Jacket of The Atheist's PrimerThe Atheist's Primer by Michael Palmer (Lutterworth Press)

Atheists. What are they like? Wading in to decent god-fearing folk and bludgeoning them repeatedly in the faith with a sock full of a posteriori scientific “facts” until their feet bleed. It’s too easy just to flick open Occam’s razor like a theological Sweeney Todd and declare that we don’t need God, not now that we have natural selection of random mutations, the human genome, the Big Bang and the Higgs Boson. Anyone who has tried to grasp the entirety of, say, the Standard Model of particle physics will feel that Occam’s razor would prefer God to bosons. But it’s current fashion to beat theists with the science shtick. On and on and on. Time for another tune.

Michael Palmer’s got a whole compendium of them to offer, with lyrics to match. In his concise, comprehensive and accessible grand tour d’horizon of atheistical arguments, he surveys the philosophical approach that started 2,500 years ago, in the extraordinary explosion of critical thought in 5th-century BC Athens. One moment we had Hesiod’s gloomy Theogony, a family tree of the gods with all the charm of a rural curmudgeon threatening incomers with his knotty cudgel; the next, Socrates, the first atheos, who said he couldn't find out anything about the gods and anyway life was too short to waste on such stuff.

Game on.

From the late 5th century BC, reasoning and philosophical investigation were the way to oppose the gods (or, later, God), and anyone who wants to be a good atheist today should read The Atheist’s Primer as a powerful toolkit to add to, and in many cases supplant, the mere recitation of scientific evidence for a god-free universe, not least because neither approach necessarily works on its own. If you believe that, culturally, God is what’s left after the stuff we know, it’s true that the space he occupies is shrinking rapidly; but only in terms of the boundaries we impose on that space. If God is the set of all null sets, then God is not necessarily a nullity himself.

Take, for example, the Paradox of the Stone. It’s a simple question: Could God create a stone too heavy for him to lift?

The usual answer, cited by Palmer, is twofold. (1) If he can, he’s not omnipotent, or (2) if he can’t, then he’s not omnipotent either. In what terms, though? The Paradox of the Stone occupies an entirely physical realm; but an omnipotent God must a priori be able to transcend the laws of material logic. The embattled theist could reasonably counter by saying, “God can deploy his omnipotence by existing perfectly well within an apparent paradox.” If mathematically inclined, he might cite Cantor, who showed that one infinity might be greater than another though both are still infinite. If cosmologically inclined, he could adduce the “multiverse” in which all possible universes exist. And so the atheist might counter that, if it’s paradox you want, that would include one in which there is no multiverse at all.

Deuce.

It always is.

Palmer, whose command of the concise précis is remarkable, leads us through the argument but leaves us to shout “Bollocks!” and then shamefacedly realise we've betrayed our own atheism. But if you survey atheism’s philosophical horizon in one grand panorama, as here, shouting “Bollocks!” is inevitable. It’s chastening to see what a cock-up we've made of it – or, perhaps, what a tremendously intractable argument it is. The old technologist’s maxim declares: “You can’t make anything foolproof because the fool is always one step ahead.” God, or the idea of God, appears here as just such a fool.

The atheist arguments are as shaky as the “leap of faith” arguments for God, or Stephen Jay Gould’s bizarre “dual magisteria”. We see endless collision of the argument with either the elided or the unknown unknowns, whether it’s the immateriality of God (as favoured by Thomas Aquinas), the counter that since God didn't have a material body he was incomplete and thus imperfect, or, indeed, Pascal’s wager: behave as though you believed in God even though you don’t. That one requires God to be stupid or hoodwinked for the bet to pay off. But doesn’t the Paradox of the Stone tell us that a truly omnipotent God could allow himself to be both stupid and hoodwinked?

And all too frequently, the roll-call of great philosophers is interrupted by the crash and clang of colliding epistemologies as a priori reasoning is applied to a posteriori questions and vice versa.

Two things are clear: very often people rail against God when their true target is religion; and the matter – whether evil, omnipotence, miracles or anything else – is not cleanly susceptible to reason. Occam’s razor drags, scrapes and draws blood.

Most persuasive is a simple, benign weariness. My favourite of all the Romans, Pliny the Elder, who cheerfully points out that God can’t be omnipotent because he can’t stop twice ten being twenty, can’t rewrite history, and can’t commit suicide, “that supreme boon among all the penalties of life”. These, he writes, “unquestionably demonstrate the power of nature, and prove that it is this we mean by the word ‘God’.”

Michael Palmer has written not only an elegant summary of 2,500 years of thinking-about-God, but a genuine “Primer”, locked and loaded for argument, which doubles as a reminder of our intellectual limitations. The arguments it offers are less overweening than the barrage of scientific facts we tend to rely on. But in the end, for me, it’s Pliny who disarms. The man who wanted to know everything ends his monumental Historia Naturalis by breaking off in mid-sentence of a discourse upon precious metals to cry Salue parens omnia rerum Natura: “Hail, mother of all things, Nature”. That, at least, is something all atheists can surely agree on; and, if they think carefully, theists too.