2

The Idea of the Scientific Revolution

With all its imperfections, modern science is a technique that is sufficiently well tuned to nature so that it works – it is a practice that allows us to learn reliable things about the world. In this sense it is a technique that was waiting for people to discover it.

– Steven Weinberg, To Explain the World (2015)¹

§ 1

When Herbert Butterfield lectured on the Scientific Revolution at the University of Cambridge in 1948 it was the second year in which an historian at the university had given a series of lectures on the history of science: he had been preceded the year before by the Regius Professor of History, G. N. Clark, an expert on all things seventeenth century, and the medieval historian M. M. Postan had lectured immediately before Butterfield. It was in Cambridge that Isaac Newton (1643–1727) had written his Philosophiæ naturalis principia mathematica, or Mathematical Principles of Natural Philosophy (1687), and here that Ernest Rutherford (1871–1937) had split the atomic nucleus for the first time, in 1932. Here, the historians were acknowledging, they were under a particular obligation to study the history of science. They were also keen to insist that the history of science be done by historians, not by scientists.^i 2

The historians and the scientists at Cambridge shared a common education: Latin was a compulsory entrance requirement. They met over lunch and dinner in their colleges, but they lived in separate mental worlds. Butterfield began the book based on his lectures, The Origins of Modern Science(1949), by expressing the hope that the history of science might serve as a long-needed bridge between the arts and the sciences. He hoped in vain. In 1959 (the year in which Latin was finally dropped as an entrance requirement) C. P. Snow, a Cambridge chemist and a successful novelist, delivered a lecture complaining that Cambridge dons from the sciences and the arts had now more or less stopped speaking to each other.ⁱⁱ It was entitled ‘The Two Cultures and the Scientific Revolution’ – the revolution being Rutherford’s revolution, which had led to the creation of the atomic bomb.³

In adopting the term ‘the Scientific Revolution’ a decade before Snow, Butterfield was (it is always said) following the example of Alexandre Koyré (1892–1964).⁴ Publishing in French in 1935, Koyré (a German-educated Russian Jew who had been imprisoned in Tsarist Russia at the age of fifteen for revolutionary activity, had fought for France in the First World War, would join the Free French in the Second, and would later become a leading figure in American history of science) distinguished the Scientific Revolution of the seventeenth century, which ran from Galileo to Newton, from ‘the revolution of the last ten years’; Heisenberg’s classic paper on quantum mechanics had been published exactly ten years before.ⁱⁱⁱ For Koyré and Butterfield it was physics, the physics first of Newton and then of Albert Einstein (1879–1955), which symbolized modern science. Now we might give equal prominence to biology, but they were writing before the discovery of the structure of DNA by James Watson and Francis Crick in 1953. As Butterfield was giving his lectures the medical revolution represented by the first modern wonder drug, penicillin, was only just getting under way, and even in 1959 C. P. Snow still thought the important new science was being done by physicists, not biologists.

So at first there was not one Scientific Revolution but two, one exemplified by Newton’s classical physics, the other by Rutherford’s nuclear physics. It was only very slowly that the first won out over the second in the claim to the definite article.⁵ The idea that there is such a thing as ‘the Scientific Revolution’ and that it took place in the seventeenth century is thus a fairly recent one. As far as historians of science are concerned, it was Butterfield who popularized the term, which occurs over and over again in the course of The Origins of Modern Science; but the first time he introduces it he refers to it, awkwardly, as ‘the so-called “Scientific Revolution”, popularly associated with the sixteenth and seventeenth centuries.’ ‘So-called’ is apologetic; even stranger is his insistence that the term is already in popular use.⁶ Where did Butterfield find the term, other than in Koyré (whose work would have been totally unknown to his audience), used specifically about the sixteenth and seventeenth centuries? The phrase ‘the scientific revolution of the seventeenth century’ originates, it seems, with the American philosopher and educational reformer John Dewey, the founder of pragmatism, in 1915,^iv but it is unlikely that Butterfield ever read Dewey. Butterfield’s source, surely, is Harold J. Laski’s The Rise of European Liberalism (1936), an immensely successful book which had just been reissued in 1947.⁷ Laski was a prominent politician and the leading socialist intellectual of the day; he was enough of a Marxist to have a taste for the word ‘revolution’. It was his usage, then, not Koyré’s, that Butterfield adopted with some discomfort, believing that it would already be familiar to many of his listeners and readers.

