Make your own free website on


The Myth Of The Selfish Replicator

As underscored above, living things are autocatakinetic systems that produce components by replication, and replication never occurs outside the context of the autocatakinetic system plus its nonequilibrium environment as a whole. More particularly, reviewing briefly, replication is used by living things for the synthesis of larger molecular components, or polymers, from smaller molecular components, or monomers (usually consisting of fifty atoms or less) with the principal cellular polymers produced in the autocatakinetic cycle being proteins and nucleic acids. In both cases the production of polymers proceeds with the sequential addition of monomers according to a DNA template so as to create a linear string which in the case of proteins spontaneously folds into a three-dimensional functional structure. The functional form of nucleic acids, of which there are two main kinds, deoxyribonucleic acid (DNA), and ribonucleic acid (RNA), remains a linear polymer. Proteins are built from monomeric amino acids of which there are twenty different kinds, and DNA and RNA are both built from four different kinds of monomers called nucleotides (e.g., see Darnell et al., 1986).

In the production of proteins, a complementary RNA copy (mRNA, or messenger RNA) is first made from the DNA template ("transcription") and then amino acids are strung together on ribosomes in a different part of the cell according to its nucleotide sequence ("translation"). Typically each amino acid is specified by a particular sequence of three RNA nucleotide bases (a "codon"). Following the discovery of DNA structure and its relation to the component production in cells, as a consequence of elucidating this translation process, it became popular in the 1950s to say that DNA, or the genes that its linear sequences constitute, "codes" for proteins, and it also became popular to talk of DNA (or genes) as constituting a "blueprint", coded instructions, or "program" for building the living things that contained it. The smell of autonomous agency became stronger when, because the DNA template in a cell is used as a template for the replication of DNA itself, DNA came to be called a "self-replicating" molecule. The idea that living things are built out of passive matter by DNA programs, that are also self-replicating, leads to the neo Pythagorean selfish replicator (or algorithm) theory of Dawkins and Dennett.

Putting aside the problem of dualist interactionism that fatally wounds all such schemes, living things do not contain blueprints or programs in any ordinary sense of the words, and the putative "replicators" of Dawkins and Dennett are a myth-the result, as Levins and Lewontin have correctly asserted, of fetishism and reification (see also Fleischaker, 1990, and Goodwin, 1982). DNA molecules in cells constitute a very particular kind of cellular component that along with proteins, among other things, are used as part of the end-directed autocatakinetic component producing system as a whole. The entire cell-environment system is end-directed and active, but if there is any part of it that is static, inert, or inactive relative to the rest of the system it is the DNA or genes. The myth of the replicator has been pointed out by various critics of the idealist reductionism of selfish gene theory. The DNA of living things, as Goodwin (1994, p. 35) has expressed in straightforward terms is simply "not self-replicating; it is not an independent 'replicator'". Contrary to the popular conception, writes Lewontin (1992, p. 48), "that genes make proteins, and that genes are self-replicating...genes can make nothing." Component production or replication, including the replication of DNA, is a function of the entire autocatakinetic system as a whole and not of a particular molecule in the cell. Genes "cannot make themselves any more than they can make a protein," and by referring to them "as self-replicating," continues Lewontin, "we endow them with a mysterious, autonomous [and illegitimate teleological] power" they simply do not have (see also Swenson, 1990a).

Not only is it the case that DNA is not a self-replicating molecule, that, like proteins, is replicated and produced as part of the component production process of the autocatakinetic system as a whole, but within this system it is "proteins," as Darnell et al. (1986, p. 107) have written in their authoritative text, Molecular Cell Biology, that "are the active working components" of the cell, not the DNA molecules. It is proteins, in the form of polymerases, that carry out component replication or synthesis of both protein and nucleic acids. In addition, while it is true that DNA provides a crucial template used in the component production process, it is nothing like a blueprint or set of instructions or program for building a living thing. It does not even specify the three dimensional coordinates of a single protein let alone provide instructions as to how to put proteins together to form a living thing or even a part. Even the linear sequence of amino acids that the template does specify is typically incomplete and requires modification by the cellular dynamics, by proteins (e.g., by cutting and splicing) after it is released from the ribosomes.

To say that the DNA contains a blueprint, recipe, or program would completely change the meaning of these terms in the usual (artifactual) context in which they are normally used. It would be as though instead of what is usually thought of as a blueprint (typically explicit instructions for building a house, or other structure), an architect furnished only a list of raw materials to the contractor or client. What would be missing would be the instructions for assembling the materials into a functional three-dimensional product, a subject on which the DNA in cells is entirely mute. In the case of living systems this is provided by the cellular dynamics of autocatakinetic systems, self-organizing systems that, in contrast to artifactual ones, assume their dynamic forms following from initial conditions and laws, principles of self-organization, and not, in general, from rules or any explicit representation or set of instructions. What autocatakinetic systems in general do is utterly fantastic in the context of artifactual or rule-based (algorithmically based) systems: they not only build, repair and maintain themselves, but, in the case of living things, go out and seek the raw materials to keep this dynamic process going.

The Function of DNA (Like The Words On This Page) Depends On Its Inactivity

"DNA", as Lewontin (1992, p. 33) has written, "is a dead molecule, among the most nonreactive in the living world. That is why it can be recovered in good enough shape to determine its sequence from mummies..." Biological viruses, which evolved as part of the complex process of gene exchange by which the prokaryotic world developed and has maintained, the Earth's biogeochemical cycles (the price of keeping the system loose and adaptable), are simply a dead, and not very interesting, "mixture of chemicals", as Margulis and Sagan (1995, p. 24) have put it, without an environment of active end-directed, replicative autocatakinetic systems. Likewise, computer programs or algorithms, like the words on this page, are dead and meaningless without the autocatakinetic cultural systems of which they, and the computers that run them, and other artifacts are component productions.

Finally, as discussed more fully below, replicative ordering provides the means for the world to build dynamical systems able to access otherwise inaccessible space-time dimensions, in particular, to build dynamical order from discontinuously located potentials (Swenson, 1991b; Swenson & Turvey, 1991). Towards this end replicative systems, whether living systems in general or cultural systems (second, and third-order autocatakinetic systems respectively), require "a set of internal constraints [such as the words on this page or DNA strings] that are discrete, sequential, and rate-independent relative to the rest of the autocatakinetic cycle" (Swenson & Turvey, 1991, p. 343). This insight was first expressed by Polanyi (1968) who noted that the "order [of the linear sequences of a DNA molecule] is not due to the forces of potential energy", or, more accurately, as we understand it now, to local potentials. The order of the sequences of such constraints, as Polanyi pointed out, such as the words on this page or the sequence of base pairs in a DNA string is arbitrary with respect to local potentials. Namely, the strings or sentences of equal lengths, for all practical purposes, take the same amount of potential energy to produce ("write") or use ("read").

The major point of relevance here is that the function of DNA, like the letters or words on this page, depends precisely on the local thermodynamic or energetic equivalence of the sequences. In different terms, the function of DNA, like the function of the words on this page, as with all other rule-based systems (e.g., recipes, blueprints, programs, or algorithms) depends on the fact that any sequence can be exchanged with any other without any local energetic consequence. It is the stability of the DNA molecule, in other words, regardless of the sequence-its inertness, or utter inactivity, or "deadness" on which its function crucially depends, but, alas for the reductive idealists, it is this property of deadness or inactivity which also disqualifies it a priori from the role of active agent that they would like to attribute to it.