Nature, nurture and a pinch of salt

Scientists at the Sanger Institute near Cambridge are expected within weeks to complete the mapping of all the genes on human chromosome 22. Within the next three years, they and other scientists collaborating on the Human Genome Project will have elucidated the DNA sequences of all the genes on all 23 pairs of chromosomes found in every human cell. It will be a staggering achievement. But the excitement about what is being done should be greatly moderated. "The Book of Life", as one leading scientist called it, will not provide the complete story about human nature.

Scientists at the Sanger Institute near Cambridge are expected within weeks to complete the mapping of all the genes on human chromosome 22. Within the next three years, they and other scientists collaborating on the Human Genome Project will have elucidated the DNA sequences of all the genes on all 23 pairs of chromosomes found in every human cell. It will be a staggering achievement. But the excitement about what is being done should be greatly moderated. "The Book of Life", as one leading scientist called it, will not provide the complete story about human nature.

The human genome is like a cook's larder list. Working out all the dishes which cooks might make from the ingredients available to them is another matter. If you want to understand what happens in the life-long process from conception to death, you must study the process. The starting points of development include the genes. But they also include factors external to the genome and, of course, the social and physical conditions in which the individual grows up are crucial. You would not realise this from the language employed by some scientists and the media today.

Hardly a week goes by without a new and supposedly direct link between genes and individual human characteristics being reported. Intelligence, criminality, a desire for novelty, homosexuality - all are said to have their gene. Recent examples include "Happiness seems to be largely determined by the genes", "Scientists find dyslexia gene", "Excited researchers think they have found a gene for language", and "It's the gene that explains feminine intuition".

The language of a gene "for" a particular characteristic is exceedingly muddling to the non-scientist - and, if the truth be told, to many scientists as well. What scientists mean (or should mean) is that a genetic difference between two groups is associated with a difference in a characteristic. They know that other things are important and that, even in constant environmental conditions, the outcome depends on a combination of many genes.

Unfortunately, the language of genes "for" characteristics has a way of seducing the scientists themselves into believing their own sound-bites. The language rests on a profound misunderstanding, which we have referred to in our book Design For A Life as "The Great Blueprint Fallacy".

While genes obviously matter, even a cursory glance at humanity reveals the enormous importance of each person's experience, upbringing, education and culture. The notion that genes are simply blueprints for an individual human is hopelessly misleading. In a blueprint, the mapping works both ways: starting from a finished house, the room can be found on the blueprint, just as the room's position is determined by the blueprint. This straightforward mapping is not true for genes and individual human behaviour patterns, in either direction.

Genes do not make behaviour patterns or physical attributes. Genes make proteins. Each human has about 60,000 genes, each of which is an inherited molecular strand (or set of strands) which may be translated into a protein molecule (or part of one). The proteins are crucial collectively to the functioning of each cell in the body. Some proteins are enzymes, controlling biochemical reactions, while others form the physical structures of the cell. These protein products of genes do not work in isolation, but in a cellular environment created by the conditions of the local environment and by the expression of other genes. Each gene product interacts with many other gene products.

The adult human body is made up of a hundred million million cells. Although the complete set of genes is present in virtually all cells, different genes are activated or inactivated at different stages in development. Genes are switched on and off in response to local conditions within the cell. As the cells of a fertilised egg divide to form the early embryo, the activities of adjoining cells - and, increasingly, the conditions in the environment surrounding the embryo - participate in what looks like a symphony of gene expression.

At least half of all human genes are involved in building the brain and nervous system, although many of these same genes are also involved in building other parts of the body. The adult human brain has around one hundred thousand million neurons, or nerve cells, each with hundreds of connections to other neurons. A diagram of even a tiny part of the brain's connections would look like an enormously complex version of a map of the London Underground. Since the behaviour of the whole animal is dependent on the whole brain, it will be obvious why it is not sensible to ascribe a single aspect of behaviour to a single neuron, let alone a single gene.

Each characteristic of an individual, such as a behaviour pattern or psychological attribute, is affected by many different genes, each of which contributes to the variation between individuals. In an analogous way, many different design features of a motor car contribute to a particular characteristic such as its maximum speed. A particular component such as the system for delivering fuel to the cylinders may affect many different aspects of the car's performance, such as its top speed, acceleration and fuel consumption. A broken wire can cause a car to break down, but this does not mean that the wire by itself is responsible for making the car move.

The image of a genetic blueprint also fails because it is too static, too suggestive that adult organisms are merely expanded versions of the fertilised egg. In reality, developing organisms are dynamic systems that play an active role in their own development. To some extent each individual chooses and shapes its own physical and social environment. This can have interesting consequences. People who differ in ways that relate to differences in their genes may also pick certain physical and social environments in which to live. It means individuals with different characteristics, some of which reflect differences in their genes, end up by their own actions experiencing the world in different ways.

Environmental and inherited factors often work together to produce much larger overall effects than when either factor is present on its own. One study in Sweden looked at children from families with or without a history of criminality. These children had been adopted early in life into families, some of which also had a history of criminality. Coming from a biological family with a criminal history quadrupled the individual's chances of exhibiting criminality, while being adopted from a non-criminal biological family into a criminal family doubled the risk. But the combination of criminality in both the biological and adopting families increased the risk of developing into a criminal by a factor of 14 - far more than just the product of the biological and familial influences.

The often uncanny similarities between identical twins provide striking evidence for the importance of genes in shaping physical and behavioural characteristics. But one surprising finding to emerge from studies of identical twins is that twins reared apart are sometimes more like each other than those reared together. So rearing two genetically identical individuals in the same environment can make them less similar. This fact pleases neither the extreme environmental determinist nor the extreme genetic determinist. The environmental determinist supposes that twins reared apart must have different experiences and should therefore be more dissimilar in their behaviour than twins who grew up together. The genetic determinist does not expect to find any behavioural differences between genetically identical twins reared together; if they have had the same genes and the same environment, how can they be different? Of course, one twin provides an environment for the other and siblings often hate to do what the other one is doing.

A single developmental ingredient, such as a gene or a particular form of experience, might produce an effect on behaviour, but this knowledge gives only a feeble insight into the processes that we have called developmental cooking. The best that can be said of the nature/nurture split is that it provides a framework for uncovering a few of the genetic and environmental ingredients that generate differences between people. At worst, it satisfies a demand for simplicity in ways that are fundamentally misleading.

Patrick Bateson is Provost of King's College Cambridge. 'Design For A Life: How Behaviour Develops' by Patrick Bateson and Paul Martin was recently published by Jonathan Cape