Origin of life Studies
Messages and Molecules:
What is the Essence of Life?
DAVID L. WILCOX
St. Davids, PA 19087
The concept that the nature of chemical reactions can form an adequate basis to explain or describe the essence of life is examined and rejected. The proposed alternative suggests that the real essence of life is the information or patterns which, although carried on DNA, are not determined by it. Several implications of the concept of informational entities are then explored.
Found in Perspectives on Science and Christian Faith 40 (December 1989) 227.
The Argument For a Chemical Essence
The initial premise of this paper is that the more completely we understand biochemistry, the more unlikely it seems that such chemistry represents the "essence" of life. Although such a statement runs counter to most current thought, I contend that it is amply justified by recent developments in genetics.
Nineteenth-century materialists considered life an emergent phenomenon of simple inorganic reactions, a necessary product of the laws of chemistry. Life was thus thought to be a predictable outcome of highly probable reactions. Today, the materialistic conviction of a "chemical essence" remains strong. Biological chemistries are known to be elegant and precise in both place and time, parts of highly determined and exactly governed systems of reactions. Nevertheless, the boundary conditions for the reactions are set and controlled by specific chemical constraints. No maneuvering room seems left for an elan vital, a soul, or the body as a whole. All life is chemically governed chemistry. Has the molecule preempted the seat of the soul?
In the same fashion, developmental biology increasingly points to the molecule. Bodies unroll from single cells in a pattern dependent on the distribution of chemicals in the cytoplasm of those initial cells. The early molecular patterns set the initial patterns of tissue cells, and in turn, interactive movements of those early tissue layers shape the organs and thus the organism. Morphogenetic movements and cell differentiation are controlled by gradients of inductive chemicals produced by tissues. Has development also been reduced to chemistry? Are bodies simply epiphenomena of embryonic chemistries?
Some efforts to define the essence of humanity have also looked to biochemistry. Human and non-human chemistries are often almost identical. Histological chemistries, as used in transplant tissue typing, show more than a ninety-eight percent overlap between human and chimpanzee. This means that biochemical differences within human populations are often greater than the average biochemical differences between the two species. If there is no distinguishable difference in our chemistries, our cells, tissues, and organ structures, are humans to be thought of as a rather odd sub-species of chimpanzee?
The Argument Against a Chemical Essence
Each of the above arguments contains seeds for its own dismissal. Precise control of the boundary conditions of biological chemistries requires a complex system of "top-down" control and structure. Even a quick look at complete biological entities reveals a hierarchy of such physiological or cybernetic constraints. Thus, whole-body norms set organ-function norms (boundary conditions), organs set cellular norms, and cells set the norms for their chemistries. Although chemical agents are used to control chemistries, the agents simply enforce the commands of the body as a whole, as transmitted down through its levels of structure. If the essence were truly chemical, high-level structure would follow patterns which arose from and imaged the fundamental characteristics of the chemicals themselves. But they do not.
The essence of the developmental "unrolling" process is that it follows a previously existing program. The initial molecular template was created by a holistic pattern which the fertilized ovum carries; a pattern which not only describes normal biochemistry, but also the norms for all levels of structure. (Since the new zygote ova will self-organize in vitro, it is evident that the fertilized ovum carries an ideal body pattern rather than being structured by the maternal adult body.) Even "preformed" embryos with deterministic cleavage are "fate-mapped" products of a complete pattern, a pattern which includes the path and goal of development as well as error-correction mechanisms.
In the example of human and chimpanzee chemistries, to say that there is no significant chemical difference is not to say that the two primates are identical, but rather that their differences are either not significant or are not chemical in nature. If their differences are significant, it is simple logic to say that their differences must be located elsewhere. To state that chemical identity means total identity is to assume the point to be demonstrated, that the essence of biological form is in the chemistry. One must already believe in chemical determinism.
