Citation: Hoffmeyer, J. [2001]. "Life and Reference". Biosystems. Vol. 60, pp. 123-130.
This paper is also available from Elsevier's web site
Abstract:The paper recommends a broadening of Howard Pattee's seminal distinction between a dynamic and a linguistic mode of living systems. It is observed that even the dynamic mode is always a semiotic mode although indexical and analogly coded rather than symbolic and digitally coded. The analogically coded messages corresponds to a kind of tacit knowledge hidden in macromolecular structure and shape (e.g. molecular complementarity) and in organismic architecture and communication, i.e., in the semiotic interactions of the body. It is claimed that the origin of referential processes is tied to the flow of historical singularities. The function of analog and digital codes in evolutionary systems is discussed.
Keywords: Code-duality, reference, biosemiotics, singularity, Lamerck, semetic interaction
Reflections on the emergence of life necessarily brings us to the problem of discontinuity and continuity: Should we choose to see life as just one particular instantiation of more general tendencies characteristic to the universe we inhabit, or should we rather choose to see living systems as decisively different from anything else in the known world? If we assume, as most scientists probably intuitively do, that we all live in one shared historical world, then a continuity approach would seem to be the preferable strategy. And yet, as pointed out throughout Howard Pattee's work, there is no way we can escape the fundamental epistemological problem posed by living systems: living systems are inherently engaged in measuring processes, which immediately introduces us to the paradox of "the epistemic cut":
"We must define an epistemic cut separating the world from the organism or observer. In other words, wherever it is applied, the concept of semantic information requires the separation of the knower and the known. Semantic information, by definition, is about something" (Pattee 1997). Pattee underlines that he is not suggesting a Cartesian dualism here but only a "descriptive dualism", for although a measuring process depends on choices which cannot be derived from laws, such choices are seen by Pattee as functions coded in DNA and ultimately generated by natural selection. Pattee's idea of a "semantic closure" accords with these conceptions. Measurement and observer are epistemic concepts and as such they presuppose at least an organization that can construct the measuring device and use the results for its survival. Subcellular entities such as enzymes are not sufficiently sophisticated to be counted for as measuring devices: "To qualify as a measuring device it must have a function, and the most primitive concept of function implies improving fitness of an organism. Thus, observation and measurement require an organization that (1) constructs the measuring device and (2) uses the results of the measurements for survival. This requirement I have called the semantic closure principle (Pattee 1982). This provides an objective criterion for distinguishing measurements and observations from other physical interactions. Only organizations with this semantic closure property should be called observers. The cell is the simplest natural case of an observing system" (Pattee 1996).
A major difficulty with this "epistemic cut position" is that it makes it very difficult to see how life could ever have evolved at all, and although Pattee has titled the paper from which one of these quotations was taken "the physics of symbols and the evolution of semiotic controls" he does not suggest anything more specific than a sort of downward causation through the action of natural selection: "Just as the observer's cognitive selection process is necessary for a measurement, so natural selection is necessary to generate function or meaning in the genetic DNA" (Pattee 1997). But even if we accept the idea that life's semiotic constraints on flows of energy and matter must ultimately be derived from a process of differentiation of more general constraints operating in our universe (Christiansen 1999) our need for an understanding of how this could in fact be brought about is not exhausted. How did physical constraints become semiotic controls?
A key term in answering this question will be the term "reference". What does it mean that something inside a system refers to something outside of this system in such a way as to adjust the inside processes to the outside conditions in a functional way? An E. coli bacterium for instance perpetually measures eventual changes in the external concentrations of a diversity of nutrients through the binding of nutrient molecules to chemoreceptors at the bacterium's "nose" (Parkinson and Blair 1993), and the time-weighted result of this measurement is used in the regulation of flagellar movements and thus in the regulation of the bacterium's chemotactic behavior. The referential character of this process, which actually involves dozens of topologically well-organized protein species, would be hard to deny, and as emphasized by Pattee the apparent functionality of this kind of integrated process requires it to be coded in DNA and ultimately generated by natural selection.
It should be noticed that this argument has a spatial as well as a temporal component to it. The spatial component concerns the necessity for an inside-outside asymmetry; i.e., the referential process requires a distinction between system and environment, "self" and "other". The temporal component concerns the peculiar way semiotic systems have invented for persisting through time, what Claus Emmeche and I called code-duality (see below and Hoffmeyer and Emmeche 1991). Semiotic systems are mortal which means that they only survive due to multiplication of the digitally coded self-descriptions built into the sequence structure of DNA molecules. In uni-cellular organisms survival still requires the co-production of both DNA and organism, while in multi-cellular organisms the co-production of a single (germ) cell is enough, but the basic differentiation of roles into digital self-reference and analog other-reference is basic to all kinds of life.
