientist
explains the properties of matter in bulk,
as well as other natural phenomena we observe.
There were days in the 70s when studying a subject at university and participating in a cultural and social revolution seemed like one and the same thing. When you were studying something like biology there was nothing the least bit strange in the fact that `biomass' became political student slang for the mass of biology students who constantly had to be `mobilized' against the bourgeoisie's reactionary measures directed against the experimental Roskilde University, university Marxism, long student careers and other benefits of the new society. And it seemed quite a matter of course that the biomass could not itself be a revolutionary subject, so it had to be mobilized from without, by the party or the `critical' avant-garde. I don't want to gloat over the romantic revolutionary naiveté of the period - plenty of others do that, although one is unlikely to understand history through irony and condemnation. But in this particular context the thought of the mobilizable mass came up again. Perhaps the very notion of masses - very generally - involves the idea that masses must be without a will of their own, or if they have one, that it is at any rate not very rational and must be given some logos and sense from the outside, whether it is the Truth and the Life, a guru, a God, an implicate order or a Party that is the missing factor. There have to be agents!
Not alone nor for itself
While in the scientific study of the human psyche there is a partly forgotten research tradition, mass psychology, which studied the regularities that arise when many minds are united in one movement, a mass, a crowd, a group, there does not appear to be a corresponding specialized research tradition in life studies which looks at the regularities that arise when a little life bands together and becomes mass life or biomass. And yet biomass is after all a solid ecological concept, and large parts of biology deal with the study of the mechanisms for the continued existence of life - and as William Blake said (in The Book of Thel, 2), "Everything that lives, / Lives not alone, nor for itself." So is there not all the same a `mass biology', hidden in the scientific landscape of biological disciplines and paradigms, and does it not study these very mass phenomena in biology? Let us look into it.
The dictum quoted from Blake here is in fact a quite basic biological view: life is something fundamentally relational; living organisms are defined among other ways in terms of special relations with the surrounding world, relations that have both a material-ecological side and a more communicative side. The organism seize upon its means of subsistence, but to do so it must orient itself in its world. The two sides of the organism-environment relationship, the material and the semiotic-communicative, have not been very well reconciled in modern biology - or rather the latter side has almost been sidetracked into small specializations like ethology (without disparaging Tinbergen and Lorenz).
Where is the biomass?
Put very briefly, biology has been misinterpreted in the twentieth century as a mechanistic discipline, capable only of studying living things in terms of matter and chemistry. The fact that other types of explanations were used as well as those of physics could easily be overlooked. Even when scientists did good biochemistry and molecular biology and celebrated triumphs with the mapping of the structure of DNA and the breaking of the genetic code, the biosystems could not be explained solely in terms of dyadic efficient causation. This had been the only kind of causation left by the Newtonian universe after it had rejected a richer kind of causation proposed by Aristotle, the Greek who also thought that form, matter and purpose could function as valid causal categories. Even in modern biology the researchers have used, and still use, functional explanations which implicitly see the parts of the organism as determined by the role they play in relation to the totality. Part and whole mutually determine each other, as Immanuel Kant put it. Heart, liver and kidneys have particular functions. So the basic philosophy of biology is not mechanism but an organicism where a kind of teleology (a theory explaining things in terms of appropriateness to their purposes) in the form of functional part-whole explanations governs our understanding of what a living being is. This does not prevent large parts of biology and ecology from being solidly rooted in a physical-chemical understanding of the living landscapes.
When a system ecologist casts a glance at a rain forest with its myriad species, he does not only see these individual forms; if he knows his ecology, he also sees quantities of biomass, which with the precision of a dictionary he defines correctly as "the total quantity of matter (the non-aqueous component frequently being expressed as dry mass) in organisms, commonly of those forming a trophic level or population, or inhabiting a given region". This organic matter comes at first from the green plants, the primary producers. Thanks to the ingeniously designed biochemical energy trap called photosynthesis, the leaves of the plants capture the photons from the sun and make them power the water-wheel of metabolic processes in the cells that leads to the building of carbohydrates and the important macromolecules. These macromolecules form the bulk of the plants' biomass ... which is then eaten by the primary consumers, i.e. the herbivores ... which are in turn eaten by the secondary consumers, and so on almost ad infinitum, because these `trophic levels' are in reality not separate levels but a whole network of transfers of matter and energy in the supermarket of ecologically correct products which the system ecologist ends up describing the rain forest as. This way the biomass is reduced to flows of matter and energy - a handy, firm description if one wants to understand something like trophic efficiency, the fact that only about 10% of the energy from the lower trophic level is gathered up as biomass at the next level, which is why it takes an area ten times as large to feed a population of beef-eating humans as a population of dedicated vegetarians - or rice farmers. Thus the biomass from the trophic levels of the rain forest can be illustrated as a pyramid with the primary producers as the green base. Inasmuch as ecology is interested in the individual species in their specificity, it speaks in terms of niches, where the niche is the mode of functioning of the individual species in the ecosystem, its special contribution to the network of energy and matter.