But where then is the source of the differences? Where are holistic patterns stored and read? Does this logic require an immaterial elan vital, or soul, or can the DNA hold all levels of pattern? The latter is usually assumed to be true. Would such a concession be a return to chemistry via another route? Human and chimpanzee total DNA are almost identical. Thus, how can the chemical nature of DNA be considered the essence of life?
The Medium is NOT the Message for Life
If DNA chemistry is not to be considered the essence of life, what is? Obviously the total pattern of living systems must be explained. Could one say that the essence is the pattern carried on the DNA, rather than the DNA itself? Or, is this simply quibbling, a way to slip chemical determinism in the back door? I think not. The distinction being drawn is the difference between the medium and the message, the transmission vehicle and the entity transmitted.
A few definitions will clarify this point. A medium is a channel by which information can be stored and/or transferred. A system is able to act as a medium because it can exist in several possible states, which may collectively be termed its ensemble. Any such system can be measured for its potential as a medium, its capacity to store and carry information. However, it becomes a medium only if its ensemble of various states is converted into an alphabet by being given arbitrary assignments of symbolic meaning (as in Morse code or the English alphabet). A defined ensemble therefore becomes an alphabet through which one can encode a message. Note that the content of such a message will be independent both of the nature of the medium and of the code-word assignments. However, its expression will be dependent on them. Thus, the same paper and ink, and the same letters, can be used to encode Spanish or English, and the same words (in either language) can be used to encode either the Scriptures or pornography. Given the coding definitions, the message shapes (rules) the sequence of the letters.
Such a distinction can be effectively applied to DNA and the "message" of life. As an illustration, consider the following thought-experiment. At the present time, the complete DNA sequence of certain bacterial viruses (phages) has been determined and published. Using the published genome sequence of phage Phi Chi 174, for example, one could program a "gene machine" (automated DNA synthesizer) with the phage DNA sequence. After obtaining a complete product, a naked strand of DNA, one could inject it into the bacterium E. coli. The DNA would take over the bacterial cell and cause the production and release of infectious phage particles (viruses). The phage placed in the first bacterium was their ancestor, but did it (the infecting phage) have an ancestor? Was there an ancestral virus? Of course. The pattern of information taken from a scientific paper was the ancestor. The printed phage was on the printed page. What was the essence of the phage's life? Not DNA, but a medium independent pattern! (It may be objected that viruses such as a bacteriophage are not truly alive, since they depend upon host cells for their synthesis. The point of the argument, however, is not the exact status of viruses, but rather that all living things are the realization of patterns stored on the DNA.)
The objection may be raised that DNA encodes information only for how to make proteins; thus, although a message, it is still chemical in essence. The inadequacy of such an argument should be obvious. Genomes do not produce mere proteins; they produce a hierarchy of cells, organs, and organisms made of protein. Thus, other patterns (information) must also be written there. To some extent, this is true even for a bacteriophage. The folding pattern for packaging its RNA "chromosome" is stored on the genome of an RNA virus, as well as the information to make its needed proteins.
The easiest way to approach this question is to evaluate a human language such as English. Such languages contain a hierarchy of alphabets. At the first level, written letters are assigned phonetic definitions. Sequences of such sounds (words) are then assigned second-level definitions as objects, actions, etc. In turn, sequences of words are utilized to encode real-time events such as, "I ate the bread." Finally, events themselves may be assigned a symbolic meaning, as in the Lord's Supper which "shows the Lord's death until he comes." Note that at each level, a single "word" may be assigned to represent the increasingly complex object. In the example given, the highest level is represented by the term Eucharist.
In the same fashion, descriptions of specific chemicals encoded on the DNA may themselves become letters used to write more global descriptions. In order to do so, to use a lower level of message as an ensemble and medium, the encoding conventions must themselves be written out on the medium. This is necessary because they must be available for us in decoding (reading) messages written in that language. Without definitions, a message is indistinguishable from a random sequence. For example, consider the code sequence for the 20 different types of amino acids used to make proteins. Each is defined at the protein level. These definitions are in the form of large proteins which can recognize both the code words (tRNA paracodons) and the amino acids which they represent. The amino acid code is therefore used to encode and thus describe (the protein definitions of) the amino acid code itself.
