The Century of the Gene

As the title suggests, Keller argues in this short book that the gene has had its century, and it is time for biology to focus on broader issues. This is not to say that genes are not important; Keller is very clear that they are. However, the very research that has tried to establish exactly what genes do has uncovered complications, conceptual ambiguities, and non-genetic factors that violate the simplifying assumptions that came to guide gene research from the middle of the 20^th Century. Contrary to some critical reviews of this book, Keller is not critical of genetic research in itself; more often, she praises it. She argues very strongly, however, that genetic research has shown conclusively that the gene is not what it used to be. Her main point is that the very success of genetics has (or should) teach us humility. A subsidiary point is that the rapid changes in genetics have left us with a heritage of inadequate concepts, specialised practices with little theoretical integration, and a residue of inadequate but widely accepted ideas of what genes are. Scientists working in each of these specialised fields know what they are doing, and how they understand and apply the concepts of genetics, but there is no genetic lingua franca. Unfortunately, given the social importance of genetics, past and current discourse about genes leaves the public with the opposite impression. If nothing else, this book is an antidote to that (possibly dangerous) impression.

The exposition of the book is historical. The history is a brief synopsis of highlights, and the triumphs of genetic research are perhaps given less space than many would like, whereas the problems with the gene concept and with simple views of genetic determination are emphasised. Given the large number of histories that put the triumphs of genetic research front and centre, Keller’s emphasis is both warranted and rather refreshing. The book contains a short introduction and conclusion, and four longer chapters that are arranged both historically and thematically.

The first chapter deals with the regulation of genetic stability. This is the basis for the functionality of the gene as a unit of heredity. It turns out that the gene is not stable enough on its own, and needs help. Part of this help comes from genetically controlled processes, but not all. Still, this leaves the gene as the central actor in the determination of traits, though it needs its dressers and perhaps even its directors.

The second chapter deals with the issue of what genes do. If the gene is a functional unit that that is responsible for heredity, then it determines heritable traits in the cell and the whole organism, and can replicate and transmit this capacity for determination from cell to cell and from organism to organism. Or at least that was the first version. The discovery of the structure of DNA and the genetic code confirmed earlier observations that DNA was the locus of inheritance, and that one gene coded for one enzyme (functional protein). Keller quotes Francis Crick, “DNA makes RNA, RNA makes proteins, and proteins make us”. Subsequent discoveries in genetics have shown this to be a bit of an exaggeration. Keller describes several major modifications in this “Central Dogma”. First, some DNA regulates other DNA. Second, the RNA transcribed is edited, and many sections of DNA have multiple mappings onto proteins. If segments of DNA are the units of heredity, then the units of heredity are not the genes in terms of developmental function, and vice versa. Sometimes the two coincide, and these cases are important, but they are far from being the rule. Nonetheless, much popular opinion, and even much science, works as if the Central Dogma were Nature’s own Truth.

The next chapter deals with the concept of a genetic program. This chapter shows how this somewhat more sophisticated notion of gene activity is supposed to work, and some of the problems it encounters. The most obvious one is that the way a program works depends on the type of machine it runs on. It turns out that there are both surprising similarities across diverse kinds of organism, and surprising divergences in similar organisms. The idea that the genes are a blueprint for making an organism is both disarmingly accurate where certain genes that govern eye development in flies lead to mouse eyes when transplanted to mice, and dismayingly inaccurate when we look at the genetic details of an organism as simple as the roundworm, C. elegans. 

Turning to development, and developmental stability, Keller points to the role of organisation and redundancy. Both imply no simple relation between genes and traits. Unlike current computer programs, small variations (and even some large ones) have little effect on the end result. It is interesting that IBM has recently started a program to develop systems that have the sort of autonomy that we see in organisms. Perhaps the program metaphor will become accurate eventually.

The conclusion is a rumination on why we still talk of genes, given the problems. Keller suggests that gene talk is more of a hindrance to the understanding of lay people than of scientists, since the latter typically know what they are talking about, and by and large can compensate for inadequacies in gene discourse. Unfortunately, the “land rush” engendered by genomics has, if anything, led to an even more unwarranted focus on genes, whatever they are. The advantages of laying claim to a segment of DNA are too great to allow time to contemplate complicating details, such as underdetermination or overdetermination of physiological function. Economic decisions are considerably less corrigible than scientific ones.

In any case, the Central Dogma encourages the idea, often reported in the news, of the gene for X. The problems with the Central Dogma are the problems with the notion of a gene for X. Science reporters should read Keller’s book carefully, and digest its message fully. Failing this (and given that with few exceptions reporters report what they think scientists have said, rather than the science itself, it will fail), one can hope that people of influence who read those reports will have read Keller’s book, so they are not so easily swayed by the heroic archetype of the gene.

© 2002 John Collier

John Collier is a Visiting Scientist at the Konrad Lorenz Institute for Evolution and Cognition Research. His work is in foundations of information theory, autonomy, evolutionary theory and theory change, as well as various papers in metaphysics and the philosophy of physics. He is currently working on a book on reduction in complex systems with C.A. Hooker.