R.C. Lewontin, Steven Rose, and Leon J. Kamin (1984)
J.D. 2024-04-23
The Selfish Gene (Dawkins, 1976) is one of those books that has changed the way we think about the subject, and almost fifty years later, still feels as fresh as I can only imagine it to have felt in 76. Along with its follow-up, The Extended Phenotype, it's one of the few books I find myself frequently going back to. The genes that consitute my genome come together to put 'me' together as a cohesive unit--well, at least that's how it feels to me and, safe to assume, you as well. In a segment of a specific chromosome, I carry a version of a gene associated with an elevated likelihood of contracting uveitis, a painfully irritating auto-immune condition that causes inflammation of the uvea--the part of the eye that includes the iris. The feeling is something like an eye cramp that can only be 'stretched out' with corticosteroid drops. This narrow slice of my genome carries with it particular instructions that happen to work against its irritated host. Uveitis isn't close life threatening, but carriers of this genetic marker have been sufficiently procreative that not only was I conceived and sufficiently long lived to have possibly passed on my version of the annoying gene, but almost ten percent of Chinese and a quarter of particular groups of Scandinavians are host to it. If it were life threatening, it would be an example of a gene that reduced my likelihood of surviving long enough to pass it along, working along with the rest of creation to cull the genetic pool of bad genes like mine. Other more well known mutations are the BRCA1 and BRCA2 genes, well known thanks to Angelina Jolie and the internet. These genes that are otherwise responsible for repairing DNA mutated at some point in the past and, in carriers of the offensive mutation, ceased to effectively treat damaged DNA and, as a consequence, increased risk of breast cancer in women. These are all examples of detrimental or harmful mutations, but mutations might also have positive effects. It's usually these that we talk about when we're talking about natural selection--genes responsible for a chameleon's ability to camouflage, a spider's natural ability to spin a web, a lion's strength, an antelope's speed. Genes that give a species some sort of selective advantage in gather resources or producing offspring.
The genetic information that each gene carries does not determine outcomes on its own, and this is one of the overarching points that Lewontin, Rose, and Kamin make in Not In Our Genes. Genes push and prod in specific directions, adding up to statistically meaningful influences, all else being equal. But a point worth making is that, for all practical purposes, the 'all else' is practically innumerable. My chance of suffering from uveitis can be expressed as a probability--as can any of a host of other afflictions that, at some point in life, I may or may not find myself complaining of. The probabilities that describe my possible and specific ailments depend on legions of environmental factors, too many to count and usually too complex to understand but in total. A sample of these factors might include temperature of the climate I grew up in, stress levels, diet, humidity, exposure to sunlight, upbringing and parental nurture, the remaining fifty percent of each of my parents' genomes that I happened to randomly inherit, and countless others.
When thinking about genetics and environment, a helpful analogy--and one that I recall Dawkins also having made somewhere--is of a computer program. If I were to write a piece of software that takes some text as input, the output will depend on both the steps specified in the program as well as the data fed into the software. What this means is that if I were to feed different files into the same program, the outputs would differ only to the extent that the data fed through the program's logic apply to different conditions specified in the program. And those differences can be vast. On the other hand, if the same text file were fed through different programs, the outputs would again differ, but this time according to the programs' various instructions that would cause the data to be processed differently. It's worth pointing out now that computers provide insulated environments that look nothing like the web of interrelated factors that affect and contribute to our living environment. But whether or not we can successfully appreciate the environmental factors that contribute to what amounts to our 'inputs', we can still express outcomes as probability distributions specified by a combination of those inputs and our genetic markers.
Rather than computer programs, Lewontin, Kamin, and Rose use the analogy of a recipe:
Think, for example, of the baking of a cake: the taste of the product is the result of a complex interaction of
components--such as butter, sugar, and flour--exposed for various periods to elevated temperatures; it is not
dissociable into such-or-such a percent of flour, such-or-such of butter, etc., although each and every component
(and their development over time at a raised temperature) has its contribution to make to the final product. In a
world in which such complex developmental interactions are always occurring, history becomes of paramount
importance. Where and how an organism is now is not merely dependent upon its composition at this time but upon a
past that imposes contingencies on the present and future interaction of its components.