It follows that the origin of life is inseparable from the origin of the environment and I have argued that we have strong conceptual reasons to assume that the formation of a closed semi-permeable membrane of some sort around an autocatalytic chemical system preceded the formation of replicative molecules. By creating a topological closure the membrane established an inside-outside asymmetry which must be an absolutely decisive step because it opens the door to the semiotic world of communication and function and thereby to the formation of an individualized context space or agency. One should not forget furthermore, that the asymmetry between organism and environment was initially an asymmetry between the membranes' excluded outside and it's excluded inside (Hoffmeyer 1998b). The membrane itself is the locus for individualization, it potentially points out an interest space in the world (a semiotic niche) and in the same time it opens a potential internalized environment, or Umwelt (Uexküll 1982 [1940]).
In his book Ecology. The Ascendent Perspective Robert Ulanowicz finds the origins of selfhood exhibited in the form of the "centripetal" property of an autocatalytic loop: "Taken as a unit, the autocatalytic cycle is not acting simply at the behest of its environment. It actively creates its own domain of influence. Such creative behavior imparts a separate identity and ontological status to the configuration above and beyond the passive elements that surround it. We see in centripetality the most primitive hint of entification, selfhood and id. In the direction toward which the asymmetry of autocatalysis points we see a suggestion of a telos, an intimation of final cause (Rosen 1991). Popper put it all most delightfully (Popper 1990), 'Heraclitus was right: we are not things, but flames.' Or a little more prosaically, we are, like all cells, processes of metabolism; nets of chemical pathways" (Ulanowicz 1997, 47-48). In the same vein, I suppose, Stanley Salthe's idea of a specification hierarchy leads him to see agency and semiosis as generalized properties of our world as a whole to be exhibited also in abiotic phenomena such as tornadoes (Salthe 1993; Salthe 1998).
Much as I sympathize with these ideas I also see a problem here. Autocatalytic cycles or tornadoes do indeed deserve to be understood as (ontologically) real entities, and yet there is a very important sense in which they seem to me (and Pattee?) to be logically different from living entities: they are not interpreting their environments, there is no need for introducing an "epistemic cut" separating the system from the world because there is no stable internal symbolic representation of the environment.
The critical question here is the role which DNA plays in life. As a result of the aggressive genocentric bias or DNAism of mainstream biology there may have been an exaggerated tendency among evolutionary systems thinkers (and developmental systems theorists) to tone down the particular contribution of this digital code to the dynamics of evolving life. The creation of nature's first digital code, DNA, was however a key to a very new world, the world of symbolic reference.
When Claus Emmeche and I first introduced the term code-duality and emphasized the crucial role of this duality of analogically and digitally coded survival instructions, we pointed to three characteristics of digital codes (such as DNA or natural language) which make such codes indispensable for any evolutionary process (Hoffmeyer and Emmeche 1991). These three characteristics were:
Digital codes allow for impossible messages because there is no strict binding between the code itself and the message it carries. In linguistic messages anything goes; Socrates may have lunch with Meryl Streep and the wives of pilots may give birth to children carrying wings. The same is true of genomes. Impossible genetic instructions are created all the time through processes of genetic recombination such as, e.g. crossing over, resulting in early abortions or the birth of non-vital descendants. The incredible combinatorial capacity of living systems for creating endless chains of novelties is due to this freedom.
Digital codes are codes for memory. Only because Plato wrote down Socrates' dialogues do we know them today. Had the dialogues not been coded in language but in pieces of mime they would probably have died with Socrates himself or would have survived a few generations at most. The key here is that digital codes are objective in the sense that they depend on a shared convention. Thus, Anthony Wilden pointed out the following distinction between analog and digital codes: "a digital code is 'outside' the sender and receiver and mediates their relationship; an analog code is the relationship which mediates them" (Wilden 1980, 173). Genomes are not normally thought to be outside the sender or the receiver, since they are normally exchanged via processes of mitosis or meiosis. But if, as we try to do here, life is seen as unfolding across membranes, then even the inside of the membrane is - as we pointed out above - an outside. The important thing in this context however is that DNA is actually protected from the vicissitudes of life by its relative chemical inertness and by a sophisticated apparatus of enzymatic repair systems. In sexual reproducing species furthermore the germ line is kept separate from the somatic cell lines so that no Lamarckian inheritage is normally believed to take place. Digital codes thus are necessary to assure the temporal semi-stability needed for evolution - nothing can evolve if it is not remembered (because then we talk about substitution).
