Why Is Everyone Still So Muddled About Selfish Genes?

I was recently invited to speak in England on “the science of cooperation”, which was going to be billed as an alternative to selfish genes. I cringed at the comparison. I’m well versed in selfish gene theory and I know fully well that selfish genes do not necessarily imply selfish individuals. Selfish genes typically evolve by cooperating with other genes in the bodies of single individuals, but they can also evolve by cooperating with genes in the bodies of other individuals. Richard Dawkins and other selfish gene theorists have been saying that since 1976. Why should anyone be muddled on the topic in 2012?

But muddled they were. During my trip, it was abundantly clear that the term “selfish genes” had become part of the vocabulary but was used synonymously with “selfish” in the everyday sense of the word, along with other stock phrases such as “social Darwinism”, “survival of the fittest” and “nature red in tooth and claw”. The root idea is that unselfishness is somehow less “natural” than selfishness, an evolutionary version of the religious concept of original sin.

Either the general public is so stupid that they can’t be educated about the real meaning of selfish genes, or the educators are doing something wrong. Let’s explore the second possibility. Here are some of the best-known passages from the beginning of The Selfish Gene.

We are survival machines - robot vehicles blindly programmed to preserve the selfish molecules known as genes.

These are claims that could have been made for Lorenz's On Aggression, Ardrey's The Social Contract, and Eibl-Eibesfeldt's Love and Hate. The trouble with these books is that their authors got it totally and utterly wrong. They got it wrong because they misunderstood how evolution works. They made the erroneous assumption that the important thing in evolution is the good of the species (or the group) rather than the good of the individual (or the gene).

The argument of this book is that we, and all other animals, are machines created by our genes. Like successful Chicago gangsters, our genes have survived, in some cases for millions of years, in a highly competitive world. This entitles us to expect certain qualities in our genes. I shall argue that a predominant quality to be expected in a successful gene is ruthless selfishness. This gene selfishness will usually give rise to selfishness in individual behavior. However, as we shall see, there are special circumstances in which a gene can achieve its own selfish goals best by fostering a limited form of altruism at the level of individual animals. 'Special' and 'limited' are important words in the last sentence. Much as we might wish to believe otherwise, universal love and the welfare of the species as a whole are concepts that simply do not make evolutionary sense.

Let us try to teach generosity and altruism, because we are born selfish. Let us understand what our own selfish genes are up to, because we may then at least have a chance to upset their designs, something that no other species has ever aspired to do.

These passages leave little doubt that, according to Dawkins, the primary outcome of selfish genes is selfish individuals, that evolution can explain only “a limited form of altruism at the level of individual animals”, and that a more expansive form of altruism for the good of the group or species is “totally and utterly wrong”. The take home message is that we are “born selfish” and altruism and generosity must be taught.

So, the idea that “selfish genes” more or less means “selfish” in the everyday sense of the word isn’t something that the public got wrong. It came straight from Dawkins.
Perhaps it’s true, but if it proves to be wrong, the general public is muddled because Dawkins was muddled.

The distinction between “for the good of the group” and “for the good of the species” is important. A group might be a fish school, a bird flock, an ant colony, a pasture full of lemmings, a human hunter-gatherer group, or a human nation. An entire species typically occupies a large geographical area and consists of many, many groups. Species-level selection is theoretically possible and there is an interesting recent literature on it, but I will put it aside for the moment to focus on group-level selection. In other words, for the moment, I agree with Dawkins that we should not expect universal love and concern for the welfare of one’s entire species. The question is whether we can expect traits to evolve on the strength of being good for one’s group, despite being selectively disadvantageous within groups. Dawkins considered both to be “totally and utterly wrong”.

Dawkins is not the only evolutionist who thinks that we are born selfish and must be taught generosity and altruism. George C. Williams, whose 1966 book Adaptation and Natural Selection provided much of the material for The Selfish Gene, famously said that “Mother Nature is a wicked old witch”. Both can be regarded as modern counterparts of Thomas Huxley, who regarded nature as operating at the same level as a gladiator’s show. Just as Huxley was opposed by Darwin (who tried to explain animal altruism and human morality as products of evolution) and Peter Kropotkin (who placed an even greater emphasis on the importance of mutual aid in animal evolution), the “born selfish” position of Williams and Dawkins is opposed by contemporary evolutionists such as Frans DeWaal and Sarah Blaffer Hrdy, to name just a few, who regard altruism and generosity as just as “natural” as selfishness.

It is discouraging that at some fuzzy paradigmatic level, so little has changed between the 19th and 21st centuries, especially since a solution is so readily at hand. Let’s focus on the distinction between teaching generosity and altruism, and generosity and altruism as natural. Suppose that you have a choice between two social strategies, which would be regarded as generous and selfish, respectively, in everyday terms. Now suppose that I place you in a group of selfish people. You will probably elect to become selfish yourself, if only for self-preservation. If I somehow persuade you to be generous, you will suffer from my teaching.

Next, suppose that I place you in a group of generous people who follow the rule of expelling people who are selfish. You will wisely elect to be generous. You were not taught to be generous. You always had the option in mind and merely employed it under the right circumstance. If you weren’t smart enough to become generous and I were to teach you, then you would benefit from my tutelage.

