Raymond Tallis's books are not often easy to summarise, and not always easy to finish. He worked for years as a doctor, with a scientific speciality in the treatment of epilepsy and stroke victims, and wrote for his own enlightenment. Although he is capable of writing with great clarity and force about really important things, there is a sense that he is conducting an argument with the people he has read, rather than the people who might be reading him. Nonetheless, it's clear that the purpose of this book is to rescue atheism from the currently fashionable atheists.

He attacks, among others, the philosopher John Gray for his anti-scientific pessimism and the Daniel Dennett / Richard Dawkins axis for their pseudo-scientific optimism. All seem to him to mis-state, to elide, or to conceal the absolute strangeness of being human. But their viewpoint is so much taken for granted that it is easy to mistake it for the way the world is, rather than a worldview built like any other on metaphor, assumption and a bright dusting of wish fulfilment.

Broadly, these authors agree that we are animals directed by computers. We might not like to be such things, but it is what science shows we are. The only alternative to this would be some viewpoint dismissible as "religious". Tallis doesn't like it, either; but he also believes it a completely wrong picture, scientifically flawed as well as philosophical nonsense. At the same time, he remains "a proud atheist".

Aping Mankind is a double-barrelled blast at both the computer and the animal understandings of human nature and in particular of human consciousness. The arguments against consciousness as computation have been already made in his short book Why The Mind Is Not a Computer. As a gerontologist, he has been constantly brought up against the relationship between people and their variously damaged or decaying brains. That the one depends on the other does not mean that they are the same thing, or even the same kind of thing. The brain is something that appears to consciousness. Consciousness is not something that appears in the brain.

One of Tallis's central points is the discussion of "information". This word plays a central role in the Dawkins/Dennett world view, much more important and less obvious than the nonsense about "memes". Brains, computers, and even life itself, are all said to be processing information. DNA itself is pure digital information. But the word here needs scare quotes throughout, for it has two quite different and separate senses. The older usage of the term is inextricably bound up with meaning: information is something you know that carries a meaning. It is, in engineer's jargon, signal, rather than noise. Information, in this sense, is always information to someone or some system.

But there is a second sense of "information", arising from electrical engineering, and the beginnings of computer science, in which it is entirely measurable, and can be broken into discrete chunks. This has been an important and productive understanding  I couldn't be typing and you couldn't be reading without that kind of information science  but it came at the price of breaking "information" entirely away from meaning. Tallis quotes one of the pioneers in the field: "Information, in this theory is used in a special sense that must not be confused with its ordinary usage. In particular, information must not be confused with meaning. In fact, two messages, one of which is heavily loaded with meaning, and the other of which is pure nonsense, can be exactly equivalent, from the present viewpoint, as regards information."

Despite this admirable clarity pop science writers and artisans of metaphor have been trying for years to glue the two senses of "information" together again, as if what can be measured in computers is the same stuff as meaning for humans. "By smuggling consciousness into the matter of the brain, via the computer analogy, we make a materialist account of consciousness seen plausible ... When you personify the brain and bits of brain, this makes it easy to 'brainify' the person."

Once these two meanings of "information" have been unpacked and separated, so we can distinguish between the mathematically measurable properties of a fluctuating electric current, and the stuff that makes up the world we experience, then it's possible to unpick the idea that consciousness is brain activity, or that consciousness is just the way we experience brain activity. As Tallis points out, the brain activity we can measure is present to our consciousness. It's not present to itself. The fact that you can destroy selves by destroying bits of the brain does not prove they are identical. "A person's behaviour becomes more completely explicable in neurological terms the more damaged they are. A seizure sits more comfortably within the neural model of the mind than does living with epilepsy, which requires something to bring it all together."

This kind of personhood  the capacity, in fact the compulsion, to bring things together into some kind of coherent narrative, without which experience is not just senseless, but almost impossible, is what he believes science cannot now explain. Anyone tempted to suppose that science has explained it even in principle  and that means almost all of us  should read him, and realise we're wrong.

Andrew Brown's Fishing in Utopia is published by Granta.

Chimpanzees with an extra gene or two  that's all we are. The sense of having agency is really an illusion, as is the sense of self. All "love" actually comes to is neurones firing in the caudate nucleus. Why Donne moves us is just because certain linguistic tropes stimulate certain neuronal pathways. According to claims made by supposedly respectable neuroscientists, reported by supposedly respectable newspapers, it is possible actually to pinpoint the locus of love, wisdom or whatever as parts of the brain that "light up" under functional magnetic resonance imaging (brain scans).

All nonsense, according to Raymond Tallis, professor of geriatric medicine at the University of Manchester, poet, novelist, and all-round polymath. When neuroscience and Darwinism trespass into the humanities, they become, he says, "neuromania" and "Darwinitis"  unhealthy, mad and malign.

He himself uses fMRI on his patients, so is well equipped to explain that what the scanning actually shows is widespread brain activity, which is marginally more prevalent in some areas than others. And usually the same areas  areas that incessantly crop up in different experiments, and are cited as serving quite different functions ("You could be forgiven for thinking of the brain as being managed by a crooked estate agent letting out the same bit of real estate simultaneously to different clients").

Anyway, argues Tallis, even if the scans could in theory, and did in practice, reliably isolate specific brain states to match up with specific psychological states, what would such matching amount to? Is the brain state supposed to coincide with, cause, or be the very same as the psychological state? Sometimes neuromaniacs, in cunning or confusion, shift between these alternatives; sometimes they brazenly opt for a third  the idea that nerve impulses, "by moving from one material place to another are mysteriously able to be the appearance of things other than themselves".

But if a psychological state (say, an experience of seeing yellow) and a brain state "were the same thing, the least one might expect is that they would appear as if they were the same thing"  appear to be yellow, for instance. They don't; not, anyway, to the neuroscientist observing the brain. But in principle the person whose brain is being observed could (via mirrors and advanced technology) observe their own brain while observing a daffodil outside it; and then their electrochemical activity would simultaneously provide an experience of grey sparking brain stuff and of a yellow flower. Electrochemical activity in the brain, in fact, is required to have "two sets of appearances", but its "inside" appearance is only available to the brain owner. "Inside" is surely illegitimate for neuromaniacs, as is "brain-owner", since they deny a unified self, yet in fact they do precisely what they accuse believers in immaterial mentality of doing. They smuggle a miniature inner observer (a homunculus) into the head. It is just that their homunculus is better disguised; it is broken up into a whole lot of tiny homunculi  neurons, circuits or bits of brain.

Like the clever child who shouts "it's in his pocket" at the bad conjuror, Tallis brilliantly exposes the portentous fraudulence of memes, and the way cod-sci metaphors, such as "information", manage, in anthropomorphising machines, to mechanise humans. Less convincingly, he defends free will from the Libet experiments which purport to disprove it. His goal is to "reaffirm humanity"  and without appealing to mysterious "mind-stuff". What, then, is his alternative to the scientistic account? Each of us, he says, is lifted from the organic events of our self-contained bodies and "stand-alone brains" by "a trillion cognitive handshakes"  the way humans have jointly built up a world of transmittable meanings which transcend biology.

But how did that come about? Tallis makes various explanatory gestures towards the evolution of the hand as flexible tool, but these seem incongruously biologistic. And what about subjective consciousness, the perspectival view which both unites and differentiates a human's multiple experiences? He does not pretend to explain it, but we can be grateful that he shows up the pretensions of those who think they can, and who reduce the glory of being human to brain circuitry and the survival tactics of early hominids. With erudition, wit and rigour, Tallis reveals that much of our current wisdom is as silly as bumps-on-the-head phrenology.