Bad genetics and bad evolutionary psychology make for a dangerous cocktail, says Gabby Dover

Dear Mr Darwin, I've been stimulated to write to you again. You might recall how I suggested that our previous correspondence on modern genetics and evolution be published, with your permission. You never got back to me on that, I suspect because we were immersed in a discussion on the genetic determinism of human nature, which did not leave us much time to make plans for the future.

That discussion ended on the high falsetto note of desire of the great Russian bass singer, Feodor Chaliapin, as he spontaneously exercised an act of free will in the rendition of a particular song on a particular day, without so much as a glance at his genetic makeup or at his cultural upbringing. No nature; no nurture - just Chaliapin.

That was my way of expressing the argument that our particular behaviours and psychologies are singular acts of free will, day on day, exercised by uniquely formed individuals.

ÖйúAƬ

A person is no more predictable as a result of the particular constitution of the genes than from the interstices of a particular culture.

There is no blueprint or recipe in our genome instructing the genes how to make a human. Human development is consequential on a time-dependent unfolding of interactions among genes, and between genes and the environment that is specific for each individual. The behavioural glory of the emergent individual cannot be foretold in any interesting or accurate way, even were we to have all of an individual's genes strung out in a line before us.

ÖйúAƬ

Sir David Weatherall emphasises that even for genetic disorders under single gene control, there is a remarkable complexity. Individuals with the exact same mutation in the beta-globin gene do not necessarily suffer from beta-thalassaemia. Such individual differences are not explicable either by environmental factors.

How much more difficult is it to define a causal predictive link between genes and organisms for complex behavioural or disease traits, which are influenced by many genes.

Arguments in support of the determinism of generic aspects of human nature by specific genes - which have supposedly manipulated human behaviour since hunter-gatherer days - inspire little interest. To say that, for example, men are more aggressive than women is a statement only about the non-existent average man and average woman.

Some believers in evolutionary psychology (EP) claim that the potential for aggression or rape is a genetically determined "universal" of male behaviour. But variation, not universals, is the name of the game of genetics. There is no such thing as an average, normal man and woman, supposedly surrounded by a bath-ring of deviants.

It is both bad EP and bad genetics to extrapolate from such universals how a particular individual might shape up. The overlap between the sexes in the distribution of, say, heights is extensive. We could do only marginally better at predicting an individual's sex from a randomly selected measure of height than were we to toss a coin over it. Society consists of unique, unpredictable individuals, not averages.

The same goes for other shaky propositions of bad EP that have raised their heads from the murky swamps of our evolutionary past - IQ, shyness, risk-taking, extroversion, creativity, bullying, post-natal depression, obesity, criminality, aggression, sexuality, altruism, suicidal tendencies, vulnerability to stress, rape, beating up step-children, ability to sing love songs, physical beauty, and so on.

I recognise that some of the more sensationalist just-so stories cannot be laid at the door of EP professionals. There is an ignorance stalking the land about genes and their role in evolution and development, which is in danger of retarding the scientific study of biology and human nature. EP needs to start putting its house in order to sort out the good from the bad or downright ugly.

At the heart of the debate lies the relationship between the genes and the organism. In our earlier correspondence, Charles, I introduced you to the bizarre ways of genes in development and to the subtle ways by which a number of evolutionary processes have shaped and conditioned life around us, including ourselves. New genetic discoveries over the past two decades have told us that most genes are shared by most organisms. The famous HOX genes, first discovered in insects, where they busy themselves controlling early development, exist in species as "high" as humans and as "low" as worms.

ÖйúAƬ

What distinguish one species from another are the patterns of regulation as genes switch each other on and off. A given gene can be switched on and co-opted to interact with subsets of genes involved with a wide variety of developmental processes; and a given process can be influenced by a wide variety of genes. Such combinatorial promiscuity among genes is a consequence of genetic redundancy (genes and their controlling elements having multiple copies) and modularity (like Lego - a few basic genetic building blocks can give rise to an infinite number of structures).

