No, The Selfish Gene does not predict that organisms should be selfish

Photo by Sangharsh Lohakare on Unsplash

Consider this recent response from Huemer when asked whether evolution can fully explain our moral intuitions. This starts at 6:05 in the video, which I’ve timestamped:

Starting at 6:26 Huemer claims that evolution would predict that selfish behavior and that this is something you’d learn from the book The Selfish Gene by Richard Dawkins. This is not true.

His initial response to the question of whether our moral intuitions and judgments can be fully explained by evolutionary psychology is:

Sort of and sort of not […]

I am sympathetic to this reaction. I think evolution can go some way in explaining moral judgment, but it cannot fully explain it…at least not directly. Part of the reason for this is that I do not believe every facet of human cognition is a direct product of innate psychological mechanisms. Instead, I endorse gene-culture coevolutionary accounts which emphasize the centrality of culture in shaping human cognition.

Natural selection might predict we will communicate, but won’t predict that we’d speak French or Malay in particular. It may predict that we’d fight, but won’t predict the specific weapons we’d use or what we’d make them from. It may predict that we’d create games or other forms of recreation, but won’t predict that we’d invent chess or baseball in particular.

The vagaries of culture fill in the gap, and the specific details of the culture and practices of any given population will be contingent on historical idiosyncrasies that cannot be directly inferred from knowledge of our evolutionary history unless this included all the fine-grained details of each culture that arose. My own perspective is that the notion of a distinctive normative domain of “morality” is historically contingent, and arose only in some populations but not others (see Machery, 2018 for a similar view).

Furthermore, I don’t believe there is any principled way to distinguish moral from nonmoral terms (Stich, 2018); moral judgments don’t represent a natural kind (Machery & Stich, 2022; Sinnott-Armstrong & Wheatley, 2012), we don’t have an innate capacity for distinctively moral cognition (Machery & Mallon, 2010), and much of the emphasis on morality is a byproduct of a WEIRD/Western-centric approach to normative theorizing (Henrich, Heine, & Norenzayan, 2010).

I don’t know if Huemer would agree with any of that, but there may be some overlap in our skepticism about the degree to which evolution can explain human morality. Even so, my main concern is not where Huemer and I may find overlap in our views, but in how Huemer characterizes The Selfish Gene. Huemer goes on to say:

You can come up with evolutionary explanations for virtually everything, right? […] You can come up with explanations for morality in the same way that you could come up with an explanation for anything that happened […] But could you come up with explanations that you would have anticipated before you knew what our actual moral beliefs were? No, probably not.

Depending on the specifics, I might agree that there are certain distinctive moral attitudes or practices one couldn’t readily predict just by understanding human evolution, because culture may lead people to hold unusual and highly specific moral practices for which there is little direct (though there could be an indirect) evolutionary rationale, e.g., a taboo against saying specific words, or eating certain foods, or wearing certain colors of clothing. If you knew everything about human biology but nothing about a particular culture, and that culture tabooed wearing blue because they happened to associate it with death, this is not something you could predict without specific knowledge of that culture’s history. But that’s not the direction Huemer takes this. He continues:

You learn from evolutionary theory that evolution selects for selfishness. This is what you learn if you read great books like The Selfish Gene.

This is not accurate. Dawkins argues that genes are the fundamental unit of selection, and describes them in an anthropomorphic and metaphorical way as “selfish.” However, Dawkins stresses that while genes may be selfish, it does not follow that the organisms that genes give rise to are selfish.

Dawkins draws a distinction between genes as the fundamental unit of selection, and organisms as the “vehicles” genes cooperate to build so as to maximize the amount of themselves that are copied into future generations. This has at least two implications.

First, it means that from a “selfish” perspective, genes actually have an incentive to cooperate with other genes. Dawkins likens this to a rowboat: you all have to row together at the same pace and speed or you’ll be thrown off course. Likewise, the destiny of any one gene is bound up in the destiny of the other genes in a given genome, and as such it is in their individual interest to cooperate.

Second, once you focus on genes as the unit of selection, this actually helps explain why some organisms are not entirely selfish: because copies of a given gene are not only located in the body of the vehicle in which they reside, but in the organisms around oneself, and because making deals with others can be mutually beneficial, it is adaptive for genes in some situations to promote altruistic or cooperative behavior at the level of individual organisms.

