‘Science is boring,’ proclaims Michael Strevens airily, setting a maverick tone. Then he tells a story about rival endocrinologists Roger Guillemin and Andrew Schally, who mashed up tonnes of animal brains in their mission to isolate a milligram of the thyroid-stimulating hormone TRH and to unravel its molecular structure. Like the Curies, they won a Nobel Prize for their labours, although Strevens emphasises the labour rather than the glory, because his response to why scientists have such superhuman tenacity is the scientific method’s demand for empirical evidence. But there is an ambitious project behind this deceptively simple answer: Strevens thinks this demand for evidence is so remarkable, so ‘unreasonable’, that it merits a new thesis to explain why modern science works so brilliantly, and why it took so long to emerge.

A philosopher rather than a scientist, Strevens takes a long and winding road to his destination. Scientists will likely get there much faster, for they have the advantage of hindsight. It takes someone with a philosophical bent to unravel that hindsight – that habitual way of thinking that for most of human history wasn’t obvious at all – and to ask deeper questions. Yet The Knowledge Machine is accessible to non-specialists, offering a cornucopia of interesting snippets – historical, biographical, philosophical, and scientific – within an overarching quest to understand what makes science such a powerful method of enquiry. In fact, says Strevens, it’s more than a method: it’s a ‘knowledge machine’ with the unparalleled ability to generate new knowledge.

To demonstrate just how ‘unreasonable’ the scientific method is – for in his view it takes all the humanness out of it, all the philosophising, theologising, and artistry that people naturally bring to pondering the nature of the world – Strevens begins by showing that not even the famous philosophers Karl Popper and Thomas Kuhn captured its secret. He often makes his arguments through counter examples, such as the 1919 eclipse expedition led by Arthur Eddington and designed to test Einstein’s prediction that gravity bends light. Strevens wants to show that Popper’s concept of ‘falsification’ – that empirical evidence stringently weeds out false theories – is compromised by scientists’ subjective interpretation of that evidence, and he accuses Eddington of selecting the eclipse data to favour his own personal and intellectual agenda. (Unfortunately, Strevens’s principal source for this unfair accusation is a 1980 book that scientists have repeatedly challenged.)

Strevens does ultimately build a terrific account of how the scientific method melds scientists’ natural subjectivity with the objectivity demanded of their published research. But it’s a pity he takes such an adversarial approach, not only to Eddington but to science as a whole. True, he is deliberately provocative: often, it’s clever, as when he builds cases only to undercut them to make his points; but sometimes it feels too clever, undercutting so much that it’s difficult to know what he’s on about. Strevens’s quest assumes that science is extraordinary, and he makes a significant contribution by establishing that those who argue that science is inherently subjective confuse the process of doing science with the rules ensuring that evidence is found and written up objectively. But he doesn’t seem to like scientists very much or to understand the joy that many derive from doing science. He cherry-picks quotes, such as E.O. Wilson’s ‘So many scientists are narrow, foolish people’, and thinks science is mostly inhumanly dull work that succeeds because its explanations are ‘shallow’ and its modern practitioners rarely ‘think outside the box’: their emphasis on empiricism ‘starves’ them of all other meaning, ‘clamping’ their minds, ‘sucking the air out’ of them like a ‘merciless’ experimenter proving that a bird dies without oxygen. Some of this antagonism is rightly directed at the elevation of STEM over humanities; he also strikes a suitably ambivalent note about science’s role in today’s climate-changed, hyper-technological world. Still, I wonder how many scientists Strevens has actually met. Although he is writing a popular book, I would have preferred less hyperbole, more nuance and context.

That aside, there is a lot that is both fascinating and informative: the philosophy, the debates over the age of the earth and the nature of heat, how quantum theory can be so inexplicable and yet so practical, and much more. He sets up ingenious thought experiments, inviting readers to imagine how they would decide which of two rival theories to accept, or how they might determine the best approach to understanding nature. In this way, and by discussing thinkers from Aristotle to Descartes to Murray Gell-Mann, he gradually builds up his thesis. It is essentially Isaac Newton’s thesis, as Strevens eventually acknowledges, but he aims to generalise it, teasing out the hidden assumptions in various experiments, taking selections from the history and philosophy of science in order to divine its enigmatic power, and exploring why it arose when and where it did.

Philosophers like to challenge our thinking, and Strevens certainly does that. But there’s much else to enjoy in this unique intellectual mystery tour.