Thus in this respect the Scientific Revolution is not like the American or French revolutions, which were called revolutions as they happened; it is a construction of intellectuals looking back from the twentieth century. The term is modelled on the term ‘Industrial Revolution’, which was already commonplace towards the end of the nineteenth century (and which originates, it seems, in 1848, with Horace Greeley – now famous for supposedly saying, ‘Go West, young man!’),⁸ but which is also an after-the-fact construction.^v And this of course means that some will always want to claim that we would be better off without such constructions – although it is worth remembering that historians constantly (and often unthinkingly) use them: ‘medieval’, for example, or the Thirty Years War (terms that could, necessarily, be introduced only after the fact); or, for any period before the Renaissance, ‘the state’, or, for any before the mid-eighteenth century, ‘class’ in the sense of social class.

Like the term ‘Industrial Revolution’, the idea of a scientific revolution brings with it problems of multiplication (how many scientific revolutions?) and periodization (Butterfield took as his period 1300 to 1800, so that he could discuss both the origins and consequences of the revolution of the seventeenth century). As time has gone on the idea that there is something that can sensibly be called the Scientific Revolution has come increasingly under attack. Some have argued for continuity – that modern science derives from medieval science, or indeed from Aristotle.^vi Others, beginning with Thomas Kuhn, who published a book on The Copernican Revolution in 1957, followed by The Structure of Scientific Revolutions (1962), have sought to multiply revolutions: the Darwinian revolution, the Quantum revolution, the DNA revolution, and so on.⁹ Others have claimed that the real Scientific Revolution came in the nineteenth century, in the marriage of science and technology.¹⁰ All these different revolutions have their utility in understanding the past, but they should not distract our attention from the main event: the invention of science.

It should be apparent that the word ‘revolution’ is being used in very different senses in some of the examples above, and it helps to distinguish three of them, exemplified by the French Revolution, the Industrial Revolution and the Copernican revolution. The French Revolution had a beginning and an ending; it was an enormous upheaval which, in one way or another, affected everyone alive in France at the time; when it began, nobody foresaw how it would end. The Industrial Revolution is rather different: it is rather difficult to say when it began and when it ended (conventionally, it runs from about 1760 to sometime between 1820 and 1840), and it affected some places and some people much more rapidly and extensively than it affected others, but everyone would agree that it began in England and depended on the steam engine and the factory system. Finally, the Copernican revolution is a conceptual mutation or transformation, which made the sun, not the earth, the centre of the universe, and placed the Earth in movement around the sun instead of the sun around the earth. For the first hundred years after the publication of Copernicus’s book On the Revolutions of the Heavenly Spheres in 1543 only a fairly limited number of specialists were familiar with the details of his arguments, which were only generally accepted in the second half of the seventeenth century.

A failure to distinguish these senses, and to ask which of them the first users of the term ‘the Scientific Revolution’ had in mind, has caused a tremendous amount of confusion. The source of this confusion is simple: from its first appearances, the term ‘the Scientific Revolution’ was being used in two quite different ways. For Dewey, Laski and Butterfield, the Scientific Revolution was a lengthy, complex, transformative process, to be compared with the Reformation (which Laski called a theological revolution) or the Industrial Revolution. For Koyré, it was, following on Gaston Bachelard’s concept of an ‘epistemological break’, identified with a single intellectual mutation: the replacement of the Aristotelian idea of place (in which there was always an up and a down, a left and a right) by a geometrical idea of space, a substitution which made possible, he argued, the invention of the idea of inertia, which was the foundation of modern physics.¹¹ Koyré had a vast influence in America, and his Bachelardian conception of an intellectual mutation was adopted by Thomas Kuhn in The Structure of Scientific Revolutions. Laski and Butterfield had a comparable influence in England on works such as Rupert Hall’s The Scientific Revolution (1954), which denied any connection between the Scientific and Industrial revolutions, and J. D. Bernal’s Science in History, whose second volume, The Scientific and Industrial Revolutions (1965), insisted on the closeness of the connection.

