RE: memetic fitness and suicide

Gatherer, D. (
Thu, 22 Jul 1999 13:21:55 +0200

Date: Thu, 22 Jul 1999 13:21:55 +0200
From: "Gatherer, D. (Derek)" <>
Subject: RE: memetic fitness and suicide
To: "''" <>


One more related question. Natural selection is a tautology if fitness is
an absolute frequency measure - survival of fittest (which is a lousy
description, because it is neither about survival or fitness in any commonly
understood manner), becomes survival of the survivors.

So surely evolutionary (memetic) fitness must be a relative concept (tied
to a particular ecological niche/context),


In evolutionary genetics, the usual method is to sample a population at a
certain time, and then sample it again at a later time. In the lab, this
can be done by rearing, say, fruit flies and scoring each generation for
phenotypes. If one knows the relationship between phenotype and genotype,
one can construct the Fisherian table [Note that this is the problem for
standard Dawkins B memetics - how do we know the relationship between
memotype and phemotype? - answer: we don't - so that is where it falls down.
Cavalli-Sforza and friends get round this by not having a memo/phemo
distinction, they tag the cultural trait onto the genotype to give a
phenogenotype of mixed genetic-memetic elements] In the wild one can
collect butterflies one summer and then again the next summer and score
likewise (EB Ford's 'Ecological Genetics' first published, I think, in 1940
contains wonderful descriptions of evolutionary genetics in the wild, mostly
on English butterflies)

these calculations are summed over the whole population one has collected.
It would however be possible to do it niche-specifically. Indeed this has
been done in human population genetics to demonstrate that the ecological
niche can be important.

For instance yam cultivators tend to get malaria in West Africa, because the
yam fields are mosquito-breeding grounds. The yam farmers consequently have
a high frequency of sickle cell anaemia, which gives malaria resistance to
the heterozygote. Their non-agricultural neighbours have less malaria
resistance, because they haven't been exposed to the selective agent (ie. no
puddles, no flies, no bites in the night) (Wiesenfeld 1967)

and critically a *propensity* to
contribute to future generations, not an absolute measure of contribution.


It's measured now, so it is an absolute measure of the fitness of present
phenogenotypes, but then it is used to project into the future. So
evolutionary genetic textbooks are filled with graphs of how allele
frequencies would change over generations given certain selective pressures
etc. This has been done in the Cavallian tradition, for instance Feldman
and Laland (1996) Fig.1 have a 4000 generation projection of how meme
frequencies for female infanticide and the genetic frequency of female
births would evolve given a certain initial assumed set of selective
pressures. So yes a propensity, but derived with fingers crossed from
current absolutes.

To translate this into a concrete memetic example - why does suicide qua
acquired (social learned rather than inherited -which it may also be)
behavioural strategy persist? Because, of course it is, evolutionarily fit.
Please can we have a stab at unpacking this and provide a proto-memetic
explanation for suicide.

The Cavallian explanation is that cultural selection (ie. highly effective
transmission) can override natural selection. Where the cultural trait is
horizontally transmitted, it doesn't matter if vertical transmission slows
to zero. So runaway cultural selection could drive a population to

To the rescue (perhaps) comes Takahasi (1998), who develops the Cavallian
approach to make a model incorporating genetic preferences with memetic
cultural traits. The result is that genetic aversions to highly culturally
selected, but maladaptive, memes tend to halt the runaway cultural selection
process before extinction is reached.

Thus suicide may be culturally highly selected (and it certain can seem like
that during a suicide epidemic), but the reason suicide epidemics don't wipe
us out completely is that there are enough genetically aversive individuals
around, and such genetic aversion has evolved as a response to culturally
imposed selection pressure in the past.

Feldman MW and Laland KN (1996) Gene-culture coevolutionary theory. Trends
in Ecology and Evolution 11, 453-457.

Takahasi K (1998) Evolution of transmission bias in cultural inheritance.
J. Theoret. Biol. 190, 147-159

Wiesenfeld SL (1967) Sickle-cell trait in human biological and cultural
evolution. Science 157, 1134-1140.

This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)