Stimulating the habits of the agent
But biomass is after all organized in other, far more ingenious ways than the simple dyadic-ecological relations of the type illustrated by the figure "tiny fish is eaten by little fish is eaten by fish is eaten by large fish is eaten by huge fish". The fact that the fish food chain is not just simple and dyadic but complex, and dependent on constant communication among organisms of different or the same species, represents the semiotic dimension of the North Sea or the mangrove swamp. For fish have to be able to find one another before they can at all eat, mate and do their other business. The very fact that a large proportion of the living organisms, indeed perhaps all of them in reality, have some "business" - i.e. exhibit purposive behaviour, search, want something, do something, experience something, etc. - shows that they are not ecological objects but biosemiotic agents which have particular habits ("whether they know it or not!" as Marx put it brusquely in a slightly different context) - habits they are either born with or have learned.
As a collection of semiotic agents the biomass appears in its complex triadic aspect as embedded in a network of signs, understood in the radical sense that it is not only the human subject that ascribes to the organisms of nature the ability to produce, transfer, preserve and interpret signs - they do it themselves! That is at least the interpretation that so-called biosemiotics proposes. It represents semiotic realism; it is precisely nature itself that, thanks to its power of self-organization, has crossed a threshold of complexity, past which things do not just spontaneously degenerate, but past which evolution gathers speed, assumes a creative character, and forms a semiotic network of relations which combines the ecological consumers and producers of the biomass in a dramatic market where signals are exchanged and semiotic actions are performed for dear life.
Just as the human economy is not just a purely material or economic matter, but only comes into play thanks to the human mind's ability to represent very abstract relationships (like a general equivalent for the value of human work), so the prior emergence in natural history of an ecological sphere of circulation and production only comes into play when matter or `mass' on the earth has reached a level of organization that permits the formation of agents, not with consciousness but with a capacity for semiotic action and the formation of an Umwelt (which is semiotic and a counterpart of the niche, and which has a subjective aspect inasmuch as it is an experienced world). This semiotic action takes place at two levels: an endosemiotic level, i.e. within the cells, on the basis of the duality that has arisen between a digital genetic code (physically carried by DNA) and a dynamically analog metabolism (which is not in itself a code, but which is implicitly and partially coded in DNA); and an exosemiotic level which is concerned with the production of signs and their communication to the outside world, and the way signals from the world are interpreted by the cells (not as `conscious interpretation' but as triadic semiotic action in Peirce's sense). It is thus the cells that are the primary semiotic agents; with its closed organization (think of the cell membrane) the cell creates the primary difference between outer and inner, and thus indirectly between living and dead. It is this primary `organismic information' which is the precondition of both the endosemiotic and the exosemiotic. And it is the whole network of sign relations, the semiosphere in the global sense, that determines in the first place whether the biomass can be organized in something as ingenious as an ecosystem, and the biosphere in a global system of ecosystems.
