William Hamilton is widely acknowledged as one of the most important
theoretical biologists of the 20th century. In 1964, at the age of 27, he published
The Genetical Evolution of Social Behaviour. For several years it was
the most cited paper in the Journal of Theoretical Biology ever, and
this may still be so. Hamilton died in 2000 after contracting malaria in
Africa. The following interview took place at the University of Oxford, December
1996.
Some of your ideas were foreshadowed by Sir Ronald Fisher's explanation, in the thirties, of the distastefulness of some insects. In what way?
Fisher saw there was a difficulty about the evolution of distastefulness
in insects, as also in the further matter of the insects becoming brightly
coloured in order to advertise their distastefulness. He realized that if the
insect is actually eaten by the predator in the course of learning to avoid
them, then whatever made that insect conspicuous to the predator is obviously
disadvantageous. So he reasoned that the only way in which you could see that
kind of selection getting started would be if firstly, the insects were gregarious,
secondly, the group was a group of siblings, and thirdly, having tasted one and
found it awful, the predator would then leave the rest of the group alone. The
genes of the one eaten would then be promoted. Fisher also realised that this
was not such a strong form of selection, not as if it were the individual
itself that had a form of protection. He made some remark about the selection
going ahead at half the speed than it would have if it were direct selection.
And that was one key early statement of the selection principle concerning the
closeness of relatedness that I later came to develop.
You describe in part two of your 1964 article the post-reproductive behaviour of two kinds of moths. What was the general phenomenon being illustrated?
I noticed that someone had written about the post-reproductive spans of
two kinds of moths, one kind being cryptic, and the other kind warningly
coloured as in the case we have just discussed. The author had noticed that
the cryptic insect tended to die very soon after it had laid its eggs, whereas
the warningly coloured ones often have a long life after they had laid their
eggs. Again this could be interpreted in terms of the kinship principle in a
rather neat way. In case of the cryptic one, if there are any relatives around
in the neighbourhood at all, it is advantageous for the moth to give up its
life as soon as it finished its own main business, laying its eggs. Because if
it is around and the predator detects the moth and eats it, then that is a step
in the predator learning to detect other moths, perhaps including those which
have not yet laid their eggs. So by causing itself to die soon after it has
laid its eggs, it is actually doing a service to its cousins which are in the
neighbourhood. Quite the contrary holds for the warningly coloured one. Once
it has laid its eggs, it is in a position freely to use its warning-colours to
warn everyone it can. So it should continue to live and wait, actually expose
itself to being tasted by the predator, because that would be a step in
teaching the predator to avoid its relatives. It was gratifying that all this
theory fitted very well with the findings.
In moths both sexes have wings, but in some other insects either the males or the females are wingless. Why?
Gene dispersal is a very crucial evolutionary phenomenon, because where
you are, your descendants almost certainly will have to die out. So there has
to be a continual search for other places to colonise. And therefore for a
female to sacrifice wings without having some other way of dispersal for her
offspring would be a deadly mistake sooner or later. In fact we find that in
all the cases where females have become wingless, there is some other way in
which they or their offspring are dispersed to other localities. Most commonly
they have a young stage, a larva, which is very mobile. Either it may climb
onto other insects, or onto a twig and from there be dispersed by wind. Often
such larvae have long hairs which enable them to balloon on the wind very
satisfactorily. In some species the wingless female is carried by the winged
male, -in his arms almost literally. During this flight-time he is mating her,
and finally he drops her off in a place suitable for egg laying.
As to the males becoming wingless, if the male can inseminate a female
who is mobile herself, then he doesn't have to worry about wings too much,
because his genes can be carried off by the female. This often happens in cases
where wingless males mate their close relatives. So I think we can find some
sort of a rationale for many particular cases; but as far as I know there is no
very sweeping theory that explains why in some groups there is male
winglessness, and in others female winglessness.
You went to Brazil in 1975 to study the fig-wasp, where the males don't have wings, and often also don't have a mouth.
I didn't actually go to Brazil to study them, I went to study life in
rotting wood, but I ended up studying fig-wasps because the rotting wood was
dry and not in good condition. The fig is a little world in itself. In some
species of fig-wasps the males indeed have no mouth; these are entirely
fighting and mating machines, a very strange kind of an animal with a very
short life as an adult and mating as quickly as possible, and fighting
lethally whenever rivals intrude and compete.
Before going to Brazil I had known there was something very special for
my theorising about fig wasps. I knew how they participate in this incredible
symbiotic pollination system in the fig-trees. In adapting to living inside
the fig-fruit they developed various marvellous adaptations, including completely
wingless males with very specialized habits, and with exactly the biased
sex-ratio's which my theory predicted they should have.
Why should they have biased sex-ratio's?
Well, it is a case where the males are in a unique confining situation
inside the fig. They hatch out as adults from the gall flowers, and the easiest
-indeed the only- females to mate are the females who are also hatching inside
the same fig. So they do so. Theory says that if males mate with their sisters,
and their sisters are capable of storing sperm, then you must expect the
proportion of males to be cut down drastically. It is adaptive because a female
mother who programs her eggs to become either males or females, should do much
better if she produces a lot of females, and just enough males to fertilize
them. In that way she would get more descendants. That bias is very nicely
illustrated in what happens in figs. Other people that followed and tested my
reasoning further have shown that it is illustrated with an accuracy that I
would never have dared myself to expect. So this became a really strong and
well-upheld prediction out of evolutionary theory.
Why do males sometimes engage in lethal fighting with other males, while at other times they tend to bluff?