This latter point is one of the central themes of the book, which Dawkins points out is more about how to account for altruism than selfishness. I’ve gathered some excerpts from the introduction of the 40th anniversary edition of The Selfish Gene to illustrate what the book’s key themes are, and to provide evidence of Huemer’s mischaracterization. First, we can see that the actual contents of the book have more to do with explaining altruism than selfishness:

The best way to explain the title is by locating the emphasis. Emphasize ‘selfish’ and you will think the book is about selfishness, whereas, if anything, it devotes more attention to altruism. (viii)

Dawkins is also explicit that the only firm attribution of predicted “selfishness” applies to genes, but is not necessarily applicable to other levels of selection (individuals, groups, species, etc.):

The correct word of the title to stress is ‘gene’ and let me explain why. A central debate within Darwinism concerns the unit that is actually selected: what kind of entity is it that survives, or does not survive, as a consequence of natural selection. That unit will become, more or less by definition, ‘selfish.’ Altruism might well be favored at other levels. (viii)

Dawkins is also clear to note that non-selfishness at the level of individual organisms is a result of selfishness at the level of genes, then goes on to reference the two primary mechanisms that explain how non-selfish behavior is selected for:

Or does natural selection, as I urge instead here, choose between genes? In this case, we should not be surprised to find individual organisms behaving altruistically ‘for the good of the genes’, for example by feeding and protecting kin who are likely to share copies of the same genes. Such kin altruism is only one way which gene selfishness can translate itself into individual altruism. This book explains how it works, together with reciprocation, Darwinian theory’s other main generator of altruism. (vii-ix)

Given that so much of the book is dedicated specifically to outlining, in excruciating detail, how and why a gene-centric understanding of evolution makes sense of altruism at the individual level, it is fair to say that this is what one ought to predict, if one has a proper (from the book’s perspective) understanding of evolution. The Selfish Gene categorically does not predict selfishness at the individual level. It explicitly argues for the opposite of this. The rationale for the title is, in part, rooted in the quasi-paradoxical observation that selfishness at a lower level (genes) can give rise to and directly account for non-selfishness at higher levels, i.e., selfish genes predict non-selfish behavior (at least some of the time). This isn’t just some auxiliary point; it is arguably the central thesis of the entire book. And Huemer not only gets it wrong, but gets it exactly backwards.

Dawkins also explicitly excises misleading passages from the first chapter that suggest selfishness at the individual level:

[...] I do with hindsight notice lapses of my own on the very same subject. These are to be found especially in Chapter 1, epitomized by the sentence ‘Let us try to teach generosity and altruism because we are born selfish.’ There is nothing wrong with teaching generosity and altruism, but ‘born selfish’ is misleading [...] Please mentally delete that rogue sentence and others like it [...] (ix-x)

And when discussing titles, considers stressing altruism or cooperation as viable alternative titles:

The Altruistic Vehicle would have been another possibility. (x)

Another good alternative to The Selfish Gene would have been The Cooperative Gene. It sounds paradoxically opposite, but a central part of the book argues for a form of cooperation among self-interested genes. (x)

Here’s a clip of Dawkins talking about another possible title, The Immortal Gene:

I cannot stress enough: The Selfish Gene is not about selfishness. It does not suggest in any way that organisms should be expected to be selfish. The term “selfish” was used as a way to anthropomorphize genes, allowing one to characterize how natural selection operated in a quasi-agentic fashion. The title of the book, and the contents of the book, don’t really have anything to do with predicting that organisms would be selfish, though the book does spend a great deal of time explaining why some organisms will be under certain conditions unselfish. It was a poor choice of title, and it looks like Dawkins himself suspects this may be the case.

Finally, it’s worth noting that one of Dawkins’s first remarks is to mention philosophers mischaracterizing his work:

Many critics, especially vociferous ones learned in philosophy as I have discovered, prefer to read a book by the title only. (viii)

I worry that something like this may have happened. I find it hard to believe anyone could read the book and come away with the impression Huemer expressed in this video. Either way, this is a lose-lose for Huemer. Either he made assumptions about the contents of the book without having read and understood it, or he did read the book but somehow catastrophically misunderstood its contents.

(An earlier version of these remarks was initially posted as a comment to the video).

References

Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world?. Behavioral and Brain Sciences, 33(2-3), 61-83.

Machery, E. (2018). Morality: A historical invention. In K. J. Gray & J. Graham (Eds.), Atlas of moral psychology (pp. 259-265). New York, NY: The Guilford Press.

Machery, E., & Mallon, R. (2010). Evolution of morality. In J. M. Doris & The Moral Psychology Research Group (Eds.). The moral psychology handbook (pp. 3-47). New York, NY: Oxford University Press.

Machery, E. & Stich, S. (2022). The Moral/conventional distinction. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Summer 2022 ed.). https://plato.stanford.edu/archives/sum2022/entries/moral-conventional/

Sinnott-Armstrong, W., & Wheatley, T. (2012). The disunity of morality and why it matters to philosophy. The Monist, 95(3), 355-377.

Stich, S. (2018). The moral domain. In K. Gray & J. Graham (Eds.), Atlas of moral psychology (pp. 547- 555). New York, NY: Guilford Press.