My interim conclusion is that a biomass psychology does in fact exist, although as a slightly off-centre, partly new, but not wholly established tradition in biology alongside the established disciplines like cell biology, evolutionary theory, genetics etc. That this biosemiotics has nothing to do with mass psychology beyond a slight formal resemblance (manifested by the interest in "downward causality" - see below), is due to the distinctiveness of its object. For mass psychology is a kind of social psychology that looks at the interplay between the psychological and the social in the human sense. It is about the critical points where a mass of `psyches' or individuals depart from an institutionally established framework for `normal' interrelations and at the same time from their own isolated individuality, and thus form politically unstable and socially uncontrollable phenomena, such as mass hysteria. Mass hysteria is not necessarily something that simply breaks out spontaneously, a kind of bottom-up self-organized structure; history has plenty of examples of carefully staged situations where the aim is the negation of the individuals' limits of rationality and normal relations and the creation of a collectivist and manipulable crowd where the crowd dynamics of social habit as the emergent phenomenon works `downwards on' the individuals (downward causality) by temporarily or permanently eradicating their individuality, thus distorting a collectivity of people in a collectivist, i.e. conformist direction. Biomass psychology is not really psychology, but in fact biosemiotics, which basically tries to explain phenomena of self-organization at a much lower level, where there is no question of anything psychological in the human sense, but of agency, understood here as semiotic action (the production, exchange and interpretation of sign relations): an organism, a single cell, indeed even a biological macromolecule in a cell - such as a protein which `recognizes' another protein because the surfaces of the two materials match and weak interactions (in the form of hydrogen bonding) are established - can behave as nodes of a sign relation, because these entities are associated in the actual interchange with the agent, the biological organism as a functional whole. The individual macromolecule always has one or more functions (this is evident from the general biofunctional description), but this can also be described relationally and semiotically as the association of the part (the molecule) with the whole (the agent, the organism) in a complex semiotic relation.
As the second conclusion we can see here that the biomass on Earth is organized as a massively ramified network of signs: the signs act - by virtue of their associations (`sign-links') with the sign-producing, sign-bearing and sign-interpreting semiotic agents - as what links these otherwise organizationally closed and self-sustaining (autopoietic) systems with an evolutionarily and ecologically open network of life; in fact, as a pure bio-internet. This biosemiotic account of things is certainly not hard science, inasmuch as established biology has not absorbed the fundamental biosemiotic concepts. The fact that this is a "philosophy of nature" interpretation of established biology rather than a fully-fledged `alternative experimental paradigm' should not however blind us to the fact that biosemiotics is an important bridgehead if we are one day to construct a proper theory that explains the natural history of meaning, i.e. how something like agents with a psyche in the human sense (and thus also the potential for a mass psychology) could arise in a universe that was originally - who knows? - devoid of meaning.
The agent who came into the world
It is thus the claim of biosemiotics that the organism - as it is understood in organicism and in good modern mainstream non-reductionist and evolutionary biology - has a subjective side, which means that all organisms, even single-cell organisms, can be regarded as agents. Agens after all means simply acting, setting in motion, doing something under one's own power (for Aristotle organisms were precisely those bodies that moved by themselves). The concept of subject has been used narrowly by philosophers since the seventeenth century of that which unifies human consciousness, and which is at the basis of all emotion, sensing, thinking, and volition. Biosemiotics asserts the deep roots in natural history of the human subject by pointing to agency as something far more universal in life as such. And this immediately raises a whole set of questions of origin, about where this agency comes from (whatever version we are talking about - sensing, perception, recognition etc.); does biosemiotics pull it down like a deus ex machina, a hallelujah that magically explains what traditional biology neither can nor will explain? How can agency arise in a world of causes and effects determined by natural laws?
The philosophical mass of riddles associated with this is something we shall not reach the bottom of here, but - in terms of philosophy of language - we can remember to consider that the fundamental everyday concepts in language that we use to talk about causes, actions, choices, experience etc. form the point of departure for more refined concepts in science, and if we are to understand these at a deeper level, we must in the end resort to this pith of language that consists of ambiguous and vague meanings. Trying to find the meaning of concepts like mass, action and cause is a veritable morass. The philosopher Jennifer Hornsby says at one point that one can assume that the concept of agency as the subjectively experienced power of action which human beings acquire precisely from their experience of acting is prior (in some sense) to the concept of causality. Collingwood claims for example that our original concept of cause is derived from the concept of agency.
In the premodern world any causation (any cause-and-effect relationship) in the absence of human beings was either something that took place through divine intervention or it was an agency-like action in an object whose nature was to realize particular ends. In the modern world final causes were eliminated. The astronomer Kepler maintained that the word anima had to be replaced in treatises on physics with the word vis (force), and later the Cartesians rejected the concept of force as metaphysics and maintained that only matter in motion existed, that there was no room for purposes in explanations of causality. But the postmodern thinker, scientist and metaphysician C.S. Peirce took a new interest in final causation, now embedded in a semiotically realistic metaphysics. In this view the relationship between what we have here called biomass and biosemiotics is more or less to be understood as follows. There are two kinds of actions in our universe, dyadic and triadic. Dyadic action is mechanical or dynamic, and is concerned with efficient causation as described for example in ecology in connection with the biomass. The triadic action type is semiotic, or intelligent; it concerns final causation as described in biosemiotics. The two kinds of action are irreducible, but inseparable and superimposed. They can also be called...
