Before Tomorrow- Epigenesis and Rationality

by Catherine Malabou

  Cognitive organization is constituted of several levels. The most concrete and elementary is the perceptive level; the most abstract is the symbolic level. These levels are distributed hierarchically in the neuronal architecture. From the elementary circuits of the spinal cord, brain stem, and ganglions to the frontal cortex, everything combines to form “the neural architecture of reason.”26 At their highest level, the neuronal “assemblies” or “populations” code cognitive operations (“population coding”).

  At each level of brain organization, “matter” elements combine, like Darwinian variations, to engender the “forms” at the next level. Some of these forms are then stabilized after being selected on the basis of their functional efficacy. Hence, “the function feeds back into the [matter–form] transition.”27 Mental representations are thus elaborated on the basis of the operations of the selection–stabilization pairing.

  The Example of Mathematics

  The theory of brain epigenesis by synaptic selection and stabilization makes no mention of a priori data in the constitution of mental objects and their agreement with cognitive structures. The eliminativist cognitive model substitutes the idea of a gradual, biologically determined agreement for the notion of a transcendental agreement of the categories with objects. Mathematical truths are the products of this type of adaptation. In Conversations on Mind, Matter, and Mathematics, a collection of conversations with mathematician Alain Connes, Changeux states that mathematical objects are indeed “coded in the brain as forms,”28 but that they have a material existence and show themselves to “correspond to physical states”29 perfectly. It is these “states” that are subject to evolutionary processes. Existence, reality, even the truth of mathematics, are “a posteriori results of evolution.”30

  It is clear that this position quite naturally prolongs the oldest evolutionist thesis analyzed by Bouveresse. Against Frege and Husserl, Changeux defends a cognitive psychologism according to which the reality of mathematical objects and logical idealities is the reality of mental states and processes, which, again, exist materially in the brain. “Mental representations – memory objects – are coded in the brain as forms in the Gestalt sense, and stored in the neurons and synapses, despite significant variability in synaptic efficacy.”31 Mathematical axiomatization, which is also defined as a brain process, is only possible on the basis of this materiality of forms.

  Mathematical representations are selected according to a contingent evolutionary process. They constitute mathematical objects as “cultural objects, [. . .] public representations of mental objects of a particular type that are produced in the brains of mathematicians and are propagated from one brain to another [. . .].”32 From this it follows that according to Changeux there can be no mathematical ontology: mathematical truth is in fact the result of an aleatory process that only becomes necessary through the action of selection. Thus, “the science of ‘why?’ isn’t theology, it’s evolutionary biology. And the ‘why?’ of the existence of mathematics has as much to do with the evolution of our knowledge acquisition apparatus – our brain – as it does with the evolution of mathematical objects themselves.”33

  The brain, whose embryogenesis and development are subject to the dynamic of stabilization by elimination, can therefore be defined as a biological machine produced by evolution. The agreement of the “forms” and mental objects, whether they are “pure” mathematical objects or the objects of experience in general, could not therefore be constituted before this dynamic. According to Changeux, there is nothing in brain organization that might correspond to the “transcendental” one way or another. The economy of rationality concurs entirely with the dynamic of brain epigenesis.

  Edelman’s Theory of Systems of Recognition

  It might be objected that there are less reductive, less hard-wired versions of neural Darwinism. On these grounds, American neurobiologist Gerald Edelman argues simultaneously for the theory of brain epigenesis by synaptic selection and stabilization and the existence of a priori structures of cognition. The epistemology that Edelman proposes also assumes the epigenetic constitution of all cognitive processes from the elementary biochemical and cellular level up to the appearance of consciousness. This is the “theory of neuronal group selection” (TNGS). This theory nevertheless allows for the existence of “value-categor[ies].”34 But is the vision of epigenesis that underlies it compatible with that of Kant?

  In Bright Air, Brilliant Fire: On the Matter of the Mind, Edelman explains that synaptic selection is a dual process that operates both at the “topobiological” level and at the level of “recognition systems.”

