Use of the term "phenocopy"
Antoine Danchin 唐善 • 安東
Slightly modified from a text published in Language
and Learning, The debate between Jean Piaget and Noam Chomsky
Harvard University Press, 1980
This text, resulting from a discussion between Noam Chomsky and Jean Piaget, at a meeting organized by the Centre Royaumont pour Une Science de l'Homme, of which Jacques Monod had asked me to become one of the coordinators, discusses the concept of phenocopy much used by Piaget. The idea of assimilation of the environment into the genotype (genetic assimilation) supposed that there was a directed action of the environment on the genotype, and Piaget, who had studied the Limnaea of the Lac Léman in his thesis in Zoology thought that he had demonstrated an explicit instructive role of the environment. Today, we know that this is not so, even in the case of « adaptive » mutations. What happens is that living organisms are constructed to behave as information traps, so that they can use contingent fluctuations in the genotype and retain them when they bring about information which is significant in the context of a specific environment.
An experiment describing how phenocopies can be used to look for specific mutants is reported in:
Insofar as confusion still seems to persist in the use of some terms, I believe it necessary, in speaking of Piaget's metaphorical treatment of the concept of "phenocopy," to review the definitive and explanatory contribution of molecular biology to this concept in order to avoid any future misunderstandings.
Actually the whole opposition that characterized the biology of the Stalinist period between Lysenko and the proponents of Mendelian neo-Darwinism is involved here. Indeed, in Lysenko's view, the phenotype can exercise a considerable influence on the genotype, for "in natural selection . . . heredity, variability, and capacity of survival (autoregulation) of organisms are always included," and this autoregulation ". . . an elective faculty of organisms, organs, and cells, is the result of the historical adaptation of earlier generations to the conditions of their external environment . . . Thus, change in the nature of a living organism is due to change in the type of assimilation, in the type of metabolism. External conditions, once absorbed and assimilated by the living organism, are no longer external but internal conditions; they become, in other words, elements of the living organism, and require for their growth and development both food and the conditions of the external environment such as they were in the past. The living organism is, in a way, composed of elements of the external environment that it has assimilated." One could continue Lysenko's argument against the innate nucleus of the "Morganists "* by calling on the necessity of equilibration in living organisms. Let us mention at this point that Piaget's description of the evolution of a Limnaea  is strictly parallel to that of Lysenko for wheat . Now, this whole attitude is the result of a profound ignorance of the role of chance** and variability at the level of the genome, where it produces families of variations capable of interacting differently with the external environment. Let us return, for example, to the notion of phenocopy used in a metaphorical fashion by Piaget (see ) and offered as a reference leading to an analogy with the constructivist hypothesis. Although it may have been possible, before the existence of molecular biology, to believe in an "instructive" or "creative" principle that would explain the determination of traits in a living organism, producing an adaptive phenocopy (similar to Lamarck's notion of the inheritability of acquired characters), this point of view is today merely an episode in the history of ideas.
Let us be more precise. Until about 1940, what genetics had mostly accomplished was to enable us to define the existence of a fixed heredity preserved in complex chromosomal matter. Most geneticists had therefore realized that the influence of the environment exerts itself only in the expression of the genome and not in its identity: mutations occur before the selective event. A latent controversy persisted, imbued with a certain interpretation — which was inexact, moreover — conceming Lamarck's ideas on this effective and stochastic preexistence of mutations. Following the particularly simple and conclusive experiments of Luria and Delbrück , which proved this absolute influence of chance**, observations were eventually substantiated by the discovery of the material carrier of heredity.
Around this time, in fact, the relationship between genetics and biological chemistry became more precise: Morgan, Ephrussi, Beadle, and Tatum arrived at the simplified conclusion of the central correspondence "one gene/one protein." The relationship between the individual and its environment became clearer, and the distinction between genotype and phenotype acquired precision; the discovery of the structure of DNA enabled research scientists to provide a material support to the principal characteristic that crystallized around the genotype, namely, the genetic program .
This central concept — the abstraction of all observable data linked to the genotype — was to have only a limited usefulness as long as this abstraction did not permit researchers to find a direct relationship with phenotypic expression. It is in fact this lack of relationship that led Lysenko — because he observed an enormous phenotypic variability in plants (sensitivity to cold, practice of vernalization) — to reject completely the concept of genes.
The discovery of the concrete basis for genes, and especially the clarification of mechanisms regulating genetic expressions (in particular those that bear on the stepwise processing of hereditary information from the sequences of DNA nucleotides to the proteins) was to give meaning to the concept of a genetic program, for these regulations introduce relationships of order between the various elements of information contained in the genes. These order relations are then revealed during the temporal expression of the genetic program. They can lead therefore, in the case of a given program, to a plethora of particular outcomes, because the number of possible combinations rapidly becomes enormous. Since the discoveries of Jacob and Monod , it is clear that the genes corresponding to the regulatory functions will be responsible for phenotypic variability, especially for the apparently ideal adaptation of a living organism to its environment, and more generally for all the genotype-bound manifestations as they unfold in time.