The brutal and the final
The point is now this. There are roughly speaking two different views in play, an old and a new one. Let us look at the old one first. Both classical and modern physics have been subject to a metaphysics that only operated with dyadic relations as in the classic cause-and-effect relation (what Peirce called "brute force", i.e. that something is affected by something else without mediation, has this dyadic character). This view fits with the notion that a system can always be described in terms of an initial state plus a set of deterministic equations of motion corresponding to the laws of nature, where the state of the system at the point in time 0 and the equations of motion uniquely determine ("are the cause of") the state of the system at the point in time x. This makes it hard to imagine that mass phenomena can be governed by anything but the laws for the particles of which they are composed. Thus mass becomes a macroscopic phantom with no meaning, for it is the dynamics at the micro level that have been the causal moving force for the system as a whole. Put differently, this should imply that any mass phenomenon, whether psychological, social, biological or physical, can be described in accordance with the theory of dynamical systems as the motion of a set of particles through a state space. But here we come up against the problem that the theory of dynamical systems places narrow limits on the category of systems we can actually deal with within such a descriptive framework; and complex non-linear and evolutionary phenomena where new, emergent properties arise at the higher level, the level of mass, drop out of the description. In principle the whole can be captured by the deterministic description, but in practice it cannot be done. In the face of this problem so-called deterministic chaos represents a borderline phenomenon - that is, it can be forced into the deterministic description by means of differential equations, but in practice, for the real system represented by the equations, we cannot be sure of predicting what will happen, or for example which attractor system it will be caught up in. Chaos theory is thus in a certain sense a field that exposes anomalies in the old view: prediction and control are no longer possible in the classical sense. Another difficulty for the old view is, as suggested, that it tends to make all emergent or supervenient phenomena into pure epiphenomena, because it is the causally closed microphysical level that alone determines the development of all causally potent factors in the system. Consciousness too becomes an epiphenomenon.
The new view is then that the two levels of description - of the part and of the whole, the level of the individual and the mass - must necessarily be seen as complementary, where we cannot totally reduce the mass level to the microphysical description. Again, this applies within as well as outside physics. Within physics - where `complementarity' is used in the strict sense - we know about Niels Bohr's demonstration of the epistemological situation in which we stand with quantum mechanics: that while with quantum formalism we have access to an understanding of microscopic phenomena, in this understanding our old concepts of causality, place and time are no longer enough; on the other hand we can take comfort in the fact that the precondition of the actual microdescription is that we can maintain a classical, macroscopic description of - for instance - our measuring apparatus: we can never reduce the whole measuring situation to a quantum mechanics system, nor can we understand the quantum universe simply in classical terms; these descriptions are complementary. Outside the descriptive domain of physics, for example in biology and mass psychology, `complementarity' (used in the broad sense of levels of description) simply has the crucial consequence that the mass phenomena can be studied "in their own right" without having to be regarded sceptically as some sort of mystical phenomena not yet reduced to microphysics. If we are to offer some sort of physically intuitive notion of the biological phenomenology (of cells, DNA, proteins, indeed organs and organisms) as mass phenomena compared with the atomic and subatomic levels, we must say something like this: while it is true that an organism - from a particle physics point of view - is only a set of interacting particles, these are still so highly and complexly organized that these macroscopic levels constitute independent constraining boundary conditions for the dynamics of the lower particle levels, and that these boundary conditions themselves have a profound historical (and thus also biofunctional and biosemiotic) character, which is why the particle description can never be adequate. This may sound rather "physical", and so it is, but this new physical view (which is widespread in `the physics of complex systems', chaos theory, artificial life etc.), also has a more metaphysical, or one could say biosemiotic interpretation: it is the meaning in nature as an emergent phenomenon - that is, something genuinely new created through the history of evolution, which is yet continuously (synechistically) connected with the old - which constitutes the true constraining conditions which ensure that the habits nature has acquired at the level in question remain relatively stable.