There is a fundamental difference between these two conceptions of the Scientific Revolution. Copernicus, Galileo, Newton, Darwin, Heisenberg and others who have been responsible for particular intellectual reconfigurations, mutations or transformations in science had a very good grasp of what they were doing as they did it. They knew that if their ideas were adopted the consequences would be momentous. It is thus easy to think of scientific revolutions as deliberate acts, conducted by people who achieve what they set out to achieve. Butterfield’s Scientific Revolution was not that sort of revolution. Comparisons between the Scientific Revolution and political revolutions are not entirely misleading, for they were both transformative of the lives of all they touched; they both had identifiable beginnings and ends; they both involved struggles for influence and status (in the Scientific Revolution between the Aristotelian philosophers and the mathematicians who favoured the new science). Above all, both political revolutions and the Scientific Revolution have had unintended, not intended, outcomes. Marat aspired to liberty; the outcome was Napoleon. Lenin, when he published State and Revolution just two months before the October 1917 revolution, genuinely believed that a communist revolution would lead to the rapid withering away of the state. Even in the American Revolution, which came closest to realizing the ideals which first inspired it, there is a vast gap between Thomas Paine’s Common Sense (1776), which envisaged a democratic system in which a majority could do more or less whatever they chose, and the complex checks and balances of the American Constitution as analysed in The Federalist(1788), which were designed to keep radicals like Paine trussed and tied. In the Scientific Revolution, Bacon and Descartes were amongst those with plans for thoroughgoing intellectual change, but their plans were castles in the air, and neither of them imagined what Newton would achieve. The fact that the outcome of the Scientific Revolution as a whole was not foreseen or sought by any of the participants does not make it any the less a revolution – but it does mean it was not a neat epistemological break of the sort described by Koyré.^vii So, too, when first Thomas Newcomen (1711) and then James Watt (1769) invented powerful new steam engines, neither foresaw that the age of steam would see the construction of a great railway system girdling the Earth – the first public steam railway did not open until 1825. It is this sort of revolution, a revolution of unintended consequences and unforeseen outcomes, that Butterfield intended to evoke by the term ‘the Scientific Revolution’.

If we define the term ‘revolution’ narrowly as an abrupt transformation that affects everybody at the same time, there is no Scientific Revolution – and no Neolithic Revolution, or Military Revolution (following the invention of gunpowder), or Industrial Revolution (following the invention of the steam engine) either. But we need to acknowledge the existence of extended, patchy revolutions if we want to turn aside from politics and understand large-scale economic, social, intellectual and technological change. Who, for example, would object to the term ‘the digital revolution’ on the grounds that it is not a singular and discrete event, localized in time and space?