Digital codes are eminent tools for the construction of meta-messages, i.e., messages necessary for interpreting other messages. Gregory Bateson showed that meta-messages may also be communicated in the analog as when young monkeys were engaged in so-called "play", i.e., an activity in which they exchanged signals similar to those seen during combat. As Bateson observed, when the monkeys snapped at one another while creating an imaginary combat situation, this snap would actually signify the following "meta-message": "this is not a bite" (Bateson 1972, 177-193) The absence of a bite is announced by the presence of the snap. The snap is an indication of something that is not there. But Bateson also commented that this is probably as far as an analogically coded communication can go in the direction of the abstract category of "not". For real abstractions to take place digital codes are needed. We still do not know the full syntactic structure of the genetic code, but regulatory genes are examples of meta messages, and the occurrence of atavisms such as the three toed horse seems to indicate that exclusion is not the only way to get rid of outmoded ontogenetic instructions. Negation may suffice. Abstraction thus furnishes plasticity in the absence of which the evolutionary process might perhaps not have been as rich as it actually is.
These three essential aspects of digital codes not only make these codes indispensable tools for the evolutionary process but they also explain why digital codes are fundamentally passive. We do not believe in spells because there is no consistent way that the mere pronunciation of words could cause desired physical events to take place, and likewise we should not believe that genes by themselves do anything. In both cases it takes an interpretant to mediate between the message and the active world, in both cases large amounts of "tacit knowledge" (in the sense of Polanyi Polanyi (1967) are required by the system if the digital message shall be of any use (and the never-failing availability of this "tacit knowledge" was of course "taken for granted" when the message was first coined in DNA).
The invention of "digitality" I suggest was the step which some 4 billion years ago allowed certain swarms of communicating closed membrane systems floating in the prebiotic mud to escape the indifference of the mere moment and to enter a temporal world of genuine selfhood.
The tacit knowledge aspect of cellular (or organismic) activity is the strangely overlooked key to biosemiosis. I often get the feeling that genocentrism is only the surface of a yet more general cultural bias towards what I have called "digitalism", the preferential allocation of realness to digital aspects of the world, numbers and sequences (Hoffmeyer 1998a). However, as Michael Conrad has pointed out, for a biochemist the world consists of molecular shapes. Biological information processing is largely based on the recognition capabilities of macromolecules such as proteins and nucleic acids. And molecular shapes play an important role in these recognition processes. The biochemist's world of shapes does not easily mingle with the computer scientist's world of switches.
In principle, it might be possible to simulate enzyme-controlled systems with digital computers but as shown by Michael Conrad "the computational demands imposed by an adequate model would rapidly outstrip even the most powerful digital computers" (Conrad 1992). This is because "pattern recognition capabilities of enzymes cannot be characterized in terms of a well defined set of rate constants, but must instead depend on the detailed physics of the recognition processes, including the quantum nature of sub molecular processes." Thus at the sub cellular level information processing is still mostly analogical rather than digital, based on shapes rather than based on switches (Hoffmeyer 1997).
More than 25 years ago Pattee was fully aware of the importance of what I here call "tacit knowledge" when he wrote that "developmental controls in cells may be executed by 'ordinary' molecules to which we give titles such as activator, repressor, or hormone, but the control value of these molecules is not an inherent chemical property; it is a complex relation established by a collective hierarchical organization requiring the whole organism" (Pattee 1973). But how does the semiotic dynamics of this hierarchical organization, i.e., its capacity for exercising cellular and developmental control, enter our understanding of the dynamics of living systems? Here I would suggest that Pattee's distinction between a dynamic and a symbolic mode (Pattee 1977) were broadened to cover also those less highly developed forms of semiosis which C. S. Peirce called iconic and indexical sign processes. The idea of the concept code-duality was to fill in the no-mans-land left by Pattee's distinction, since it implies that even the dynamic mode is basically a semiotic mode although based on indexical or iconic reference (i.e., analog coding) rather than symbolic reference (i.e., digital coding).
An enzyme reaction inside a cell is not just an ordinary chemical reaction but is most often also a link in a topologically channeled chain or cascade of reactions coupling the individual enzymatic reactions into the other-referential activity of the cell membrane. The referential or semiotic dimension of this process becomes clear only when seen in the proper temporal perspective as a process, which is logically chained to the survival project of the lineage to which the organism belongs. "Tacit knowledge" of cellular and organismic architectures only makes sense in a historical perspective.
