Books: From stone age to space suit

For most of our history, humans were hunter-gatherers. Steven Pinker's stimulating book argues that this left us with a brain better adapted to solve the problems of the Stone Age than to meet the challenges of today

WHATEVER THEIR weaknesses, one thing Darwinists possess in abundance is confidence. Jared Diamond subtitled a recent book A Short History of Everybody for the Last 13,000 Years. Daniel Dennett has published a book entitled Consciousness Explained. And now Steven Pinker wants to tell us How the Mind Works.

Actually, Pinker concludes his book by suggesting that we can never know how the mind works - not completely anyway. If the mind is the product of natural selection, he asks, "why should we expect it to comprehend all mysteries and to grasp all truths?" Among the mysteries that he expects not to understand are consciousness, sentience and free will. But then, Pinker seems to believe that these are philosophical, not scientific concerns. And his aim is to sweep away philosophical speculation from psychology and provide a truly scientific account of the mind. Perhaps unsurprisingly, in this he fails. But he has succeeded in producing a witty, erudite, stimulating and provocative book that throws much new light on the machinery of the mind.

There are two key arguments here. First, Pinker wants us to think of the mind as a natural computer, which works by using a set of rules or algorithms to process data. By specifying these rules, and understanding how they are implemented, we can begin to learn how the mind works.

Second, Pinker argues that these rules have been selected for in the course of evolution. Each set of algorithms constitutes a "module" or separate organ within the brain, designed by natural selection to carry out a very specific task: to learn a language, for instance, to recognise faces or to behave romantically.

Pinker's attempt to stitch these two elements into a cohesive narrative is not entirely convincing. The first part of the book, on the psychology of cognition, is brilliantly argued and wondrously written. The second half, which examines the evolutionary antecedents of the mind, often descends into cod-psychology. (Anger, Pinker informs us, "protects a person whose niceness has left her vulnerable to being cheated"; guilt can "rack a cheater who is in danger of being found out".)

By far the most persuasive part of Pinker's book is his discussion of what has been dubbed the Computational Theory of Mind. At its heart is the idea that intelligent systems, like the human brain, cannot simply be stuffed with trillions of facts. They must be equipped with a smaller list of core truths and a set of algorithms to deduce their implications.

For example, what we call common sense embodies an immeasurable number of facts about the world that we take for granted. We know that when Edna goes to church her head goes with her. We know that if Doug is in the house he must have got in through an opening, unless he was born there and never left. Such notions cannot be specified, fact by fact, in the brain. They must derive from a tacit understanding of the workings of the world. The mind, Pinker writes, solves its unsolvable problems by "a leap of faith about how the world works".

Pinker demonstrates the power of the computational model in a brilliant exposition of human vision. The act of seeing, Pinker explains, is not simply a matter of recording sensory data from the eye, but of transforming that data into meaningful and coherent information. Pinker's description of the "Mind's Eye" is one of the most elegant pieces of writing I have come across about the mechanics of living processes.

How many sets of algorithms does the brain possess? It used to be thought the brain was a single general-purpose processor that applied the same set of rules to every situation. This is now regarded as highly improbable. As Pinker puts it, "only an angel could be a general problem solver". The minds of mere mortals must be stuffed with specialised problem-solving devices, each designed to carry out a specific task. Designed by who or what? By natural selection. In the same way as the eye or the kidney has evolved, so have the organs of the brain. And what tasks are they designed to accomplish? Since humans have lived for most of our history as hunter-gatherers, so, Pinker argues, evolution has designed our minds to solve the problems, not of the modern age, but of the Stone Age. The mind, Pinker writes, "is a system of organs of computation designed by natural selection to solve the kind of problems our ancestors faced in their foraging way of life".

Recent research has given credibility to the idea of a "modular" mind. Child psychologists, for instance, have shown that infants as young as three months have knowledge about what constitutes an object and about how objects move that is unlikely to have been learnt. Psychologists have dubbed this innate knowledge "intuitive physics". Similarly, many psychologists now believe that infants have an instinctive understanding of the difference between animate and inanimate objects ("intuitive biology"), and that other human beings have minds too ("intuitive psychology").

Meanwhile, studies of brain-damaged patients reveal their incapacities to be astonishingly specific. Some, for instance, are unable to name objects, but otherwise are perfectly normal. Others do not recognise faces but have no problem in recognising material objects. These studies suggest that such information is held in different domains ("modules") in the brain.

The idea of a modular mind is not new, having been first developed by an Austrian neuroanatomist, Franz Joseph Gall, in the early 19th century. By the postwar era, however, the modular picture of the brain had become defunct. At least in part this was because the idea of an innate human nature was seen as dangerous and racist. Now that modular notions are becoming fashionable again, many critics are once more warning of the dire social consequences.

