How much does the Modern Synthesis explain?

This  is a thought-provoking review article by Jonathan Bard of Oxford on both James Shapiro’s Evolution – a View from the 21st Century  and Transformations of Lamarckism: from Subtle Fluids to Molecular Biology edited by S.B. Gissis & E. Jablonka, which is a historical assessment of Lamarck and his intellectual successors. Those of my acquaintances who struggled to understand Shapiro will be comforted that Bard agrees you need a biology degree to make much sense of it.

After succinctly describing the Evolutionary Synthesis in classical population genetics terms, Bard says this:

The enormous amount of molecular information that has emerged during the last couple of decades is making us review this story, partly because we now know that the relationship between the phenotype and genotype is not as simple as previously assumed, partly because the genome is a richer, more complicated world than the scientists who put together the modern synthesis could ever have supposed and partly because there is data that does not fit comfortably within the synthesis.

This in itself is a tacit admission that despite the “evolution is fact” polemic, we are much farther from an adequate understanding of the process than is usually claimed. To the extent we don’t understand it, assertions of its undirected nature are also simply so much froth. Bard is even more explicit in the opening sentence of his abstract:

The evolutionary synthesis, the standard 20th century view of how evolutionary change occurs, is based on selection, heritable phenotypic variation and a very simple view of genes. It is therefore unable to incorporate two key aspects of modern molecular knowledge: first is the richness of genomic variation, so much more complicated than simple mutation, and second is the opaque relationship between the genotype and its resulting phenotype.

He goes on to critique Shapiro’s book for an inadequate treatment of the second point (how the genotype relates to the phenotype), and in the rest of the article explores some of the possibilities, particularly (in the light of his focus on Lamarckism) ways in which the environment could affect evolution quite apart from genomic involvement. It’s a good read. But it also drew my attention to a rather simpler aspect (in the sense of being a simple view of something inordinately complex).

The heart of population genetics on which the Neodarwinian Synthesis was developed is the simple genetic model that one gene = one protein = one trait. In other words, it is a slightly more sophisticated version of Mendel’s genetics (which is why some have said it should really be called the Neomendelian Synthesis). At least in popular discussions (such as BioLogos threads) the increasing complexity found in real life is simply accommodated to this understanding at a stroke. What if two or three genes are involved in a trait? It makes the actual calculations more difficult, but the underlying principles are the same. And if non-coding DNA is beginning to reveal its importance as the source of control mechanisms, we just apply the population genetics to that level rather than simply to protein synthesis.

But population genetics, in order to make useful predictions, has to assume that genes are unlinked. Or to put it another way, that the gene is a black box whose contents, however complicated, can be treated as a single unit. As Bard puts it:

There is a serious underlying problem with the evolutionary synthesis: it is based on a minimalist Mendelian view of genetics which assumes that a very small number of genes underpin a trait and a mutant gene leads to an abnormal phenotype. While the advantage of the formulation is that it provides a model for evolutionary genetics, the disadvantage is that the approach assumes a naively simplistic view of how genes generate traits, as Waddington pointed out in the ‘50s. If more than about three genes (nature unspecified) underpin a phenotype, the mathematics of population genetics, while qualitatively analyzable, requires too many unknown parameters to make quantitatively testable predictions.

He goes on:

The inadequacy of this approach is demonstrated by illustrations of the molecular pathways that generates traits : the network underpinning something as simple as growth may have forty or fifty participating proteins whose production involves perhaps twice as many DNA sequences, if one includes enhancers, splice variants etc. Theoretical genetics simply cannot handle this level of complexity, let alone analyse the effects of mutation.

This seems to assume, at least, the conventional notion of genes coding for proteins. The sheer number of proteins involved in the mechanisms of real living organisms makes population genetics of no practical use. But as Bard then reminds us, even without the intriguing role of epigenetics and so on, DNA has been found to have more than 50 functions so far: gene = protein = trait is as relevant to evolution as adding up is to quantum physics.

A quote I rather like, from the guitarist Robert Fripp,  may provide a simple analogy:

With a note of music, one strikes the fundamental, and, in addition to the root note, other notes are generated: these are called the harmonic series… As one fundamental note contains within it other notes in the octave, two fundamentals produce a remarkable array of harmonics, and the number of possible combinations between all the notes increases phenomenally. With a triad [three notes], affairs stand a good chance of getting severely out of hand…

It’s the old problem restated: you don’t develop complex systems by thoughtlessly changing individual components. Because as every technophobe knows, those individual components are overwhelmingly likely to louse up the whole system if you don’t know what you’re doing. Baby applies pencil to paper – maybe a creative result to stick on the fridge door. Same baby, same pencil, into back of computer – less creative in every way.

The more you look into the way that cells and organisms actually work, the less adequate the Modern Synthesis is to account for what we actually see in life. It simply has to go – or maybe more accurately, it has to stay in the nursery. Whether people like James Shapiro or Jonathan Bard will be able to cast any fundamentally better light on it remains to be seen.

Avatar photo

About Jon Garvey

Training in medicine (which was my career), social psychology and theology. Interests in most things, but especially the science-faith interface. The rest of my time, though, is spent writing, playing and recording music.
This entry was posted in Creation, Science. Bookmark the permalink.

Leave a Reply