There is a certain irony in Butterfield’s adoption of the retrospective term ‘Scientific Revolution’, and an even greater one in his choice of The Origins of Modern Science for his title. In 1931 he had published The Whig Interpretation of History, which attacked historians who wrote as if English history led naturally and inevitably to the triumph of liberal values.¹³ Historians, Butterfield argued, must learn to see the past as if the future were unknown, as it was to people at the time. They must think their way into a world in which the values we now hold, the institutions we now admire, were not even imagined, let alone approved. It was not the historians’ job to praise those people in the past whose values and opinions they agreed with and criticize those with whom they disagreed; only God had the right to sit in judgement.^viii Butterfield’s attack on the liberal tradition of historical writing in England was salutary, although he soon grasped that the sort of history he was advocating would be unable to make sense of the past, since without hindsight it would be impossible to establish the significance of events; history would become like the Battle of Borodino as experienced by its participants – at least according to Tolstoy in War and Peace – and both the readers and the historians themselves would stumble about, unable to make sense of events. Tolstoy, of course, as an omniscient narrator, also provides a running commentary, establishing what it was that the combatants were all, willy-nilly, conspiring to bring about. But later historians have naturally turned the phrase ‘Whig history’ back against Butterfield himself, accusing him of taking for granted the superiority of modern science over all that went before. The very idea of a book about ‘origins’ seems to them contrary to the principles he established in The Whig Interpretation of History.^ix 14 Indeed it is; but the fault lies with Butterfield’s early principles, not his later practice, for we really do need to understand the origins of modern science if we are to understand our own world.

§ 2

For the most part, scholars in recent years have been reluctant to adopt the term ‘the Scientific Revolution’, and many have explicitly rejected it. The opening sentence of Steven Shapin’s The Scientific Revolution (1996) is often quoted: ‘There is no such thing as the Scientific Revolution,’ he wrote, ‘and this is a book about it.’¹⁵ The main source of their discomfort (once one has cleared away confusions over the meaning of the word ‘revolution’) points to a feature of the study of history that Butterfield simply took for granted and saw no need to discuss: that language is ‘the principal working tool’ of the historian.¹⁶ The whole of Butterfield’s Whig Interpretation of History is a critique of anachronistic thinking in history, but Butterfield never discusses a fundamental source of anachronism: the language in which we write about the past is not the language of the people we are writing about.^x When Butterfield’s arguments were restated by Adrian Wilson and T. G. Ashplant in 1988, the central feature of the historian’s enterprise had become the fact that texts which survive from the past are written in what amounts to a foreign language.^xi Suddenly it seemed that there was a hitherto unacknowledged problem with the word ‘revolution’ and indeed with the word ‘science’, too, for these are our words, not theirs.^xii

The word ‘science’ comes from the Latin scientia, which means ‘knowledge’. One view to take, a view that derives both from Butterfield’s rejection of Whig history and from Wittgenstein (to whom we will turn later in this chapter), is that truth or knowledge is what people think it is.^xiiiOn this view astrology was once a science, and so of course was theology. In medieval universities the core curriculum consisted of the seven liberal ‘arts’ and ‘sciences’: grammar, rhetoric and logic; mathematics, geometry, music and astronomy (including astrology).¹⁷ They are often now referred to as the seven liberal arts, but each one was originally called both an art (a practical skill) and a science (a theoretical system); astrology, for example, was the applied skill, astronomy the theoretical system.^xiv These arts and sciences provided students with the foundations for the later study of philosophy and theology, or of medicine or law. These, too, were called sciences – but philosophy and theology were purely conceptual explorations that lacked an accompanying applied skill. They had practical implications and applications, of course – theology was applied in the art of preaching; and both ethics and politics, as studied by philosophers, had practical implications – but there were no university courses in applied theology or philosophy. They were not arts, and it would have been incomprehensible to claim then, as we do now, that philosophy belongs with the arts, not the sciences.^xv