Let me illustrate this claim by the kind of interaction I have termed "semetic interaction" (Hoffmeyer 1995): a well-described case is exhibited by caterpillars munching upon the leaves of corn seedlings (Hoffman 1992; Krampen 1992). Here the munching process evokes an interpretant to be formed in the damaged leave cells in the form of a (unknown) molecular message that is carried around to all the leaves. Healthy leaves understand the message as an instruction to produce a certain volatile substance (a terpenoid) which is then emitted to the air. Wind will carry the terpenoid away and eventually it will be absorbed by certain wasps where it may then eventually provoke the formation of an interpretant in the form of muscular contractions destined to bring the wasp to the emitter of the volatile substance. When the wasp arrives here it senses the presence of the caterpillar and its body is induced to perform the chain of movements necessary for injecting its eggs into it. Upon hatching the wasp larvae eventually eat and kill the caterpillar.
Seen from above what happens is that the wasp and the corn seedling has a common, if opposite, interest in the caterpillar and that they have found a productive way to cooperate in this interest by entering a shared dimension of the semiosphere (Hoffmeyer 1996; Hoffmeyer forthcoming). Most interestingly however, wasps are not attracted by the terpenoid the first time they come upon it but have to learn how to interpret the substance. Or to state this differently, a regularity of nature, i.e., the corn seedling's emission of a certain terpenoid when attacked by caterpillars, has become a sign leading the wasp to the caterpillar.
Semetic interactions depend on a capacity for associative learning, which is typical to living system at whatever scale. That a gradient of terpenoid molecules refers to the direction that will lead a wasp to caterpillars is not understandable from pure biochemistry but requires the recognition of biochemical events as traces or measuring marks left by singularities of past conjunctions. Seen in this light evolution consists in the perpetual incorporation of a flow of singularities into receptive systems as referential marks. It may be true, that science cannot deal with singular, irreproducible events, but science certainly can, and should, study the preconditions for such singularities to occur as well as the consequences of these events for universal processes such as those causing evolutionary change.
It is important here to stress the interdependence of the analog and the digital as two equally necessary forms of referential activity arising like twins in the individuation of that logic which we call life: had it not been for digital coding there would have been no stable access to the temporal world, the unidirectional continuum of pasts and futures, and therefore there could have been no true agency or communication. On the other hand, had it not been for the analog codes there could have been no interaction with the world, no other-reference, and no preferences. To claim that only the digital twin is semiotic whereas the analog twin remains in the sphere of classical dynamics is to block the only possibility for transcending the semantic cut.
Code-duality and semiosis opens for a dimension of our world and its evolution, which is left, underdetermined by thermodynamics. Context space expands with an accelerating rate in proportion to the increase in semiotic sophistication of species; for, simply put, there are so many more different ways to be smart than there are different ways to be simple (may this be the reason why the speciation rate among mammals is five times higher than the speciation rate among lower vertebrates?)
Robert Root-Bernstein and Patrick Dillon seem to arrive at a somewhat similar position in a recent challenging and very thorough analysis of molecular complementarity as a key factor in the origin and evolution of life (Root-Bernstein and Dillon 1997). They use the term complementarity to mean "non-random, reversible coupling of components of a system" and explain that such complementarity could exist between the energy fluxes in a cyclical, or self-organizing thermodynamic system, but could have many other manifestations as well: "Molecular complementarity refers specifically to non-random, non-covalent interaction between molecules. Thus van der Waals forces, hydrogen bonds, pi-pi stacking bonds, ionic bonds, charge transfer complexes, and similar reversible chemical interactions can all be involved in molecular complementarity...Higher order levels of complementarity are also possible in living systems: cell organelles, cells, tissues, organs, and organisms all couple non-randomly. One can even think of symbiosis as being complementarity between organisms."
This complementarity refers to precisely the kinds of interactions, which I would describe as indexical semiosis or analogue sign action and although Root-Bernstein and Dillon do not themselves adopt a semiotically informed terminology it seems obvious to this author that their concepts of complementarity and coupling presupposes a semiotic setting of the systems under consideration. Their study therefore establishes a much-needed exploration of the creative power inherent in life's tacit or analogically coded knowledge. And here I guess do we find a way to deal with Pattee's "measuring problem": "Life originated with semiotic controls. Semiotic controls require measurement, memory, and selection, none of which are functionally describable by physical laws that, unlike semiotic systems, are based on energy, time, and rates of change" (Pattee 1997). Searching for the origin of referentiality will take us to the inside-outside asymmetry of membranes and the self-organizing chemistry of interaction across the membrane based on a growing potential for indexical reference in terms of molecular complementarity. The formation of symbolic reference in the form of replicative molecules grew out of the already established indexical referentiality. Digital codes grew out of analog codes, not vice versa.
