Excerpt from my book Epigenetics and Public Policy The Tangled Web of Science and Politics now available from my publisher here and at Amazon
Although the physical conditions for mutation are well known, so-called sporadic or spontaneous mutations and other aberrations that occur during cell division are for the most part treated in conventional genetics as more or less random events. That said, there are significant debates as to what exactly is meant by “random” in this context. In more sophisticated discussions random is said only to indicate the nature of the statistical sampling of a population and not “indeterminism or any technical mathematical meaning.” However, such technical explanations do not explain the links between the earlier invocation of random causes of variation made necessary by the assumptions of the insulation of organisms from their environment described at length in previous chapters, or the ongoing and now mostly tacit assumptions that mutations can be treated as random qua uncontrollable or unforeseeable events.
As discussed before, in general cancers are identified as the product of genetic mutations that are either passed on through the germline or acquired during one’s lifetime. Only a relatively few specific cancers are identified as the result of mutations from exposure to known carcinogens in the immediate environment, and none of these mutations are presumed to be heritable. In all other cases, these cancer-causing mutations are assumed to be functionally random events, both disconnected from their environments and not inherited. This dual recognition of some cancer-causing mutations as the product of environmental influences, but of all other mutations as somehow functionally insulated from the environment, is an awkward distinction that to my knowledge escapes not just scrutiny but general recognition within mainstream genetics and life sciences.
The origins of randomness
The primary scientific justification for this ultimate resort to randomness has its roots in the early days of the Modern Synthesis, and through population genetics in particular. Given the basis of population genetics in statistics, the assertion that mutations are inherently random seems to have originated primarily as a requirement of the mathematical formalisms being used to model how genes work (i.e., a rate of change needed to be specified in order for the equations being used to yield a result, and the assumption of randomness provided a convenient, mathematically simple, and empirically reasonable heuristic).
This approach was summarized in George Gaylord Simpson’s seminal Tempo and Mode in Evolution (1944), which combined paleontology and population genetics through Sewall Wright’s theory of random genetic drift. This book was a pillar of the Modern Synthesis because of the way Simpson applied the statistical formulas from population genetics to examples from the fossil record and preliminary results from genetics. This assumption of the inherent randomness of mutations was then extended in the 1950s through the work of Motoo Kimura and his application to genetics of the Kolmogorov backward diffusion equation from physics, which models heat conduction as a random dynamic process and would go on to become a standard tool in population genetics.
That the empirical results were fit to the mathematical formulas, and not the other way around, was a virtual necessity as there was not at the time any way to verify whether or not genetic mutations were in fact random qua causally disconnected from their environment—a theoretical difficulty which notably continues to this day. In other words, the assumption of the actual randomness of mutations was not driven by any observations of the actual randomness of mutations (in fact, if anything, the available empirical evidence was that genetic mutations were actually produced by specific mutagens such as x-ray radiation). Instead, this fundamental assumption was and is still driven by a priori theory and mathematical necessity, which are then used to inform subsequent theoretical and methodological choices in the development of the science of genetics.
Randomness and political ideology
As discussed elsewhere regarding Lamarckism and the emergence of modern genetics, this assumption of genetic mutations as random events—particularly as functionally disconnected from the environment—was one of the principal cudgels of neo-Darwinism against competing accounts of evolution and biological development. The conceptualization of mutations as random as “proven” by mathematics was a powerful way to justify describing the fundamental autonomy of humans from changes in their environments, in contrast to the neo-Lamarckism that was prevalent at this time which proposed that humans were quite malleable to their environments. As such, this notion of mutations as necessarily random (though, again, established as such more by a priori theory than through empirical observation) became a pillar of the Modern Synthesis of genetics, and still remains one of the central principles of contemporary genetics.
This is why, as also discussed elsewhere, one of Waddington’s primary motivations in initially proposing epigenetics more than 80 years ago was to expressly counteract this cornerstone assumption of genetic mutations as “totally unconnected with any ambient circumstances.” Waddington saw this as such a fundamental concern in part because this scientific assumption of organisms and their environments as having “as little essential inter-relation as a sieve and a shovelful of pebbles thrown on to it,” in turn carried with it substantial political implications, particularly regarding the relationships of humans with their environments and our responsibilities to each other.