Moreover, these sciences were organized into a hierarchy: the theologians felt entitled to order the philosophers to demonstrate the rationality of belief in an immortal soul (despite the fact that Aristotle had not been of this view: philosophical arguments against the immortality of the soul were condemned by the theologians of Paris in 1270); the philosophers felt entitled to order the mathematicians to prove that all motion in the heavens is circular, because only circular movement can be uniform, permanent and unchanging, and to demonstrate that the earth is the centre of all these heavenly circles.^xvi A basic description of the Scientific Revolution is to say that it represented a successful rebellion by the mathematicians against the authority of the philosophers, and of both against the authority of the theologians.¹⁸ A late example of this rebellion is apparent in Newton’s title Mathematical Principles of Natural Philosophy – a title which is a deliberate act of defiance.^xvii An early example is provided by Leonardo da Vinci (d.1519), who in his posthumous Treatise on Painting ^xviii wrote: ‘No human investigation can be termed true science if it is not capable of mathematical demonstration. If you say that the sciences which begin and end in the mind are true, that is not to be conceded, but is denied for many reasons, and chiefly the fact that the test of experience^xix is absent from these exercises of the mind, and without it nothing can be certain.’ In saying this, Leonardo, who was an engineer as well as an artist, was rejecting the whole of Aristotelian natural philosophy (which is what he means by ‘the sciences which begin and end in the mind’) and confining true sciences to those forms of knowledge which were simultaneously mathematical and grounded in experience; arithmetic, geometry, perspective, astronomy (including cartography) and music are the ones he mentions. He realized that the mathematical sciences were often dismissed as ‘mechanical’ (that is, tainted by a close relationship to manual labour), but he insisted that they alone were capable of producing true knowledge. Later readers of Leonardo could not believe he had meant what he said, but he surely did.¹⁹ And, as a consequence of this rebellion of the mathematicians, philosophy in modern times has been demoted from pure science to mere art.

A key part of philosophy, as that discipline was inherited from Aristotle and taught in the universities, was the study of nature – ‘nature’ coming from the Latin word natura, for which the Greek equivalent is physis. For Aristotelians, the study of nature was about understanding the world, not changing it, so there was no art (or technology) associated with the science of nature. And because nature was the embodiment of reason it was, in principle, possible to deduce how things had to be. For Aristotle, the ideal science consisted of a chain of logical deductions from incontestable premises.^xx

When an alternative to Aristotelian natural philosophy, an alternative calling itself at first the ‘new philosophy’ (a term we have seen John Donne taking up in 1611), developed in the course of the seventeenth century, there was an obvious need to find a vocabulary to describe the new knowledge.^xxi The word that we use in modern English, ‘science’, was too vague: as we have seen, there were already lots of sciences. One option – the one most frequently adopted – was to continue to use the terms of Latin origin: ‘natural philosophy’ and ‘natural philosopher’.^xxii Since these were terms associated with higher status and bigger salaries, it was inevitable that the new philosophers tried to lay claim to them:²⁰ Galileo, for example, who had been a professor of mathematics, became in 1610 philosopher to the Grand Duke of Tuscany.^xxiii (Hobbes held that Galileo was the greatest philosopher of all time.)²¹ For some, the only real philosophy was natural philosophy: thus Robert Hooke, one of the first people to be paid to carry out experiments, states baldly, ‘The Business of Philosophy is to find out a perfect Knowledge of the Nature and Properties of Bodies’, and to work out how to put this knowledge to use. This was what he called ‘true science’.²² This usage of the terms ‘philosophy’ and ‘philosopher’ survived much longer than one might think. In 1889 Robert Henry Thurston published The Development of the Philosophy of the Steam Engine: by ‘philosophy’ he meant ‘science’.

But the term ‘natural philosophy’ was unsatisfactory because it implied the new philosophy was rather like the old, that it had no practical application. There was another option, which was to use an existing phrase that avoided the term ‘philosophy’ – ‘natural science’ – and this usage was common in the seventeenth century.^xxiv (It was only in the nineteenth century that ‘science’ came to be generally used as a shorthand for ‘natural science’.) An even more general term was available: ‘natural knowledge’. The student of nature needed a name, hence a new word appeared in the late sixteenth century, ‘naturalist’ – only later did ‘naturalist’ come to refer specifically to someone who studied living creatures (as late as 1755 Dr Johnson in his Dictionary defined a naturalist as ‘a person well versed in natural philosophy’). An alternative to ‘naturalist’ was ‘natural historian’, a term derived from Pliny’s Naturalis historia (78 CE): but Pliny’s reputation fell as a consequence of the new science, and unsophisticated natural histories were soon being replaced by more elaborate programmes of observation.