For some species of fig-wasps there is not much point in bluffing inside
the fig, because there is no time to 'live and fight another day' if they
accept a bluff. Everything is over in a few hours, and if you don't fight now
and try to win, then you won't be given another chance.
I think I saw bluffing with the giant Chilean stag beetle, where I did
find that in all the tournaments that I held between males, inducing them to
fight one another in the presence of a female, it always turned out to be the
second largest one who was the overall victor. This suggests rather that those
with the biggest 'tongs' were not actually as strong as they seemed to be. And
they live in a situation where I can imagine that bluffing would pay off. Stag
beetles are quite long-lived, and there are many flowers on the trees they
could visit where females are arriving and so it might be worthwhile to pretend
that you are bigger than you are, in the hope that a male will accept that you
are stronger and go away to other flowers. Then you could win the local
contest; bluff would have paid.
You use a lot of mathematics in your work, and you write (Narrow Roads of Gene Land I, 1995: 230): "I had realized from experience that university people sometimes don't react well to common sense and in any case most of them listen to it harder if you first intimidate them with equations."
Equations seem to frighten a lot of people; if you come at them with an
display of mathematical strength then they often back off. With me you might
call it a kind of bluff. A nice anecdote about people's fear of mathematics
occurs to me. There was an IRA bomb discovered in this town, which had never
actually been fired, and yet had not been found either for about a year. Can
you guess where it was planted? The person put it behind the calculus books in
one of the major bookshops of Oxford! I guess he choose calculus-books because
he reckoned that those would be the ones least often be pulled off the shelves.
This perhaps illustrates the general fear of mathematics, but I guess another
interpretation might be that, second to Brits in Northern Ireland, the bomber
just hated maths!
Is there not a risk that someone writes an article filled with equations but without ideas, and it is accepted as science just because it looks so scientific?
That's exactly it. A lot of work does get by, by having an aura of mathematics about it. Whether it is serious and necessary or not, it certainly impresses people. You can find biologists who are very hot on this issue, like my old colleague Richard Alexander. He doesn't like mathematics in evolutionary theory, he says you can have all the same thoughts without it. So these biologists would call it bluff, and I think very largely they are right. If you have a simple idea, state it simply, and forget about the mathematics.
Nevertheless, you use mathematics.....
I use it. Often I use mathematics because I needed to straighten out my
own ideas. I have a somewhat illogical brain, and unless I put it through the
mill of mathematics, I can continue to believe in the impossible for a long
time. But eventually, either by mathematical argument or by
computer-simulation I suddenly realise that something I thought followed in
fact does not follow, and I abandon that and come to see the truth. Sometimes
that math goes into the paper in order to convince other people.
And then there is DNA, chromosomes, genes, etcetera. This can be quite intimidating to social scientists. Do you need to know a lot about genetics and math to understand evolutionary theory?
I do think you do need to know the basics of genetics. I am not sure you
need to know a tremendous amount about modern molecular genetics. I always
found that good old standard Mendelism serves me quite well, and the modern ideas
have not really changed the picture very much. I also think in the mathematical
field you just have to know something about probability theory, to understand
how genes work in evolutionary processes, and so fundamentals of probability
and statistics are just about essential in understanding modern Darwinism.
A general question. Do living organisms behave as if they want to pass on their own genes, or do they behave as if each of their genes is trying to replicate itself, possibly at the expense of other genes of the same genome?
This is a very deep and difficult question I think. One's impression is
that there is a conflict between selfish genes, but largely it is being
over-ridden by a kind of democracy that has arisen in the genome. It combines
to suppress this intergene conflict and the outcome is that the organism acts
largely as a whole.
You write that evolutionary ideas "turn out to have, or are perceived to have, the unfortunate property of being solvents of a vital societal glue". What kind of glue is this?
The glue that I am thinking of is various myths that tend to hold
societies together. When these myths are wrong, I think that they exist because
they have helped people to be more cooperative than they would be if proceeding
with a full understanding of evolutionary theory. This is the most unfortunate
fact about teaching evolution. I also think that it is the threat that is dimly
perceived by all religious people. They think that if people 'believe'
evolution instead of believing, say, the gospels, they will no longer be able
to celebrate simple honesty, or kindly and warm feelings toward others, as
unequivocally 'good'.
I think they exaggerate the danger, but they don't exaggerate a nothing.
There is a danger of that kind.
How are evolutionists trying to deal with this problem?
They are trying to deal with it by showing that the implications of a
fully rational and evolutionary theory of behaviour, and that includes human
behaviour, is not such a nasty thing as it may seem at first. If you believe
that we evolved out of animals, -are animals-, and have the same kinds of
drives, it doesn't mean that we have to be selfish and inhumane. When you
fully work out the consequences of the rules of kinship and of reciprocation,
and ensure maintenance of the standards implied, you will see that the outcome
is in fact quite a moderate kind of behaviour, avoiding evil and as good in
holding the society together as are the religious myths. Indeed, under a
rational theory we should be able to do better for human happiness by avoiding
various naive errors.
You write that long ago there was a man who claimed that humans descended from bears, and not from apes. As his theory got rejected and the man got more and more isolated, he probably looked like an angry bear. Didn't this description for a long time fit sociobiologists? "The world doesn't understand us and doesn't like us"?
I guess it fits anyone who feels himself to be very isolated, so I
guess and hope we have gone recently from being angry bears to being rather comfortable
and accepted bears. I wish I could remember who it was that founded this
ridiculous theory about bears. I am sure that it exists, but I have not managed
to recall the book. Perhaps one of your readers can help to remind me.