Instead, for Waddington the work being done in embryology at the time clearly showed that organisms and their genes are essentially interconnected with their environments, and that an empirically valid science should reflect this relationship. As such, epigenetics was his attempt to reconcile and integrate the competing factions of genetics and embryology within a more comprehensive framework that incorporated the knowledge about the functioning of genes emerging from genetics with the knowledge about the role the environment played in the expression and function of genes. Although Waddington was manifestly unsuccessful in his attempt to realize a more holistic perspective for genetics, for all the reasons already discussed, the recent emergence of contemporary epigenetics—as demonstrated in this section in the domain of cancer—suggests that he was on the right track.
Some of the substantial political and ideological factors underpinning these choices to fit the theory of evolution and genetics to these specific mathematical formalisms, leading to this empirically unverified description of genes as practically insulated from their environments, have already been discussed. Still, it does bear noting again the degree to which these assumptions in the prevailing science of genetics mirrored similar political assumptions about the fundamental autonomy of human individuals in modern liberalism, and the role these philosophical assumptions played in the central ideological contest of this time between the liberalism of the West versus the collectivism of the Soviet Union. These geopolitical factors in turn also help to explain the specific direction the science took towards molecular genetics via this assumption of the inherent randomness of change in the genes in the absence of empirical observations—or, as in this case, actually against the only empirical proof available at the time.
Randomness, epigenetics, and the funding of science
Regardless, once this conception of mutations as inherently random gained traction and became the scientific consensus, it would go on to determine the subsequent definition of valid scientific work in genetics (i.e., work which reinforced and did not contradict this assumption of randomness qua insulation from the environment), and in this way determine the subsequent direction of research in genetics in the West, particularly as constrained by the structure of government funding for research described in chapters 8 and 10.
In particular, one concrete manifestation of this convergence of science and politics embodied in the assumption that the mutations of genes should be treated as random events is in the primary focus of cancer research over the past half-century. Instead of investigating the causes of mutations or other similarly “random” errors, the focus has been on identifying the control of genes over outcomes, or as the final or original cause in the causal chain resulting in cancer. Per Waddington this almost singular emphasis on genes as agents of control has pulled the designation of what is labeled appropriately “scientific” away from consideration of the greater contexts within which genes are nested. This scientific—and ideologically informed—conception of genes was only further reinforced by the distribution of the massive influx of government funding into science as funneled through increasingly narrow channels described in the previous chapter, which in turn influenced the subsequent direction of cancer science and research towards this specific conception of genes as controlling agents and away from consideration of the environments of the genes.
As such, the work being done in cancer epigenetics not only introduces substantial complications into the scientific understanding of cancer, and of genetics in general, in part by challenging the prevailing understanding of mutations as truly random, but in the process introduces even more substantial complications of the politics and policy of cancer because of the way it shifts attention back towards the nexus of genes and their environments. The effects of such a shift in attention could impact not only the distribution of funding, which would be a major political event within science in its own right, but also the greater policy responses to cancer which incorporate this new understanding of our connection to each other and to our environments.
 Lodish, H., Berk, A., Zipursky, S. L., et al. (2000). Mutations: Types and Causes. In Molecular Cell Biology 4th edition. New York: W. H. Freeman. Retrieved January 19, 2018 from https://www.ncbi.nlm.nih.gov/books/NBK21578/.
 Millstein, Roberta L. (September 15, 2016). Genetic Drift. In The Stanford Encyclopedia of Philosophy (Fall 2016 Edition), ed. Edward N. Zalta. Retrieved April 20, 2017 from https://plato.stanford.edu/archives/fall2016/entries/genetic-drift/
 Simpson, G. G. (1944). Tempo and mode in evolution (No. 15). New York: Columbia University Press.
 Watterson, G. (1996). Motoo Kimura’s Use of Diffusion Theory in Population Genetics. Theoretical Population Biology. 49 (2): 154–188.
 Provine, William. (2001). The Origins of Theoretical Population Genetics: With New Afterword. 2nd ed. Chicago: University of Chicago Press; Millstein, R. L. (2008). Distinguishing drift and selection empirically: “The great snail debate” of the 1950s. Journal of the History of Biology, 41(2), 339–367.
 Gleason, Kevin M. (1926–1927). Hermann Joseph Muller’s study of X-rays as a mutagen. Embryo Project Encyclopedia (2017-03-07). ISSN: 1940-5030. Retrieved from http://embryo.asu.edu/handle/10776/11441
 Waddington, C. H. (1957). The strategy of the genes. London: Routledge, 188.