If Latin offered no perfect solution, what about Greek? The obvious solutions were ‘physic(s)’ (or ‘physiology’) and ‘physician’ (or ‘physiologist’).^xxv Both sets of terms, like their Greek originals, included the whole study of nature, animate and inanimate – thus Boyle’s Physiological Essaysof 1661 are about natural science as a whole. But both had already been claimed by the doctors (medicine was for a long time the only ‘art’ based on a science of nature), which was a considerable inconvenience. Nevertheless, English intellectuals in the second half of the seventeenth century used ‘physicks’ to mean ‘knowledge of Nature’ or ‘natural philosophy’ (as opposed to ‘physick’, meaning medicine). For the Presbyterian minister Richard Baxter, ‘[T]rue Physicks is the Knowledge of the knowable works of God,’ and for John Harris, who was giving public lectures on the new science from 1698, ‘Physiology, Physicks, or Natural Philosophy, is the Science of Natural Bodies,’²³ although he acknowledges that some also use the term ‘physiology’ to refer to ‘a Part of Physick that teaches the Constitution of the Body’. Harris was here still using ‘physiology’ in a sense which remained commonplace until late in the eighteenth century – it is the original sense of the word, preceding its use to refer to the study of human biology. Someone who studied natural philosophy was a ‘physiologist’. It was not until the nineteenth century that ‘physiology’ was definitively ceded to the doctors, while the natural scientists redefined ‘physics’ to exclude ‘biology’ (a word invented in 1799) and, alongside the word ‘physics’, a new one was introduced, ‘physicist’.²⁴

A further solution was to invent a term which reflected the way in which the new knowledge crossed over between the traditional disciplines of natural philosophy (which included what we now call physics) and mathematics (which included mechanics and astronomy). Hence the use of terms such as ‘physico-mathematical’ and ‘physico-mechanical’, as in ‘physico-mechanical experiments’, and even the peculiar hybrids ‘mechanical philosophy’ and ‘mathematical philosophy’.^xxvi

Thus we are not dealing with a transformation reflected in a single pair of terms – ‘natural philosophy’, which, in the nineteenth century, became ‘science’.²⁵ Instead, there is a complicated network of terms, and a change in the meaning of one term results in adjustments in the meaning of all the others.²⁶ The most striking innovation of the nineteenth century, as far as the language of science is concerned, was the introduction of the word ‘scientist’. But the fact that there were no persons called ‘scientists’ before 1833, when William Whewell coined the term, does not mean that there was no word for someone who was an expert in natural science – they were called ‘naturalists’, or ‘physiologists’, or ‘physicians’; in Italian they were scienziati, in French savants, in German Naturforscher, and in English virtuosi.²⁷ Robert Boyle’s The Christian Virtuoso (1690) is about someone who is ‘addicted to Experimental Philosophy’.²⁸ As terms such as virtuosi came to seem old-fashioned they were replaced by the phrase ‘men of science’, which in the sixteenth and seventeenth centuries was used to refer to all those who had a liberal or philosophical education (‘men of a science, not a trade’), but which in the course of the eighteenth century began to be more narrowly used to refer to the people we call ‘scientists’.^xxvii

The word ‘scientist’ was very slow to become established for the straightforward reason that it was (like our word ‘television’) an illegitimate hybrid of Latin and Greek. The geologist Adam Sedgwick (d.1873) scribbled in the margin of his copy of a book by Whewell, ‘[B]etter die of this want than bestialize our tongue by such barbarisms.’²⁹ As late as 1894 Thomas Huxley (‘Darwin’s bulldog’) insisted that no one with any respect for the English language would use the word, which he found ‘about as pleasing a word as “Electrocution”’ (a Greek–Latin, rather than a Latin–Greek, hybrid) – and even at that time he was not alone.^xxviii We can usefully contrast ‘scientist’ in this respect with the uncontroversial word ‘microscopist’ (1831), a word properly formed because it was made entirely out of Greek materials.³⁰ If we look at other European languages, only Portuguese has followed English in creating a linguistic hybrid: cientista. The claim that ‘[t]he word “scientist” was not coined until 1833 because only then did people realize it was needed’ is thus mistaken: there had long been a perceived need for a word to do the job.³¹ The problem was that finding a suitable word – one that did not already have a different usage and was properly constructed – was a genuine obstacle, so that only when the need became absolutely pressing was the obstacle overcome, and only then by breaking what was regarded as one of the basic rules of word formation. Fundamentally, though, the word ‘scientist’ was merely a new and useful word for a type of person who had long been in existence.³²