The term autopoiesis has already been used for a different purpose and the term autogenesis carries confusing connotations, so I can perhaps be excused for using the rather ugly term selfication here. C. G. Jung used the term individuation to indicate the processes whereby a person comes to accept and understand his own individuality and in the same time his being one among others, but in general philosophy individuation refers to a process whereby an individual mode of existence develops out of the general. In this sense for instance Aristotle saw matter as individuating. What I have tried to describe here is the formation of life as a special kind of Aristotelian individuation, namely the formation of selfhood, or in other words a selfication.
Since human beings are prototype selves let us for a moment consider the evolution of human selfhood. In Signs of meaning in the Universe I argued at length for the view that language existed as a mental or conceptual resource a long time before it finally became spoken, probably only around 125.000 years ago (Hoffmeyer 1996, chapter 10 and 12). Inspired by Merlin Donald's idea of a "mimetic culture" among our Erectus ancestors (Donald 1991) I suggested that words were only slowly extracted from the syntactic structure of ritualized or otherwise standardized mimetic communication. The general picture was:
meaning --> mimed 'sentences' --> single words --> fixed phonemes,
Or in other words, from analog communication to digital. When small children learn to talk they seem to repeat this scheme to some extent when for instance single words are in the beginning used to denominate whole phrases (holophrastic speech). The appearance of digital conceptual communication presupposed and in the same time required the abstract idea of "not" from which derives the basic mental asymmetry of the self and the world, i.e., human self-hood, human love and fear.
The pattern of processes out of which life arose may have reflected the same general logic. As described here the process begins as an "interiorisation" in which the membranes build up an asymmetry between their excluded interiors and exteriors. Prebiotic membranes "choose to prefer" their insides from their outsides, or one might perhaps see this as a sort of colonization of the interior space (note 1). The membranes thus scaffolded themselves through the formation of a multitude of autocatalytic metabolic loops reflecting the available resource space (i.e. analogue mapping - or other-reference) and finally of replicative molecules mapping constituents of the internal autocatalytic system (digital mapping or self-reference). Thus persistent architectures appeared as entities engaged in the trick of conjuring up a virtual reality at their insides for the purpose of coping effectively with the their outsides. We get the following selfication scheme:
interiorisation --> 'communication' --> analogue maps --> digital maps
I find it more than a little interesting to note that Jean Baptiste Lamarck, who in 1802 coined the term biology to sort out those two kingdoms possessing "perfectibility" (i.e., evolutionary potential) from the non-"perfectible" mineral kingdom, had a scheme in his mind very similar to what I propose here. He saw "La marche de la nature" essentially as obeying the following scheme:
changed conditions --> new habits --> inner feelings --> new forms
In a modern vocabulary Lamarck's "inner feelings" (sentiments interieur), which he himself thought should be explained by the action of "subtle fluids" (like those that at the time were supposed to cause magnetism) would be nothing but self-organization. It was the Darwinians who a century later identified evolution narrowly with genetic changes and thus condemned Lamarck for having a false theory of evolution because he held a false view of heredity. He certainly did hold a false view of heredity, like everybody else did at his time, but he had greater things on his mind: the idea of 'perfectibility' that the more perfect grew out of the less perfect. If the unfolding of semiotic freedom is in fact what evolution is all about (cf. Hoffmeyer 1996) he was in a way right there - if, that is, we consider Beethoven's string quartets to be more perfect in the sense of exhibiting more semiotic freedom than the cries of macaques or the cAMP concert of slime moulds.
1. One should not overlook the formal analogy between the prebiotic membrane and the proto-human "not". Both served to generate a kind of self (biological or mental) by a process of alienating an "internal" core from it's "external" source. The myth of the lost paradise springs to mind.
Bateson, G. 1972. Steps to an Ecology of Mind. (Ballantine Books, New York).
Chandler, J., and Vijver, G.v.d., eds. forthcoming. Closure. Emergent Organizations and Their Dynamics (New York Academy of Science, New York).
