Dawkins's Biggest Mistake: Giving Genes a Mind

Richard Dawkins is one of the clearest scientific communicators alive. The Selfish Gene (1976) is a genuine landmark — it took the gene-centric view of evolution and made it unforgettable. That is the achievement. Unforgettable prose has a way of becoming the thing people mistake for the underlying fact.

Then he handed the gene a personality.

The Metaphor That Got Away

Read the opening chapters of The Selfish Gene and you will find genes described in ways that load them with agency:

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

“Now they swarm in huge colonies, safe inside gigantic lumbering robots, sealed off from the outside world, communicating with it by tortuous indirect routes, manipulating it by remote control. They are in you and in me; they created us, body and mind; and their preservation is the ultimate rationale for our existence.”

“A predominant quality to be expected in a successful gene is ruthless selfishness.”

“The genes are the master programmers, and they are programming for their lives.”

Selfish. Ruthless. Master programmers. Programming for their lives.

That is not shorthand for a mechanism. That is a description of intent.

And this is not accidental phrasing scattered through an otherwise careful book. The anthropomorphism is the engine. Strip out the intent language and you lose the narrative drive, the sense that something is at stake. Without a selfish gene there is just chemistry, and chemistry does not sell books the way characters do. Dawkins understood that. He made the choice.

What Genes Actually Have

A gene is a sequence of nucleotides that codes for a protein or regulates expression. It has:

Machinery. Promoter regions, transcription factors, splicing signals, regulatory sequences. Chemistry that produces effects.

Response mechanisms. Expression patterns that change based on environment — methylation, histone modification, feedback loops. Things that react, not things that decide.

That is the whole list. There is no preference structure. No goal. No drive to survive. Natural selection filters which genes persist based on their phenotypic effects in real environments — but the gene itself does not know it is being filtered. It does not know anything. It does not want anything.

He Knew — and Included a Disclaimer

Dawkins included a disclaimer. In The Selfish Gene he wrote:

“I am using the word ‘want’ as shorthand for ‘has a phenotypic effect which… tends to result in.’”

And later:

“Be warned that if you wish, as I do, to build a society in which individuals cooperate generously and unselfishly towards a common good, you can expect little help from biological nature. Let us try to teach generosity and altruism, because we are born selfish.”

One sentence of caveat against three hundred pages of master programmers and ruthless selfishness is not a correction. It is cover. A writer who genuinely wanted to prevent the anthropomorphism would have written the book differently — returned to the caveat every time the metaphor got particularly vivid, made the disclaimer impossible to miss. Instead, he tucked it in early and spent the rest of the book making the metaphor more vivid.

The disclaimers are there. They did not do the work he needed them to do, and he was good enough at his craft to have known that.

Why It Matters

The anthropomorphized gene has been doing damage ever since. It is the scaffold for every “genes for X” story in popular science — genes for altruism, genes for addiction, genes for risk-taking. Genes become agents pursuing outcomes. Evolution becomes a kind of purpose. The metaphor Dawkins deployed with such precision is now the water everyone swims in.

It also muddies what natural selection actually is: a filter on variation, operating blindly, with no direction and no goal. The direction we see is retrospective — we call survivors “fit” because they survived, not because fitness was a target.

Genes did not “want” to survive. The ones that produced effects leading to reproduction are the ones we observe — maybe. The population genetics math is less clean than the story implies. Most gene fixation happens through genetic drift, not selection: Kimura’s neutral theory showed that the overwhelming majority of molecular variation is selectively neutral, propagating by chance rather than fitness. Haldane worked out that the cost of selection puts a hard ceiling on how many beneficial substitutions a population can fix per unit time — far fewer than the diversity of adapted traits would suggest. So even the core claim, that the traits we see are the ones selection drove to fixation, is doing more work than the math supports. Collapsing drift and selection into a single “survival” story is how you end up with evolutionary just-so explanations dressed up as mechanisms.

The gene-as-unit-of-selection is a defensible view. Dawkins’s error was not in the claim but in spending fifty years illustrating it with language borrowed from purpose and desire — and then being surprised when people took the language literally.

Though once you strip away the intent language and look at what remains, there is a quieter question. Natural selection says: the traits we observe are the traits whose bearers reproduced. That is true. It may also be almost entirely a restatement of what we already saw — a description of the pattern, not a cause of it. Whether that constitutes an explanation or a very precise way of cataloguing outcomes is something the vivid metaphors have been quietly helping us not ask.

Selfish Gene quotes are from the 1989 Oxford University Press edition.