If you're trying to bake a cake, exposure to more-or-less precise duration of elevated temperatures yields specific outcomes when applied to specific ingredients. The effect of environment (temperature) on those ingredients has a predictable range of payoffs--all of which happily look, more or less, like a cake if you were paying attention to the recipe. The discrete range of possible outcomes given certain ingredients may also have overlapping or similar outcomes to a combination of different ingredients. I remember asking my mom and aunt for a recipe for the same dish at different times only to get what looked to me like completely different sets of instructions. The end results tasted almost identical even though, I have to say, my mother made things much more difficult.
So the genes that persist between generations in our "genetic software" influence the probability of some trait being expressed. It is a statistical influence rather than a guarantee. Maybe an all-knowing being, Laplace's demon, could take inventory of the thousand and one factors that cause a specific behavior to be manifest, but were it so, it's not among Dawkins' claims--and we'll have to settle for probabilities as well. What Lewontin, et al. do help us understand, though, are the problems that arise when the probability density function with respect to some genetic variants under some random environmental factors tends to favor the carrier of one allele over another, yielding, on average, unequal outcomes. What this means is: if evolution has produced, for example, sexual dimorphism, male and female, in mammals such that the average male is usually, or have a higher chance of being larger or stronger than the average female, a conclusion of 'might is right' is problematic--evolution and mother nature are value-free. The evolutionary dance that has been going on between lions and antelopes for millenia has built magnificant creatues though at immeasurable cost and a bone chilling amount of suffering. In other cases, the chance of successful procreation might even have been aided in our less enlightened forebears, not hindered by the byproduct of unequal outcomes when considering some dimension like strength or size. Sexual dimorphism might be an example of this, where, on average, males tend to be taller and stronger than their female counterparts. The present evolutionary outcomes in sexual differences do not come with values but they do help to explain the environmental pressures placed on our ancestors.
While I can't speak for E.O. Wilson, Maynard Smith, and others, it was surprising to read charges of biological determinism brought against Dawkins by Lewontin and company. But they help to clarify what they see as the "reductionist and biological determinist propositions" that Dawkins is alleged to hawk:
The reductionist and biological propositions that we shall examine and criticize in the pages of this book are:
- Social phenomena are the sums of the behaviors of individuals.
- These behaviors can be treated as objects, that is, reified into properties located in the brain of particular
individuals.
- The reified properties can be measured on some sort of scale so that individuals can be ranked according to the
amounts they possess.
- Population norms for the properties can be established: Deviations from the norm for any individual are
abnormalities that may reflect medical problems for which the individual must be treated.
- The reified and medicalized properties are caused by events in the brains of individuals events that can be given
anatomical localizations and are associated with changed quantities of particular biochemical substances.
- The changed concentrations of these biochemicals may be partitioned between genetic and environmental causes;
hence the "degree of inheritance" or heritability of differences may be measured.
- Treatment for abnormal amounts of the reified properties may be either to eliminate undesirable genes (eugenics,
genetic engineering, etc.); or to find specific drugs ("magic bullets") to rectify the biochemical abnormalities
or to excise or stimulate particular brain regions so as to eliminate the site of the undesirable behavior. Some
lip service may be paid to supplementary environmental intervention, but the primary prescription is "biologized."