Christiansen, P.V., 1999, "Downward Causation. From Macro- to Micro-levels in Physics". In Downward Causation, edited by P.B. Andersen, N.O. Finnemann and C. Emmeche. (Aarhus University Press, Aarhus), forthcoming.
Conrad, M., 1992, "The Seed Germination Model of Enzyme Catalysis". Biosystems 27, 223-233.
Donald, M. 1991. Origin of the Modern Mind. Three Stages in the Evolution of Culture and Cognition. (Harvard University Press, Cambridge).
Hoffman, M., 1992, "The Enemy of my Enemy is my Friend". American Scientist, november-december, 536-537.
Hoffmeyer, J., 1995, "The Swarming Cyberspace of the Body". Cybernetics & Human Knowing 3(1), 16-25.
Hoffmeyer, J. 1996. Signs of Meaning in the Universe. In Advances in Semiotics, edited by T.A. Sebeok (Indiana University Press, Bloomington, IN).
Hoffmeyer, J., 1997, "Semiotic Emergence". Revue de la Pensée d'aujourd'hui 25-7(6) 105-17 (in Japanese, English version available upon request).
Hoffmeyer, J., 1998a, "Life: The Invention of Externalism". In Emergence, Coplexity, Hierarchy, Organization. Selected and Edited Papers from the ECHO III Conference. Acta Polytechnica Scandinavica no. 91, edited by G. Farre and T. Oksala. (Espoo, Finland), pp. 187-196.
Hoffmeyer, J., 1998b, "Surfaces inside Surfaces. On the Origin of Agency and Life". Cybernetics & Human Knowing 5(1), 33-42.
Hoffmeyer, J., forthcoming, "Seeing Virtuality in Nature". Semiotica.
Hoffmeyer, J., and Emmeche, C., 1991, "Code-Duality and the Semiotics of Nature". In On Semiotic Modeling, edited by M. Anderson and F. Merrell. (Mouton de Gruyter, New York), pp. 117-166.
Krampen, M., 1992, "Phytosemiotics Revisited". In Biosemiotics: The Semiotic Web 1991, edited by T.A. Sebeok and J. Umiker-Sebeok. (Mouton de Gruydter, Berlin), pp. 213-220.
Parkinson, J.S., and Blair, D.F., 1993, "Does E. Coli Have a Nose?". Science 259, 1701-1702.
Pattee, H., 1973, "Discrete and Continuous Processes in Computers and Brains". In The Physics and the Mathematics of the Nervous System, edited by W. Guttinger and M. Conrad. (Springer Verlag, New York).
Pattee, H., 1977, "Dynamic and Linguistic Modes of Complex Systems". International Journal for General Systems 3, 259-266.
Pattee, H.H., 1982, "Cell psychology: An evolutionary approach to the symbol-matter problem". Cognition and Brain Theory 5(4), 325-341.
Pattee, H.H., 1996, "The Problem of Observables in Models of Biological Organizations". In From Evolution, Order, and Complexity,, edited by E.L. Khalil and K.E. Boulding. (Routledge, London).
Pattee, H.H., 1997, "The Physics of Symbols and The Evolution of Semiotic Controls". In Santa Fe Institute Studies in the Sciences of Complexity, Proceedings Volume. (Addison-Wesley, Redwood City, CA).
Polanyi, M. 1967. The Tacit Dimension . (London).
Popper, K. 1990. A World of Propensities. (Thoemmes Antiquarian Books, Bristol).
Root-Bernstein, R.S., and Dillon, P.F., 1997, "Molecular Complementarity I: the Complementarity Theory of the origin and Evolution of Life". Journal of Theoretical Biology 188, 447-479.
Rosen, R. 1991. Life Itself. A Comprehensive Inquiry into the Nature, Origin and Fabrication of Life. (Columbia University Press, New York).
Salthe, S. 1993. Development and Evolution. Complexity and Change in Biology. (MIT Press, Cambridge, Mass./London).
Salthe, S., 1998, "Naturalizing semiotics". In Semiotics in the Biosphere: Reviews and a Rejoinder. Special Issue of Semiotica, 120(3/4), 381-394.
Uexküll, J.v., 1982 [1940], "The Theory of Meaning". Semiotica 42(1), 25-87.
Ulanowicz, R.E. 1997. Ecology, the Ascendent Perspective. In Complexity in Ecological Systems edited by T.E.H. Allen and D.W. Roberts (Columbia University Press, New York).
Wilden, A. 1980. System and Structure. (Tavistock, New York).