Thus we are led to consider three particular characteristics that allow for the representation of living beings: a program, which summarizes the hereditary constraints and is, in fact, the abstract notion underlying the definition of all individuals in a given class (species); an initial state of the system  which represents the context in which the program must express itself at the time of the individual's birth; and a particular outcome or concretization of each program, which coincides with individual development. The set of all concretizations constitutes the genetic envelope and allows the essential characteristics of the program to be obtained by induction. Theoretically, a structured set equivalent to the aforementioned one could be reconstructed by providing the program, the initial state, and the totality of the events of the external environment, producing any interaction whatever with the individual.
The very definition of a species, at that, corresponds to the intersection of a particular family of individual concretizations; it is thus only a global and therefore precarious approximation, insofar as the program may not have been grasped to the full extent of its implications (if the individuals in question, say, were part of an environment having exceptional characteristics). Thus some species have been confused with others because, in a given environment, individuals looked remarkably similar, and it is only after observing a particular variability of one class of individuals in relation to another that one can draw the line between such classes. Let us note, moreover, that this consideration easily supplies an explicative model of some aspects of the evolution of species: a family of regulator genes, for example, in an unusual and highly specialized environment for that species, allows for the production of a "phenocopy" that is rather distinct from the usual parental phenotype (an anaerobic environment, for example, for individuals generally accustomed to an aerobic life, but which have the option of becoming anaerobic. In such cases, after many generations, one often finds: (1) an invariant type whose phenotype always mirrors the stable phenocopy, as well as (2) the original type whose descendants revert to their standard expression as soon as they return to the usual environment. This fact often led to the belief that one had demonstrated an instructive effect of the environment which led in some way to making acquired traits hereditary, whereas in fact one was witnessing a simple degeneration of the initial type which had lost the regulatory aptitudes that allowed it to change its phenotype according to the environment and had only retained one aspect. This aspect is that which is adapted to the specialized environment in which a collection of individuals happened to be located as they moved about randomly. This loss could take place without damage only because the environment in question remained constant long enough; there was no selection of the fittest in any general sense, but only a preservation of all those individuals which could manage to carry on in such a peculiar external environment, including those which, through the randomness of mutations, had lost one of the adaptive properties of the original species.
The general regulatory patterns affecting the expression of one's genetic endowment, such as they have been known for a few years, are probably sufficient to describe a very large number — if not all — of the properties of individual phenotypes without allowing the intervention of even the least instructive notion on the part of the environment. The adaptive properties seem to arise simply from the fact that at every moment the program supplies a possible choice among various interactions (due to various molecular movements and fluctuations of form and position) and also from the fact that the laws of thermodynamics lead to the selection of the most stable, hence of the most durable, options which, through appropriate amplifying mechanisms, yield suitable counterparts in the overall structure and functioning of the individual. Thus the environment makes a systematic selection of individual characteristics which are precisely those that best suit the environment itself, taking into account, of course, the heavy constraints of the program — a fish can live in a rather large number of aqueous environments but usually not in the air!
This brief review has, of course, considered only a very simple phenotype, determined by a single regulator gene, but the coordinated expression of the sum of the components in the individual phenotype is the result of a trade-off between one (or several) regulator genes and one (or several) environmental characteristics ("sum" refers here to general behavior, morphology, or metabolism). Thus, a phenocopy is in no way a construction but simply a particular realization of a given program according to a strict determinism: there is neither preformation nor acquisition, but only diachronic expression. The realization will take place at a different level depending on whether one considers individual phylogenesis, ontogenesis, or epigenesis. Since mental representation is the ultimate stage, it already represents the result of the evolution of species, the diachrony of cellular differentiation and individual development, and finally the diachrony of the epigenesis of the central nervous system.
Notes and Bibliography
* A derogatory term used by Lysenko to designate geneticists (of whom
T. H. Morgan was one of the most famous representatives).
** Contingency would be preferable to indicate meeting of independent deterministic causes, rather than chance
1. T. Lyssenko, Agrobiologie, Éd. de
Moscou, 1953, p. 171. ; Ibid., p. 242.; Ibid., p. 404.
2. See Jean Piaget Biologie et Connaissance, op. cit., p. 416-421.
3. S.E. Luria et M. Delbrück, Mutations of bacteria from virus sensitivity to virus resistance Genetics, 28, 1943, p. 491.
4. G.S. Stent, Molecular Genetics. An Introductory Narrative, San Francisco. W. H. Freeman, 1971.
5. F. Jacob et J. Monod, Genetic regulatory mechanisms in the synthesis of proteins Journal of Molecular Biology, 3, 1961, p. 318.
6. JP Changeux, P Courrège, A Danchin
A theory of the epigenesis of neuronal networks by selective stabilization of synapses
Proc Natl Acad Sci U S A (1973) 70: 2974-2978