If social phenomena cannot be described as a collection of individual behaviors that are first initiated in the brains of those individuals, I'm unsure what social phenomena are claimed to be. Photons hit photoreceptors in your eyes, perturbations in air pressure are perceived as speech patterns that carry and convey information, and other signals come together as electrical impulses that, in turn, fire off subsequent spikes that cause further actions, neurological and neuromuscular. I may be missing the point or am leaning on twenty-first century reference frames to pen a response to a fourty year old book, but if social behavior cannot be analyzed as neurological events, I'm not sure they can withstand analysis at all. This is different from claiming that social phenomena are 'just' action potentials--we would be missing higher relevant levels of analysis. To explain complex phenomena with respect to various component parts is standard practice and it does not necessarily follow that those phenomena are merely the sum of its component parts. So this isn't reductionism anymore than claiming that fluid dynamics can be analyzed at the molecular level. While fluid dynamics can be analyzed as average pressures, flow velocities, and energy of countless individual molecules, the computational complexity of doing so would be immense and measurement all but impossible. There would also be no practical use in doing so, nor would the aggregate effects be anywhere reflected in molecular Brownian motion. In the same way, evolutionary processes are the product of countless but explicable environmental factors in which certain genetic variants, on average, tend to thrive more than others. The claim about treatment of individuals with abnormalities (point four) seems to be a conclusion that they themselves come to--it's an 'ought' that Dawkins doesn't dip his toes into. In points five and six, this is exactly the point about genes exerting some measureable degree of influence on behaviors. This much is true. Whether we know the locus--or if there is even a calculable or discernible number of loci--where particular behaviors are generated, unless we are claiming a supernatural medium for the genesis of some behavior, it seems to me an uncontroversial statement to claim that those behaviors have "anatomical localizations [...] associated with changed quantities of particular biological substances." Whether the biochemical substances are DNA or hormones, no abstractions need reification to construct a framework for analysis on sound biological grounds. In their last point of biological determinist propositions they claim that the conclusion of all of this is either eugenics or medical intervention. This sounds terrible, anti-humanist, not to mention antithetical to how I understand Dawkins' work.
Building on these propositions, Lewontin, et al. go on to say that biological determinism is a way of explaining away or justifying inequality, defending inequality by the reinforcement of inevitability:
Biological determinism (biologism) has been a powerful mode of explaining the observed inequalities of status,
wealth, and power in contemporary industrial capitalist societies, and of defining human "universals" of behavior
as natural characteristics of these societies. As such, it has been gratefully seized upon as a political
legitimator by the New Right, which finds its social nostrums so neatly mirrored in nature; for if these
inequalities are biologically determined, they are therefore inevitable and immutable."
A similar quote, later in the book:
Sociobiology is a reductionist, biological determinist explanation of human existence. [...] The academic and
popular appeal of sociobiology flows directly from its simple reductionist program and its claim that human
society as we know it is both inevitable and the result of an adaptive process.
This seems to me to be impossibly far from the actual views of Dawkins, (and, as far as I'm aware) John Maynard Smith, and E.O. Wilson. In fact, a reading of Dawkins' introduction to The Selfish Gene would dispell quite a few of these concerns, but two points that Dawkins makes are especially salient, from the first and concluding chapters of The Selfish Gene. Dawkins begins by articulating "the first point [...] about what this book is not. I am saying how things have evolved. I am not saying how we humans morally ought to behave. I stress this, because I know I am in danger of being misunderstood by those people, all too numerous, who cannot distinguish a statement of belief in what is the case from an advocacy of what ought to be the case." Nature does not have anything to say on ethics. Nature is as mute on the ethical shortcomings or virtues of the lion and the antelope as it is about the selective pressures that we find ourselves the outcome of. And while we rightly bristle with Lewontin, et al. at inequities, biases, and other all too human shortcomings that we have inherited, we happen to be the only species capable of introspection with respect to any of these. Dawkins wraps things up in the concluding chapter:
It is possible that yet another unique quality of man is a capacity for genuine, disinterested, true altruism.
[...] The point I am making now is that, even if we look on the dark side and assume that individual man is
fundamentally selfish, our conscious foresight our capacity to simulate the future in imagination could save us
from the worst selfish excesses of the blind replicators. We have at least the mental equipment to foster our
long-term selfish interests rather than merely our short-term selfish interests. We can see the long-term
benefits of participating in a 'conspiracy of doves', and we can sit down together to discuss ways of making the
conspiracy work. We have the power to defy the selfish genes of our birth and, if necessary, the selfish memes of
our indoctrination. We can even discuss ways of deliberately cultivating and nurturing pure, disinterested altruism
something that has no place in nature, something that has never existed before in the whole history of the world.