Pinker rightly dismisses these fears. "A denial of human nature," he points out, "no less than an emphasis on it, can be warped to serve harmful ends." He adds that "the debate over human nature has been muddied by the intellectual laziness and unwillingness to make moral arguments when moral issues come up. Rather than reasoning from principles of rights and values, the tendency has been to buy an off-the-shelf moral package (generally New Left or Marxist) or to lobby for a feel-good picture of human nature that would spare us having to argue moral issues at all."

I have considerable sympathy with Pinker's complaint. The trouble is that evolutionary psychology tends towards an intellectual laziness as crass as that of its critics. Take, for instance, the claim that brain modules are analogous to body organs. This is simply not so. The heart is located in a specific place, its boundaries are well-drawn and it has an easily defined function. Not so the brain modules. The brain is made up of anatomically distinct regions, but these regions are not autonomous organs. Rather they are integrated in complex ways we do not yet understand. Psychologists talk of a "language module" but even the simplest linguistic task involves several different brain regions working simultaneously in an intricate, flexible fashion. The brain is functionally specialised but not in the way the body is. The analogy with body organs is misplaced.

Again, Pinker suggests that the brain's functional specialisation arises from innate mechanisms. But much recent evidence suggests that modularity may, at least in part, be the product of learnt rules and knowledge. Psychologists like Annette Karmiloff-Smith have proposed a much more nuanced relationship between innate and learned knowledge in the creation of modules, arguments that Pinker by and large ignores.

Moreover, Pinker continually conflates the concepts of "intuitive" and "folk" knowledge. All cultures, for instance, classify the living world into groups (land animals, birds, fish, and so on) which bear some resemblance to scientific classification, and most have a concept of "species". Pinker takes the existence of such folk knowledge as demonstrating an innate understanding of biology. But why should it? It simply suggests that humans have a common capacity to categorise and that the empirical reality of the world leads us, at some minimal level, to categorise living beings in a similar way.

The most problematic of Pinker's notions is the idea that modern behaviour is adapted to a Stone Age way of life. "Our brains," he writes, "are not wired to cope with anonymous crowds, schooling, written language, governments, police, courts, armies, modern medicine, formal social institutions, high technology, and other newcomers to the human experience." Why not? Our brains created all these things. It seems bizarre to hold that the brain is "wired up" to invent modernity but not cope with it.

The claim that we are Stone Age Men living in a Space Age world is based on a thoroughly unDarwinian methodology. Darwin wrote that "the present is the key to the past". Reversing the method and using the past as the key to explain the present is a fatal mistake. There is no reason why we should regard what once explained human behaviour in evolutionary terms as sufficient to explain human behaviour now.

Paradoxically for a book that claims to have solved the mind-brain problem, the consequence of Pinker's approach is that it unknowingly rehabilitates old-fashioned Cartesian dualism. Pinker is wary of committing what is called the "naturalistic fallacy" - the idea that because something is natural, it must be right. He therefore proposes that ethics should be separate from the scientific study of behaviour. Science and ethics, he argues, are "two self-contained systems played out among the same entities in the world". The "science game treats people as material objects, and its rules are the physical processes that cause behaviour through natural selection and neurophysiology". The "ethics game" on the other hand, "treats people as equivalent, sentient, rational, free-willed agents, and its rules are the calculus that assigns moral value to behaviour through the behaviour's inherent nature or consequence."

For Pinker, then, a human being is "simultaneously a machine and a sentient free agent". This allows an individual to behave in a thoroughly unDarwinian way and if their "genes don't like it, they can go jump in the lake".

But what allows humans to behave in this fashion? After all, natural selection would soon dispense with any tendency among non-human animals to tell their genes to go jump in the lake. Ethics, presumably, are not some metaphysical entities, but an aspect of human behaviour. How do they originate if not through "natural selection and neurophysiology" which Pinker holds to be the basis of all other behaviours? Pinker's division of human into "simultaneously a machine and a free agent" is a sleight of hand to avoid such questions. Descartes, unable to comprehend how science could explain the mind, divided the human into a mechanical body and unknowable soul. Pinker has done much the same - except that he has relabelled the soul as "ethics".

How the Mind Works is an important book. Its weaknesses arise from Pinker ignoring his own injunction about how one should understand the mind. At the start of the book Pinker notes that "any explanation of how the mind works that alludes hopefully to some single master force of mind- bestowing elixir like 'culture', 'learning' or 'self-organisation' begins to sound hollow, just not up to the demands of the pitiless universe we negotiate so successfully." It's a pity he did not add evolution to that list. There is more to the human mind than its evolutionary heritage.

8 'How the Mind Works' is published by Allen Lane, pounds 25.