The word ‘scientific’ falls between classical ‘science’ and nineteenth-century ‘scientist’. Scientificus (from scientia and facere, knowledge-making) is not a term in classical Latin; it was invented by Boethius early in the sixth century. In English, apart from a couple of occurrences in a text of 1589, ‘scientific’ does not appear until 1637, after which date it becomes increasingly common. It has three main meanings: it can refer to a certain type of expertise (‘scientific’, as opposed to ‘mechanical’; the learning of a scholar or gentleman, as opposed to that of a tradesman); to a demonstrative method (that is, by Aristotelian syllogisms); but, in a third sense (as in ‘the scientifick measuring of Triangles’, 1645, in a work on surveying), it refers to the new sciences of the Scientific Revolution. In French the word scientifique was introduced earlier, in the fourteenth century, in the sense of knowledge-making; in the seventeenth century it was used to refer to the abstract and speculative sciences, and it only begins to be used as the equivalent of the English word ‘scientist’ – un scientifique – in 1895, around the same time as the English word began to be widely used.³³

In each European language, of course, the pattern was slightly different. In seventeenth-century French we find the equivalent terms to the English ‘physician’ (physicien) and ‘naturalist’ (naturaliste). In French, physicien had never been a term used for doctors, so the word was conveniently available to mean a natural scientist, and then to evolve to become the French equivalent for ‘physicist’.^xxix In Italy, by contrast, the link between fisico and medicine was already strong in the sixteenth century, and the new philosophers rarely called themselves fisici;³⁴ but then Italian already had to hand a word, scienziato (man of knowledge), lacking in English, and still lacking in French (scientiste is nearly always used pejoratively to mean someone who makes a fetish of being scientific).

To claim, as is often done, that there was no science until there were ‘scientists’ is therefore simply to betray an ignorance of the evolution of the language for knowledge of nature, and for the knowers of nature, between the seventeenth and the nineteenth centuries.³⁵ Those who hesitate to use the words ‘science’ and ‘scientist’ for the seventeenth century, convinced that they are anachronistic, do not understand that all history involves translation from one language into another, and that ‘science’ is simply an abbreviation for a perfectly commonplace seventeenth-century term, ‘natural science’, just as ‘scientist’ is simply a substitute for ‘naturalist’, ‘physician’, ‘physiologist’ and ‘virtuoso’. The first formal meeting of the group that would become the Royal Society discussed forming an association to promote ‘Physico-Mathematicall-Experimentall Learning’: they were making perfectly clear that their enterprise was not natural philosophy as traditionally understood but the new type of knowledge that had resulted from the mathematicians invading the territory of the philosophers.³⁶

It has also been claimed that there were no scientists in the seventeenth century because there was no professional role for a scientist to occupy. ‘There were no scientists in Stuart England,’ we are told, ‘and all the men we have grouped together under that heading were in their varying degrees dilettantes.’³⁷ By the same argument, Hobbes, Descartes and Locke were not philosophers, in that no one paid them to write philosophy; the only proper philosophers in the seventeenth century, it would follow, were scholastic philosophers, employed by universities and Jesuit colleges. Some of the new scientists were, like the new philosophers, in this sense indeed amateurs, not professionals: Robert Boyle, after whom Boyle’s law is named, was independently wealthy and a profession would have been beneath his dignity as the son of an earl. John Wilkins, who wrote extensively on scientific questions, was a clergyman and eventually a bishop, but when the Royal Society was founded in 1662 he had already been Warden of Merton College, Oxford, and Master of Trinity College, Cambridge (appointed by the regime of Oliver Cromwell), although his university career had been wrecked by the Restoration, and he was then forced to fall back on ecclesiastical preferment.^xxx Charles Darwin, too, of course, was an amateur not a professional scientist.^xxxi