At the next level, recognition codes (enhancer sequences) for more complex structures are tagged to the descriptions of the proteins needed to make them, and new proteins are described (homeobox genes) which will be able to read the new recognition codes. One more level, and the recognition proteins are themselves tagged with a set of recognition codes, and so on. What this reveals is a complex series of hierarchically encoded languages, with independent messages written at each level. Presumably, each level of the organic structural hierarchy would be described at a specific linguistic level of that information hierarchy.
In this complex mesh of information, the expression of each level of coding is controlled by the level above it. By analogy, in English the event to be described dictates the words to be used, and they in turn dictate the letter sequence. Likewise, the DNA information concerning a specific protein's structure is used only because those particular proteins are the appropriate components needed to build some larger structure. Each level encoded enforces boundary conditions on the level below it. Governance thus moves from whole-body pattern to cell pattern to molecular pattern rather than the reverse. Since higher-level norms govern lower-level activities and structures, the essence of life is the total organismic pattern, not the component biochemistries. Life is first a resident of a world of information before being embodied in physical creatures. The appropriate metaphor is language or computer programs, not the machine. Chemistry is about as relevant to understanding a whole organism as an analysis of the plastic of a record album would be to understanding a recording of a Brahms concerto.
Implications of an Informational Reality
The concept of living essences as patterns of information raises a variety of questions about humans and other living things. For example, an informational pattern might be considered a sort of encoded elan vital. This would imply that informational patterns might be real entities, living patterns engaging in specifically controlled observable activities during development, etc. By extension, such coherent patterns could exist above the organismic level. A tissue cell expresses only a small part of the larger body pattern which it contains. Might not an organism be expressing only a small part of a population level pattern which it contains? In its simplest case, both males and females carry the full pattern for both sexes. Thus, an entire population could be viewed as a coherent entity, an obedient creature responding to its Creator's commands. More precisely, the population-level pattern expressed in individual organisms could be so viewed. The "command" addressed to the population could be expressed as: increase and fill both ecological and morphological space.
The concept of living things as systems specified by coherent patterns would also change the complexion of natural selection. The total pattern carried by a population apparently is a "hyperspace" of alternative functional morphologies. A specific organism would be canalized (specialized), embodying a certain location in that hyperspace, but yet could be carrying in its genome a considerable variety of alternative morphological states. For example, there is some suggestion in the literature that there might be a single avian morphological package, variously expressed in different species. If so, directional selection (or explosive radiation, not to speak of mutation) could be viewed on the population level as a "deliberate" exploration of morphological hyperspace, looking for morphologies better matched to available niche space. The "command" is to increase and to fill both spaces. Individual organisms would act as sensory probes for such a population-level pattern. Selection would indeed be a process by which environmental information was collected, but the internal pattern would act as the collecting agent. Note that this is not group selection, nor is it neo-Lamarkian. Rather, it is organism-level selection by the population-level pattern. This proposal leaves open, however, the question of whether a process that collects and condenses environmental information is capable of building new morphological map structure. In other words, how does selection produce new morphologies?
The concept of linguistic levels also clearly defines a major difficulty with models of the origin of life. Not only must initial organisms (presumably simple cells) have their pattern encoded onto a replicable medium, they must also "know" the decoding conventions (definitions) to read that pattern. Cells can only "know" the definitions of the code words, however, if they are written down in the very language that the definitions describe. Thus, the central quandary: How is a message to be generated prior to the existence of the language used to speak it? The same question reappears again at each level of language encoded onto DNA. Thus, how are one hundred new bauplans generated in the Cambrian if organismic-level language is not yet encoded? But if it is, where are the organisms which were encoded in it?