The habits and bad habits of the masses
Some habits are literally `elementary', since they have to do with the chemical elements we find at the quantum-mechanical level, where they ensure that matter does not collapse but remains stable (quantum mechanics represents these fundamental habits as expressed in the periodic system); other habits have been formed later at a far higher level, the level of the biological agents, where one of the biosemiotic functions of DNA is to remember which amino acid combinations produce the right, life-sustaining proteins. At this level the habits are not so terribly rigid; there we find an interplay between environmental necessity and mutational and recombinatory randomness. In this interplay the habits of the agents are an example of what Peirce would call thirdness, for the more refined habits an organism can master, not only implicitly coded in a closed behavioural programme, but acquired through learning, the more the ontogenetic habits can react back on phylogenetic evolution: the more competent agents (behaviourally by virtue of their semiotic competence) select their own niches, and thus the selection pressure to which their phylogenetic trajectory will be subjected. This is not Lamarckism, but it is still "evolution guided by learning", an effect that is in fact well known in biology (the Baldwin effect), recently rediscovered through studies in Artificial Life which combine genetic algorithms with neural networks. Life as a mass phenomenon, the tracks of the individual species through the evolutionary hunting-grounds is therefore `led' or guided by semiotic competences, where the agents with the best developed Umwelt will also have the best chances of strong feedback from learning to evolution.
Once physics was about the atomic masses, `matter in bulk', aggregates of atoms in solid, liquid or gaseous form. Limiting its object today to atomic mass phenomena is passé. To the concept of mass adheres a connotation of something simple and disorganized. Physics and biology are today turning with shared curiosity towards the complex, towards life and processes of development, and towards the emergence of new habits in nature. Biomass is more than mass; biomass contains highly organized and superspecific structures from the molecular level all the way up to the levels of the multitudinous species, the ecosystem and the biosphere. These structures are organized as agents, and agents are characterized by having both external `objective' properties and an individual Umwelt, an inner sensory side which means that the agent, whether it is something of an odd man out or is right there in the midst of the mass, not only follows the laws of habit, but also breaks them. Semiotic agents mobilize themselves. What a bad habit!
 Marcello Alonso & Edward J. Finn 1970: Physics. Addison-Wesley, Reading, Massachusetts (p. 2). Cf. the same source, p. 13: "Matter in bulk, in the way it affects our senses, is an aggregate of a very large number of atoms or molecules. Grossly speaking, these aggregates appear to be in three physical states or phases designated as gases, liquids, and solids."
 That final causes or functional explanations play a major role in biology has not prevented people from discussing whether final causes can be understood on the basis of (and perhaps reduced to) explanations in terms of efficient causes. For a through overview of the philosophical discussion, see Colin Allen, Mark Bexkoff and George Lauder (eds.), 1998: Nature's Purposes. Analyses of Function and Design in Biology. MIT Press, Cambridge, Massachusetts.
 M. Abercrombie et al., 1992: The New Penguin Dictionary of Biology. London, sub "biomass".
 It should be noted that the presentation of ecology in this article, as focused one-sidedly on matter and energy aspects of the biomass, only applies to system ecology, not to ecology as a broad scientific field. See for example M. Begon, J.L. Harper & C.R. Townsend, 1990: Ecology. Individuals, populations and communities. (2nd ed.).
 This may sound a little mysterious to the agent who has never come across Peirce and his theory of signs before, but think about it something like this: a sign is a relation among three aspects, the physical bearer of the sign (the representamen), what the sign stands for (the object), and the effect (the interpretant) that the sign has, potentially or actually, on an organism or part of an organism. The last of these, the being or thing on which the sign has an effect, may itself be a sign in a new semiotic process. We thus distinguish technically between interpretant and organism as a possible interpreter. For an introduction to biosemiotics, see Emmeche, 1997: "Den biosemiotiske tanke" [The idea of biosemiotics], pp. 62-94 in Keld Gall Jørgensen (ed.): Anvendt Semiotik. Gyldendal, Copenhagen.
 See also Claus Emmeche 1998: Information i naturen, Nyt Nordisk Forlag, Copenhagen; C. Emmeche 1989: Det biologiske informationsbegreb, Kimære, Århus (pp. 134f); Jesper Hoffmeyer 1996: Signs of Meaning in the Universe. Bloomington: Indiana University Press; J. Hoffmeyer 1998: "Surfaces inside surfaces: on the origin of agency and life", Cybernetics and Human Knowing 5(1): 33-42.
