UP purchase from amazon purchase from ibs

Notes by RBJ on

The Selfish Gene

by Richard Dawkins

This is Richard Dawkins' first and still (when I first wrote this!) his most popular book, and is now regarded as a classic in the field. As well as giving a popular account of evolution and the idea of the selfish gene as the unit of natural selection, it also introduces the meme meme.

Chapter 1Why are People
Chapter 2The Replicators
Chapter 3The Immortal Coils
Chapter 4The Gene Machine
Chapter 5Aggression: stability and the selfish machine
Chapter 6Genesmanship
Chapter 7Family Planning
Chapter 8Battle of the Generations
Chapter 9Battle of the Sexes
Chapter 10You scratch my back, I'll ride on yours
Chapter 11Memes, the new replicators
Chapter 12Nice guys finish first
Chapter 13The long reach of the gene

Chapter 1 - Why are People

Pretty clear statement of the purpose of the book, which is, to argue the view that "the best way to look at evolution is in terms of selection occurring at the lowest levels of all", by which he means on gene's not on individuals, or species or at any other level.

Dawkins claims that genes can be expected to have the quality of ruthless selfishness and that, except in special and limited circumstances this leads the individuals carrying the genes to be selfish.

"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."

The following definition is given, for the purposes of the book:

An entity is altruistic if it behaves in such a way as to increase anothers welfare at the expense of its own.
Where welfare is: chances of survival


I had intended to comment on this book chapter by chapter as I read it but it didn't work out that way. These conclusions are added years later, and most of the detail of the book is now gone from my memory (or slotted in under other headings). Having been provoked by an unwitting enquiry about where the rest of the notes are, I am going to tell you my conclusions.

The book is a great little book, and well worth reading, but, when I stand back and ask "what is this book trying to do?" I do not arrive at a favourable conclusion.

The two most prominent messages delivered by the book are:

  1. that evolution takes place entirely through the effects of natural selection on the propagation of genes
  2. that genes are selfish, not altruistic
The emphasis on point 1 is mitigated by the material Dawkins writes about memes, in which he appears to believe that human cultural evolution may mitigate or supersede the effects of genetic selection. Viewing evolution in terms of selection on genes is very illuminating, but I have not been persuaded that all other perspectives are thereby invalidated. Dawkins concedes that memes may make a difference, but, given that something can override the effects of natural selection on genes, he gives us no basis for judging where the limits of these other influences lie. Could the transmission of behavioural traits by non-genetic means in animal species have analogous effects, even in the absence of anything we would recognise as cultural evolution? It is hard for me on the basis of this book to believe that the alleged exclusive role of natural genetic selection in evolution is other than a dogma. Point 2 is worse. The claim that gene's are selfish is presented at first as a mere tautology. Very roughly, the idea is that a gene will proliferate in the gene pool if its effects are conducive to its proliferation. How do we know whether its effects are conducive to proliferation? If the gene proliferates. So a gene proliferates if it proliferates, or it has a tendency to proliferate if it has a tendency to proliferate.

Its important to note here that it is not me that is claiming that the principle is tautologous. It is Dawkins.

Suppose we take him at his word. The consequence is that this is not a scientific hypothesis, and furthermore it is likely to be so trivial a tautology that its unlikely to be much help in reasoning. However, this is in fact the mainstay of the book, a substantial proportion of the claims in the book are justified by reference to this tautology. We are entitled, on the simplest epistemological grounds, to suspect that some equivocation (possibly unwitting) is taking place. This is a common route to fallacious conclusions, use one definition of a term when establishing your key conjectures, and then allow the meaning to drift conveniently when it comes to using the now established results.

The use of the term "selfish" in this context is particularly pernicious, because of its moral significance. Even if Dawkins' principle were not to be established as a tautology this term would be completely inappropriate. In the parlance of Ryle this is a category error, a gene is just not the kind of thing which can be considered selfish or altruistic. To be selfish is to make a decision or perform an act which is to your benefit at the expense of others. Genes do not make decisions or perform acts.

Dawkins is very definite about what he means by "selfish". Something is selfish if it behaves in such a way as to increase its own chances of survival at the expense of some others chances of survival.

Much of the book is given over to explaining how many aspects of behaviour which might at first glance appear to be altruistic can be seen from the genetic viewpoint to be selfish. It is success in this endeavour which makes the book seem so enlightening.

The effect is that we may be persuaded that all apparent altruism, when examined in the light of true wisdom, can be shown to be accounted for by selfish genes. The corrolory is that geniunely altruistic behaviour can only be a rare (maladaptive) exception to the norm, a freak of nature. The probable outcome of this theory and this misleading terminology is fatalistic pessimism about human nature and the future of society.

Altruism is an observable phenomenon, and is not nearly so rare as some would have us believe. Any theory of evolution which fails to account for this is thereby falsified. Of course, on the tautologous rendering, any genetic predisposition to altruism, if prevalent in the population, must be to the advantage of the genes which encode it. But the tautologous rendering, by contrast with its alternative, gives us no basis for judging whether altruism as a human trait will be prevalent or not.

UP HOME © RBJ created 1997/1/12 modified 2007/1/25