While genes unquestionably influence who we are and how we behave, we alone in nature are able to shuffle off these natural inclinations. Statistical influence and probability give us measures by which we can focus our attention in the interest of long-term flourishing. And the flourishing that we can and do create is of a world that includes more than parochial short-term selfish appetites. While we may not like what we see when we reflect on our species, isn't it better to have a clear-eyed picture of the reality of our kind than to shake our heads and say it ain't so? Participation in a 'conspiracy of doves' needs a clear idea of the constraints we've grown up under as a species and from what we might hope to disentangle ourselves. That is the revolutionary insight that made treaties and the illegality of war possible, womens' rights, civil rights, contraception, and many other deliberate contraventions of our selfish evolutionary heritage. In order to shed light on who and why we are and, more importantly, a map of what we might want to overcome, Dawkins spends all of The Selfish Gene discussing the many ways in which our biology might be at odds with the goal of limiting the suffering of conscious creatures, of creating and promoting peace and well-being that extends to a wider circle of humanity than our selfish interests might allow, of considering the world that we leave behind to generations that we won't have the pleasure of meeting. To this end, he wraps up with the hopeful message that we can defy our genetic heritage and indoctrination, look cooly at what we have been handed, and choose to cultivate a better world. On the other hand, Lewontin, et al. claim that "For biological determinists we are unfree because our lives are strongly constrained by a relatively small number of internal causes, the genes for specific behaviors or for predisposition to these behaviors." (p.289) To me, this is the central misunderstanding. Lewontin, Kamin, and Rose believe sociobiologists to say that behaviors are tightly constrained by our genes. Sociobiologists advocate that we better understand our biological tendencies to better escape them.
The frameworks that we construct to describe the world are always theory-laden. There doesn't exist measurement absent of theory. The fact of measuring a thing versus another, the units and scale we use to measure, the measurement methodology, are all built on a theoretical framework. Science as a discipline seeks to counteract the methodological biases that might be built into our theoretical frameworks by starting from an assumption of fallibalism. Hypotheses and theories withstand relentless efforts at disproof. But Lewontin, et al. seem to frame things a little differently. "One of the issues with which we must come to grips is that, despite its frequent claim to be neutral and objective, science is not and cannot be above "mere" human politics." This is precisely why the scientific method has the onboard goal of disproving the same theories that the process of scientific discovery help to uncover, not to undo the science but to strengthen it. The scientific revolution was the single most effective catalyst for progress precisely because it made damn sure that the claims made under its license weren't baloney. We have no better way of evaluating conjecture. Lewontin, et al. rightly go on to say, "The complex interaction between evolution and scientific theory and the evolution of social order means that very often the ways in which scientific research asks its questions of the human and natural worlds it proposes to explain are deeply colored by social, cultural, and political biases." However, Dawkins is exceedingly careful to separate advocacy from scientific claim. The fact of the genesis of our genetic heritage, the "worst selfish excesses of the blind replicators", is separate from the "long-term benefits" and of "[cultivation and nurturing of] pure, disinterested altruism" that Dawkins advocates for.
Lastly, in the foreward to the first edition of The Selfish Gene, fellow evolutionary biologist Robert Trivers says the following of the claims of biological determinism:
The recent progress in social theory has been substantial enough to have generated a minor flurry of counter-
revolutionary activity. It has been alleged, for example, that the recent progress is, in fact, part of a
cyclical conspiracy to impede social advancement by making such advancement appear to be genetically impossible.
Similar feeble thoughts have been strung together to produce the impression that Darwinian social theory is
reactionary in its political implications. This is very far from the truth. The genetic equality of the sexes is,
for the first time, clearly established by Fisher and Hamilton. Theory and quantitative data from the social
insects demonstrate that there is no inherent tendency for parents to dominate their offspring (or vice versa).
And the concepts of parental investment and female choice provide an objective and unbiased basis for viewing sex
differences, a considerable advance over popular efforts to root women's powers and rights in the functionless
swamp of biological identity.
While the questions that Lewontin, Kamin, and Rose bring up are clearly worth asking, by my lights, they're going toe to toe with the wrong target. But worse, the arguments made seem to miss the deeply hopeful and humanistic message that wraps The Selfish Gene's central arguments of our unique ability in the universe to comprehend and rise above our evolutionary heritage.