Mendel, Weldon and the uncertainty of counterfactuals

Nature has an opinion piece on Mendel's legacy by Gregory Radick who is a professor at the University of Leeds. The focus of Radick's article which is titled "Teach students the biology of their time" is on a counterfactual: what if instead of Mendel's work, it was Raphael Weldon's work which had been recognized, emphasized and perpetuated by the early pioneers of genetics?

The point Radick is making is that Mendel's followers emphasized the primacy of nature (as in 'nature vs nurture') in the form of the gene so much that the role of nurture was sidelined. Weldon on the other hand seems to have been an early proponent of the role that the environment played in the sculpting of physical and behavioral traits. An accompanying comment in Nature endorses the piece and says that if Weldon's work had actually acquired the importance it should have, we might not have gotten obsessed with finding a "gene for this" and a "gene for that".

It's always interesting to consider what ifs and counterfactuals, not in the least because they serve as a useful vehicle for taking stock of what actually happened and for analyzing various currents of thought. But the real lesson in counterfactuals in my opinion is about historical contingency. My main problem with counterfactuals is that we assign them a degree of certainty that history's messy contours always seem to thwart. Since factual events themselves by definition are well-charted and reflected in the facts, we seem to think that their counterfactuals would also be well-charted. This in my opinion is affording a kind of luxury of prediction to history that it simply does not possess.

Radick's argument, as interesting as it, also suffers from these shortcomings in my opinion. I was vaguely familiar with Weldon's work but had the renewed opportunity to take a look at it in Sid Mukherjee's new book "The Gene." One thing that the Nature pieces don't mention is that Weldon had allied himself with Francis Galton and Karl Pearson in clinging to a flawed theory of fractional genetic inheritance in which one half of an offspring's genes would come from the parents, a fourth from the grandparents and so on. William Bateson enthusiastically fought against this idea and in the end prevailed.

But more pertinently here, it's interesting to consider some of the details of the counterfactuals that Radick and Nature are discussing here. Let's say it was Weldon's body of work and not Mendel's that was emphasized. Would it have really diverted attention from the importance of the gene? And would it have really compelled observers to take the environment more seriously and not talk about "genes for this" trait and "genes for that" trait? And if Weldon had prevailed along with Galton against Bateson, would Galton's eugenics drive have acquired an even more respectable stamp? I am not sure of the trajectory of any of these potential developments. 

More intriguingly, do we seriously think that a recognition of the environment would have stopped Thomas Hunt Morgan from actually finding important genes for specific traits in fruit flies? And would it have stopped Hermann Mueller from then finding out cartloads of mutations in these genes by exposing flies to x-rays? On the other hand, Theodosius Dobzhansky did determine the effects of environmental factors in his own fruit fly experiments in the 40s. I think that if anything, mainstream science was quite attuned to the effects of the environment in influencing traits, and genetic factors took some time to be cemented as serious determinants

I think we can agree that Morgan and Mueller's work was supremely important in the history of genetics, and if Weldon's followers had actually kept them from finding what they did it would have been a great loss for science and a strike against Weldonian theorizing. Just because trying to find genes for every single trait or disorder is messy or often a doomed process does not mean the concept itself is a problem; I would say that on balance the search for genetic determinants of traits has been enormously useful and promises to provide a bonanza especially in medicine. Strangely enough, the specific case of causal genes which the Nature piece invokes to illustrate the problems with genetic immutability in fact demonstrates the opposite point: these would be genes for heart disease. The article asks what the tangible gains are in asking whether there is a "gene for heart disease", when at least two genes for heart disease (ones for HMG-CoA reductase and PCSK9) have led to two of the most important targeted therapies for the ailment, therapies which have tangibly improved and saved tens of thousands of lives. Heart disease is of course a complex mix of genes and environment, but it's also a medical disorder where the delineation of a "gene for that" has been extremely helpful.

Secondly, it's not as if the emphasis on environment over genes has ever stopped ideologues and scientists from pursuing deeply flawed ideas with tragic consequences. It was precisely a rejection of Mendelian genetic determinism that led Trofim Lysenko and his Soviet overlords to embark on campaigns to "reeducate" wheat through "shock therapy" and to reeducate dissidents through "Gulag therapy". The noted Mendelian geneticist Nikolai Vavilov was imprisoned and tortured for his theories and he died a broken man. Other socialist ideologies have also engaged in similar campaigns. The important lesson here is that the perversion of a scientific idea for ideological purposes does not make an argument either for or against the idea itself. The use of social Darwinism in supporting Nazism makes as much of a case for rejecting genetic causes as the use of environmental leveling in supporting Soviet socialism makes for rejecting environmental causes.

The piece asks whether teaching counterfactuals might be a good tool for exposing students to different schools of thoughts and provoking them to think about other directions that history might have taken. Generally speaking I am always in favor of teaching students the history of scientific ideas, but I also think that this works only if we also teach them the inadequacy - in fact the futility - of post-factual historical prediction. The actual unfolding of history is so messy and subject to so many contingent forces that its march looks linear only in retrospect, when we have its one true manifestation in the form of bare facts at hand. Any speculation regarding what ifs remains speculation at best.