  The first selection operates within a cellular mass that contains possible structural variants. Selected structures are formed by elimination. The nerve cells project a series of extensions that stimulate target cells. When the signal for an extension fails to correlate with the target, these “fascicles” retract or disappear. Successful correlations form the basis of neuronal maps. This is the “epigenetic drama.” In this drama, Edelman explains,

  [S]heets of nerve cells in the developing brain form a neighborhood. Neighbors in that neighborhood exchange signals as they are linked [. . .]. They send processes out in a profuse fashion, sometimes bunched together in bundles called fascicles. When they reach other neighborhoods and sheets they stimulate target cells

  and in this way they form connections. Others retract and the emitting cells pass on.

  Finally, as growth and selection operate, a mapped neural structure with a function may form. The number of cells being made, dying, and becoming incorporated is huge. The entire situation is a dynamic one, depending on signals, genes, proteins, cell movement, division, and death, all interacting at many levels.35

  The second level of selection, that of “recognition systems”, implements a so-called “degeneration property.” This time the selection impacts the neural maps that result from selective elimination. The systems are eliminated by “degeneration” or consolidated in response to their “reaction” to exogenous solicitation. If the adaptive response is not viable, the system degenerates. Edelman develops the example of vision, but considers that the brain as a whole is a system of recognition that encompasses all the particular systems.36

  At this second level it is essentially the behavior of the individual that causes the reinforcement or weakening of the diverse populations of synapses involved in neural maps. Behaviors also play a primary role in the formation of a secondary repertoire of neural groups. As a general rule, experience is an essential factor in synaptic selection, which modulates and modifies networks and circuits.

  However, Edelman emphasizes one key point: in the brain, everything does not take place a posteriori. Thus, for example, the systems of recognition form themselves a priori, and not after or according to the forms they recognize. The units of recognition are certainly selected a posteriori, but the arrangement from which they derive forms itself. This is the reason why Edelman believes that the existence of a priori categories is compatible with neurophysiological properties. As he says, “[T]he brain carries out a process of conceptual ‘self-categorization.’”37 This system of categories (“value-category system”) then responds to the solicitations and signals coming from the outside world. Phenomenal experience results from this interaction between concepts and objects. It “arises from the correlation by a conceptual memory of a set of ongoing perceptual categorizations”38 and thus depends on pre-existing structures.

  It is readily apparent, however, that what Edelman calls a priori, and that which relates to the antecedence of the structural units of recognition on the forms or objects recognized, clearly does not coincide with the Kantian a priori. Although, unlike Changeux – and this is a considerable difference – Edelman accepts a certain functional autonomy of the structures of recognition (“self-categorization”), it is still dependent on adjustments and is therefore not entirely spontaneous.

  Let’s return to the discussion about the init
ial status of the categories in the understanding. The categories develop “with the opportunity” of experience, but (at least according to Zöller) it cannot be said that they later undergo a series of eliminations and transformations by reaction. On first sight it appears impossible to identify in Kant’s work a process of adjustment of the categories to objects and a system of response of objects to the categories. This is why neural or mental Darwinism, even in its moderate form, contradicts the critical position and once again deprives it of ownership of epigenesis. The biologization of the transcendental appears to need to destroy it irrevocably, which presupposes that the agreement of the categories with the objects cannot be thought outside the dynamic of adaptability, which itself is subject to variability. Edelman concludes: “[M]ind, which arose from material systems and yet can serve goals and purposes, is nevertheless a product of historical processes and of value-based constraints related to evolution.”39

  From Methylation to Hermeneutics

  But is it so easy to dismiss Kant from the debate? After these two chapters devoted to neo-evolutionism and mental Darwinism, let us return, in closing, to the relation between genetics and epigenetics. It is striking to see that scientists often describe the work of epigenetics metaphorically as improvisation, practical or artistic elaboration, creative spontaneity, in a word, as if, against genetic determinism, epigenetics related to the register of interpretative freedom.