 Important pioneers (besides Peirce) in the area of biosemiotics are figures like Jakob von Uexküll, Thure von Uexküll, Thomas A. Sebeok, Martin Krampen, John Deely, Jesper Hoffmeyer, Kalevi Kull and others. For a kind of manifesto see M. Anderson, J. Deely, M. Krampen, R. Ransdell, T.A. Sebeok, & T. von Uexküll, 1984: "A semiotic perspective on the sciences: steps toward a new paradigm", Semiotica 52 (1/2), 7-47. Further material on biosemiotics can be found at http://www.molbio.ku.dk/MolBioPages/abk/PersonalPages/Jesper/Hoffmeyer.html
 That the phenomenon is emergent means that it is new, that it represents a phenomenon of a higher order, that it has emerged from the phenomena of the lower orders, and that it cannot be fully explained or understood in terms of these. Mass-psychological examples are the formation of a new Zeitgeist; the Messianic longings of millenarian sects; and the youth revolution at the end of the sixties. Phenomena like these cannot be explained simply as the sum of the changed orientations or attitudes of individuals. The phenomenon develops its own dynamic, its own life. Biological examples of emergence are the evolutionary formation of the first cells from simpler macromolecules; the formation of sexually propagated populations which constitute gene pools in the genetic sense; and the formation of mental phenomena in organisms with a nervous system.
 In social psychology one should distinguish between a collectivism, which reduces the individual's personality and as a tendency is therefore totalitarian, and a collectivity in more egalitarian and tolerant forms which permits the existence of a rich individuality.
 In the article "The Sarkar challenge to biosemiotics" (Semiotica, in press) I have tried to take up the issue of the relationship between the ways in which molecular biology and biosemiotics describe `molecular recognition'.
 Hornsby, sub `agency', p. 18 in Ted Honderich (ed.), 1995: The Oxford Companion to Philosophy. Oxford University Press. See also Olaf Pedersen, 1990: "Naturvidenskabens fødsel - et sprogligt problem", pp. 13-35 in Thor A. Bak & Erik Dal, eds.: Videnskabens Enhed-? Munksgaard, Copenhagen.
 `Il Saggiatore' (Opere, 1890, etc., vi, p. 232), quoted here from p. 102 in R. G. Collingwood, 1945 (1960): The idea of Nature. Oxford University Press.
 Peirce, C.S. (1931-58): Collected Papers of Charles Sanders Peirce, vols. 1-8, eds. Charles Hartshorne, Paul Weiss & Arthur Burks. Harvard University Press, Cambridge, Mass. An introduction to the concept of causality in Peirce is Lúcia Santaella Braga, 1994: "Peirce's broad concept of mind", European Journal for Semiotic Studies 6 (3, 4): 399-411; and Lúcia Santaella Braga, 1996: "Semiosphere: The growth of signs", Semiotica 109 (1/2): 173-186.
 Complexly organized - that is, in the first instance, as atoms and molecules, in the second as cells and genetic information (there we have some of the digital signs!) and as interactions among cells at the level of "mass", the multicellular organism. Biological information and biological organization are two intimately linked entities; cf. Emmeche 1989 op. cit.
 Bernd-Olaf Küppers has pointed out that it is a characteristic of biological systems that in them, the boundary conditions themselves are complex, because of their historical character; see B.-O. Küppers 1992: "Understanding complexity", pp. 241-256 in A. Beckermann, H. Flohr and J. Kim, eds.: Emergence or Reduction? Essays on the Prospects of Nonreductive Physicalism. Berlin & New York: Walter de Gruyter.
 Also called `organic selection', and known for example from Mark Baldwin, 1896. (Peirce knew of Baldwin, but I do not know whether he knew his theory of organic selection). There were several scholars who were more or less independently on the track of this effect; for some account of this, see Robert J. Richards 1987: Darwin and the emergence of evolutionary theories of mind and behavior. University of Chicago Press, Chicago and London.
 G. Hinton & S.J. Nowland, 1987: "How learning can guide evolution", Complex Systems 1: 495-502; see also D. Parisi, S. Nolfi, and F. Cecconi, 1992: "Learning, behavior, and evolution" pp. 207-216 in: F.J. Varela and P. Bourgine, eds.: Toward a Practice of Autonomous Systems. Proceedings of the First European Conference on Artificial Life. Cambridge, Mass.: The MIT Press.