Lastly, the idea of humanity being in "the image of God" is obviously a patterning concept, an informational definition. It seems clear that such imaging would be identified as message rather than medium or ensemble. The question therefore arises, at what biological linguistic level would God's image be encoded? In which language can the image be spoken? How many levels of structure and/or information would act simply as the medium used to carry the divine pattern? If high level structures themselves can form ensembles of code words for still higher level languages, could some characteristics seen in chimpanzees be used as a medium to write such an image? Are such characteristics only language-symbols, words which can be assigned arbitrary definitions and used to encode a higher level message? Or are chimpanzees a different story (message) told at the same level as our own story? Still, "In the beginning was the Word..."
Antonovics, J. 1987. The evolutionary dyssynthesis: Which bottles for which wine? The American Naturalist 129(3):321-331.
Belyaev, D.K. 1979. Destabilizing selection as a factor in domestication. The Journal of Heredity 70:301-308.
Davenport, R. 1979. An Outline of Animal Development. Reading, MA: Addison-Wesley Publishing Company.
Dhouailly, D. and P. Sengal. 1973. Interactions morphogenes entre L'epiderme de reptile et de derme d'oiseau ou mammifere. C. R. Acad. Sci. Ser. D. 277:1221-1224.
Eldredge, N. 1985. Unfinished Synthesis. Oxford: Oxford University Press.
Eldredge, N. 1986. Information, economics and evolution. Annual Reviews of Ecology and Systematics 17:351-369.
Erwin, D. 1987. A comparative study of diversification events: The early paleozoic versus the mesozoic. Evolution 4(6):1177-1186.
Grene, M. 1987. Hierarchies in Biology. American Scientist 75:504-510.
Gould, S. 1980. The promise of paleobiology as a nomothetic, evolutionary discipline. Paleobiology 6(1):96-118.
Gould, S. 1980. Is a new and general theory of evolution emerging? Paleobiology 6(1):119-130.
Gould, S. 1985. The paradox of the first tier: An agenda for paleobiology. Paleobiology 11(1):2-12.
HampÈ, A. 1960. Le competition entre les elements osseux du zeugopode de poulet. J. Emb. Exp. Morph. 8:241-245.
King, M. and A. Wilson. 1975. Evolution at two levels in humans and chimpanzees. Science 188:107-116.
Kollar, E.J. and C. Fisher. 1980. Tooth induction in chick epithelium: Expression of the quiescent genes for enamel synthesis. Science 207:993-995.
Lawlor, D., F. Ward, P. Ennis, A. Jackson and P. Parham. 1988. HLA-A and B polymorphisms predate the divergence of humans and chimpanzees. Nature 335:268-271.
Levin, R. 1988. A lopsided look at evolution. Science 241:291-293.
Mueller, B. 1986. Effects of skeletal change on muscle pattern formation. Bibliotheca Anatomica 29:91-108.
Parsons, P. 1987. Evolutionary Rates under Environmental Stress, in Evolutionary Biology, Vol. 21. New York: Plenum Press.
Raff, R. and T. Kaufman. 1983. Embryos, Genes, and Evolution: The Developmental-Genetic Basis of Evolutionary Change. New York: Macmillan Pub. Co., Inc.
Salthe, S. 1985. Evolving Hierarchical Systems. New York: Columbia Unviersity Press.
Stanley, S. 1979. Macroevolution: Pattern and Process. San Francisco: W.H. Freeman.
Valentine, J. and D. Erwin. 1987. Interpreting great developmental experiments: The fossil record, in Development as an Evolutionary Process. Edited by R. Raff and E. Raff. New York: Alan R. Liss.
Volkenstein, M. 1987. Punctualism, non-adaptationism, neutralism and evolution. Biosystems 20:289-304.
Vrba, E. and N. Eldrege. 1984. Individuals, hierarchies and processes: Towards a more complete evolutionary theory. Paleobiology 10(2):146-171.
Vrba, E. and S. Gould. 1986. The hierarchical expansion of sorting and selection: Sorting and selection cannot be equated. Paleobiology 12(2):217-228.
Wallace, A. 1984. Mechanisms of Morphological Evolution. New York: John Wiley and Sons.