  The most common image used to describe the work of epigenetics is indeed interpretation. Thus, Thomas Jenuwein, director of the Department of Immunobiology at the Max Planck Institute, describes epigenetics in these terms:

  The difference between genetics and epigenetics can probably be compared to the difference between writing and reading a book. Once a book is written, the text (the genes or DNA: stored information) will be the same in all the copies distributed to the interested audience. However, each individual reader of a given book may interpret the story differently, with varying emotions and projections as they continue to unfold the chapters. In a very similar manner, epigenetics would allow different interpretations of a fixed template (the book or genetic code) and result in different readings, dependent upon the variable conditions under which the template is interrogated.40

  In order to figure the relations between the genetic and the epigenetic, other scientists, such as Eva Jablonka and Marion Lamb, have recourse to the metaphor of the impact of music and its instrumental performance.41 The image of interpretation, whether textual or musical, evokes the style, individual fashioning, and endless possibilities for reading or playing in every instance. The use of this image does seem to indicate the opening of a hermeneutic dimension in the heart of the biological. It is as if the phenotypical event were in some senses an epigenetic version of the program. As if a space opened between them that calls for critical exploration. If epigenetic factors encompass physical mechanisms as much as environmental and social influences, then how, in the constitution of phenotypical individuality, could this be anything but the formation of a singularity that transcends strict determinism and places epigenesis and the development of all living beings in an intermediary space between biology and history? Following the metaphors used by scientists themselves, shouldn’t we go beyond the strict epistemological field to return to philosophy and claim the importance of epigenetics for thinking, and not just for cognition?

  And what if these remarks led us to reconsider the power of the transcendental, to reclaim it not as invariance and logical predisposition (as Zöller does, for example), but rather as hermeneutic latitude, the power of meaning, opened in the heart of the biological?

  These questions bring us back to the critical reading track and to its second line of understanding the transcendental as a historico-critical dimension of rationality that accompanies objectivity as its necessary shadow. This is a dimension that clearly cannot be assimilated by the “neo-skeptical” thesis, a dimension that resists it.

  Notes

  1. Conrad Hal Waddington, “The Basic Ideas of Biology,” in Towards a Theoretical Biology, Vol. 1, Prolegomena, Edinburgh: Edinburgh University Press, 1968, pp. 1–32, pp. 9–10.

  2. Thomas Morgan, “The Relation of Genetics to Physiology and Medicine,” Nobel lecture, 1935, p. 323 (http://www.nobelprize.org/nobel_prizes/medicine/laureates/1933/morgan-lecture.pdf).

  3. One of the ways that gene expression is regulated is through the state of the chromatin. Chromatin is either uncondensed or “open” (euchromatin), thereby allowing access to transcription and gene expression activity; or it is condensed or “closed” (heterochromatin), which prevents gene expression. The state of chromatin is dictated by post-translational modifications of the histone proteins linked to DNA.

  4. This modification involves the addition of a methyl group (-CH3) in place of a hydrogen atom. Although the four types of base can be methylated, cytosine is most frequently affected by the process.

  5. Thomas Pradeu, “Qu’est-ce qu’un individu biologique?,” in Pascal Ludwig and Thomas Pradeu, eds, L’Individu: perspectives contemporaines, Paris: Vrin, 2008, pp. 97–125, p. 120. Cf. also Henri Atlan, La Fin du tout génétique? Vers de nouveaux paradigmes en biologie, Paris: INRA Editions, 1999, p. 58: “We are witnessing a return of extreme preformationism, in the form of a new avatar, in which everything is contained in the genes.” All translations of Atlan’s work are mine.

  6. Changeux, The Good, the True and the Beautiful, p. 372.

  7. Nature, International Weekly Journal of Science, February 2001, online edition.

  8. Cf. Le Monde, February 13, 2001, “Le génome humain cache de ‘vastes déserts’,” online edition.

  9. Henri Atlan, La Fin du “tout génétique”?, p. 16.

  10. Henri Atlan, “Programme de Recherche Inter-Centres Biologie et société,” 2009, website.

  11. La Fin du “tout génétique”? Vers de nouveaux paradigmes en biologie, p. 16. See also La Recherche, no. 463, April 2012, “Épigénétique: L’hérédité au-delà des gènes,” pp. 38–54.