Hence, not all aspects of successfully evolved functions (for example, eyes, brains, consciousness) are the consequence of natural selection involving specifically designed genes. New functions could have been established as co-opted "exaptations", riding on the back of adaptations selected from a variety of alternative roles of genes. The many-on-many relationship between genes and organisms undermines any simple notions that there are specifically naturally selected genes for every behaviour we care to consider.

Furthermore, natural selection - the preferential reproduction of variant individuals passing their genes to the next generation - is not the only show in town. Genetic drift and molecular drive (the spread of new genetic variants resulting from ubiquitous, internal processes of "turnover" in species genomes) also contribute to the evolution of genetic interactions that govern development.

ÖйúAƬ

Charles, I cannot recap all the examples I gave you in our earlier correspondence. My point in this letter is that bad EP is spreading like a deranged meme.

I was half-dozing with my two sons recently, watching The Ten Best Love Songs on television, when out popped an EPist with an explanation of why men (he says) sing better love songs than women. Yes, you've guessed it: it is an expression of sexual display forged by the relentless forces of selfish gene selection.

That most men and women cannot sing for toffee is of little concern to our bad EPist. Pick a nice love song and the singer can be predicted to be male. Pick a male and he can be expected to sing an evolutionary burnished love song. Should I go on?

The Survival of the Prettiest (another bad EP offering in book form from Nancy Etcoff) received uninhibited attention by reviewers unable to spot the genetic howlers. As the title says, female beauty confers a positive advantage in the evolutionary game of mating. But even if the beautiful people have first pick, the uglies do eventually make it in a near monogamous species such as ours.

Indeed, with the inevitable random shuffling of genes and chromosomes by the sexual process, it is quite possible that some of the progeny of uglies are indeed more beautiful than those of the beauties. There is a well-known "regression to the mean" in traits governed by multiple genetic and environmental inputs.

Charles, we should not be taking such conjectures too seriously, but these examples reflect the unfortunate misappropriation of EP as it enters our contemporary popular culture.

"There's no such thing as a Jewish disease." So wrote an alarmed chief rabbi in The Times a few years ago, in response to an article on the prevalence of Tay-Sachs disease among Ashkenazi Jews. He was right of course, but for all the wrong reasons. There is no such thing as a Jewish gene, or a cow gene or a maize gene. Genes are not infectious agents that can be labelled "foreign" as they move, or are intentionally moved, from one species to another. A gene has no meaning outside of its interactions with tens of thousands of other genes that make up a species genome.

It is genetically incoherentto claim that transferring a human gene into a cow makes the cow part human; or that an "animal" gene into a plant stops the vegetarians in their tracks. Both of these sensational assumptions have been voiced in the more popular newspapers and repeated by high-school biology teachers.

A cow is a cow is a cow - the product on the end of a twig of the tree of life that took root 4 billion years ago. The individual cow (excluding environmental factors for the moment) is the product of the unfolding ofcomplex networks of interacting genes, which trace back to the Big Origin. In biology, 1 plus 1 equals 7 or 10 or 19.

Combinatorial promiscuity, redundancy and modularity mean that organisms are enormously buffered on both the long timescale of evolution and the short timescale of development. Taking a regular sort of gene from one organism to another is not going to unleash a world of hybrid, uncontrollable genetically modified organisms meting out instant grief to all surrounding flora and fauna.

Without a theory of genetic interactions during development we cannot have a comprehensive theory of evolution, and so it is premature to put too much loose evolutionary spin on human behavioural variation.

I'm prepared to say, Charles, that it might turn out that we can no more have a predictive theory of interactions than we can of the weather or the economy.

Our tragedy is that the media pressure for quick responses to issues of importance - cloning, genetic modification, our human condition - encourages much uninformed reporting. You got us off to a good start, but your insistence on accuracy, comprehensive knowledge and caution have long left the agenda.

ÖйúAƬ

Alas, poor Darwin.

Yours as ever, Gabby Dover PS: earlier correspondence is published later this month as Dear Mr Darwin: Letters on the Evolution of Life and Human Nature (Weidenfeld and Nicolson). Additional critiques of EP can be found in Alas Poor Darwin, edited by H & S Rose, to be published in July.