In this thought experiment, generosity and selfishness are equally natural as behavioral strategies and either one can be successful, depending upon the circumstances. The role of teaching is subsidiary. If a person has both strategies in mind and employs them appropriately, then teaching is unnecessary. If teaching is necessary, then it must council the appropriate strategy for each circumstance, or the person will suffer as a result.

You might want to say that it’s selfish for the individual to behave generously in the right circumstance, but then you will need to distinguish between two meanings of selfishness; a local form (one of the two behavioral strategies) and an “all things considered” form (which can be either of the behavioral strategies).

I expect that most readers, including proponents of selfish gene theory, will agree with me so far. The next step is to consider the range of circumstances under which generous social strategies evolve as the “all things considered” selfish strategy. Dawkins and just about everyone else at the time thought that these circumstances were limited to interactions among genealogical relatives (kin selection) and individuals who directly exchange benefits (reciprocal altruism). The range of circumstances associated with group selection was thought to result in the evolution of selfish social strategies as the “all things considered” selfish strategy.

What has changed over the decades is a reassessment of the conditions under which generous social strategies can evolve. The sweeping statement that selection within groups favoring selfish strategies invariably trumps selection between groups favoring generous strategies has proven to be false. The basic scenario of within- and between-group selection is correct, but the balance between levels of selection often tips in the direction of between-group selection. Whenever this happens, the generous strategy becomes the globally selfish strategy. These conditions go far beyond interactions among genealogical relatives and narrow reciprocators, especially in our own species.

Against this background, I have indicated in bold what Dawkins got wrong in the passages quoted above, in my opinion, with a few comments pointing in the right direction in brackets.

We are survival machines - robot vehicles blindly programmed to preserve the selfish molecules known as genes.
These are claims that could have been made for Lorenz's On Aggression, Ardrey's The Social Contract, and Eibl-Eibesfeldt's Love and Hate. The trouble with these books is that their authors got it totally and utterly wrong. They got it wrong because they misunderstood how evolution works. They made the erroneous assumption that the important thing in evolution is the good of the species (or the group) rather than the good of the individual (or the gene). [Lorenz and Eibl-Eibesfeldt were uncritical about invoking group selection by modern standards, but there was nothing “totally and utterly wrong” about their argument that traits evolve “for the good of the group”. Both authors deserve to be carefully read and some of their observations make excellent sense from a modern evolutionary perspective—especially their observations about humans, because we are a highly group-selected species].

The argument of this book is that we, and all other animals, are machines created by our genes. Like successful Chicago gangsters, our genes have survived, in some cases for millions of years, in a highly competitive world. This entitles us to expect certain qualities in our genes. I shall argue that a predominant quality to be expected in a successful gene is ruthless selfishness. This gene selfishness will usually give rise to selfishness in individual behavior. [This empirical claim is unwarranted from a modern evolutionary perspective. The full spectrum of human social strategies, from the most selfish to the most altruistic, can potentially evolve in a Darwinian world, depending upon the circumstances, making them equally “natural”].

However, as we shall see, there are special circumstances in which a gene can achieve its own selfish goals best by fostering a limited form of altruism at the level of individual animals. 'Special' and 'limited' are important words in the last sentence. [Altruism and other “for the good of the group” social strategies can evolve under a far greater range of conditions than Dawkins and most of his colleagues imagined in the 1970’s. Of course, it remains true that altruism does not evolve under many other circumstances. The balance between levels of selection needs to be determined on a case-by-case basis]. Much as we might wish to believe otherwise, universal love and the welfare of the species as a whole are concepts that simply do not make evolutionary sense. [Whatever we might conclude about species-level selection, a subject that I will save for future articles, love and concern for the welfare of one’s group makes perfect evolutionary sense].

Let us try to teach generosity and altruism, because we are born selfish. Let us understand what our own selfish genes are up to, because we may then at least have a chance to upset their designs, something that no other species has ever aspired to do. [The emphasis on teaching is misplaced, as I have already indicated and will return to below].

Notice that I do not disagree with Dawkins on the basic concept of gene selfishness. Any gene that evolves does better than the genes that didn’t evolve, all things considered. If you want to call that selfish, be my guest. You will, however, need to carefully distinguish your “all things considered” definition from other definitions. Dawkins wasn’t wrong about selfish genes, but he (and many others) was wrong about what he thought followed from selfish genes. Moreover, enough progress has been made over the decades that there deserves to be a consensus about what I have indicated in brackets. Until there is a consensus among the experts that is communicated clearly to the general public, everyone will remain muddled about selfish genes.

One reason that I am passionate about this subject is because it is not an arcane academic debate. It is vitally important for promoting altruism, generosity, and other “for the good of the group” behaviors in the real world—something that Dawkins hopes for as much as I do. If you think that we are born selfish and must be taught to be good, you are overlooking evolutionary science as a solution to the problem. Your efforts to teach will be ineffective and might even cause harm, if the newly educated altruist encounters a world that favors selfishness. If you think that we born with a capacity for a wide range of social strategies, then you can consult evolutionary science as a solution to the problem. You can work to create environments that favor altruism as the most successful strategy in a Darwinian world. If you succeed, then teaching altruism will be effective and might even be unnecessary.