  12. “Post-genomic” biology assumes an interdisciplinary approach that expands the field of molecular biology in order to study element systems (DNA, proteins, supramolecular structures, small molecules) interacting with each other.

  13. François Jacob, The Logic of Life: A History of Heredity, trans. Betty E. Spillmann, New York: Pantheon, 1982, pp. 9–10.

  14. Research on certain types of cress has, for example, demonstrated that being exposed to cold during the winter led to structural changes in the chromatin, which silenced the flowering genes. These genes are reactivated in the spring when the longer and warmer days become suitable for reproduction. The environment may also provoke changes that have effects on future generations.

  15. With supplements rich in methyl such as folic acid and vitamin B12.

  16. Cf. Mae-Wan Ho, Living with the Fluid Genome, London/Penang: Institute of Science in Society/Third World Network, 2003, and “Epigenetic Inheritance: ‘What Genes Remember,’” Science in Society, vol. 41, 2009, pp. 4–5.

  17. Thomas Pradeu, “Philosophie de la biologie,” in Anouk Barberousse, Denis Bonnay, and Mikaël Cozic, eds, Précis de philosophie des sciences, Paris: Vuibert, 2011, pp. 378–403, p. 398. My translation here and below.

  18. For instance, in their fundamental text, Eva Jablonka and Marion J. Lamb write: “The idea that DNA alone is responsible for all the hereditary differences between individuals is now so firmly fixed in people’s minds that it is difficult to get rid of it. When it is suggested that information transmitted through non-genetic inheritance systems is of real importance for understanding heredity and evolution, two problems arise.” Evolution in Four Dimensions: Genetic, Epigenetic, Behavioral, and Symbolic Variation in the History of Life, Cambridge, MA: MIT Press, 2005, p. 109.

  19. Mary-Jane West-Eberhard, Developmental Plasticity and Evolution, New York: Oxford University Press, 2003, p. 34.

  20. Jean-Pierre Changeux, Philippe Courrège, and Antoin
e Danchin, “A Theory of Epigenesis of Neuronal Networks by Selective Stabilization of Synapses,” Proceedings of the National Academy of Sciences, USA, vol. 70, no. 10, Oct. 1973, pp. 2974–8 and Neuronal Man: The Biology of the Mind, trans. Laurence Garey, New York: Pantheon, 1985.

  21. Jean-Pierre Changeux and Antoine Danchin, “Selective Stabilization of Developing Synapses as a Mechanism for the Specification of Neural Networks,” Nature, December 23–30, 1976, vol. 264, no. 5588, pp. 705–12.

  22. Pradeu, “Philosophie de la biologie,” p. 400.

  23. Again, on this point, see Jablonka and Lamb, Evolution in Four Dimensions.

  24. Changeux continues: “It is no longer possible then to identify one gene with one function.” The Good, the True and the Beautiful, p. 372.

  25. Jeffrey M. Schwarz and Sharon Begley, The Mind and the Brain: Neuroplasticity and the Power of Mental Force, New York: HarperCollins, 2002, p. 365. The authors write: “Although it would be perfectly reasonable to posit that genes determine the brain’s connections, just as a wiring diagram determines the connections on a silicon computer chip, that is a mathematical impossibility. As the Human Genome Project drew to a close in the early years of the new millennium, it became clear that humans have something like 35,000 different genes. About half of them seem to be active in the brain, where they are responsible for such tasks as synthesizing a neurotransmitter or a receptor. The brain, remember, has billions of nerve cells that make, altogether, trillions of connections. If each gene carried an instruction for a particular connection, we’d run out of instructions long before our brain reached the sophistication of a banana slug’s. Call it the genetic shortfall: too many synapses, too few genes. Our DNA is simply too paltry to spell out the wiring diagram for the human brain” (pp. 111–12).