However, it would be quite wrong to think of the new science as primarily an amateur – that is, unpaid – activity. In this respect it is unlike the new philosophy of Hobbes, Descartes and Locke: they belonged to no profession, but the new scientists were for the most part practising science as part of their paid employment. Giovanni Battista Benedetti (1530–90, mathematician and philosopher to the Duke of Savoy),^xxxii Kepler (mathematician to the Holy Roman Emperor) and Galileo (for eighteen years a professor of mathematics) were not dilettantes or amateurs: they were professional mathematicians, engaging with problems that were part of the university curriculum, even if their solutions to those problems were quite unlike those taught in the universities. Tycho Brahe, as we have seen, received state funding. Mathematical instrument making and cartography were both commercial undertakings (both were conducted by Gerardus Mercator (1522–99), for example).

Nor was there a shortage of such people in Stuart England. Robert Hooke (d.1703), Denis Papin (d.1712), and Francis Hauksbee (d.1713) were all paid by the Royal Society to perform experiments, although only Hooke received a regular salary.^xxxiii Christopher Wren, a founding member of the Royal Society and now best remembered as an architect, was Savilian Professor of Astronomy at the University of Oxford, holding a chair founded in 1619, having previously held the chair of astronomy at Gresham College in London (founded 1597); astronomy was universally recognized as a branch of mathematics, and architecture required mathematical skills. Isaac Newton was Lucasian Professor of Mathematics at Cambridge, holding a chair founded in 1663. In so far as there was a professional role occupied by the new scientists, it was that of mathematician, and there were plenty of people making a profession out of mathematics outside the two universities: Thomas Digges (1546–1595), for example, who played an important role in the largest engineering project of the Elizabethan era, the rebuilding of Dover harbour, and also sought to turn England into an elective monarchy, or Thomas Harriot (d.1621), whose skills as an astronomer, navigator, cartographer and military engineer led to his being hired to form part of Raleigh’s expedition to Roanoke (1585).³⁸ Thus there were plenty of mathematicians who saw the new philosophy as falling within their area of professional expertise.³⁹ And, naturally, the crucial topics for the new science corresponded neatly with the professional preoccupations of seventeenth-century mathematicians: astronomy/astrology, navigation, cartography, surveying, architecture, ballistics and hydraulics.⁴⁰

It would be perfectly sensible to avoid the words ‘science’ and ‘scientist’ when talking about the seventeenth century if the introduction of these words marked a real moment of change, but ‘science’ is simply an abbreviation for ‘natural science’, while ‘scientist’ marks not a change in the nature of science, or even a new social role for scientists, but a change in the course of the nineteenth century in the cultural significance of classical learning – a change which has become incomprehensible to those historians of science who have not received even the rudiments of a classical education.

§ 3

Although Copernicus, Galileo and Newton were well aware that their ideas were momentous, and we can legitimately describe their work as revolutionary, they never explicitly said to themselves, ‘I am making a revolution.’ The word ‘revolution’ was rarely used even in Newton’s lifetime to refer to large-scale transformations, and almost never before the Glorious Revolution of 1688, the year after the publication of his Principia; even then it was at first confined to political revolutions.^xxxiv 41 Butterfield was right to stress that the historian must aspire to understand the world from the point of view of those alive at the time,^xxxv but, as we have seen, just understanding the world from their point of view can never be enough. The historian has to mediate between the past and the present, finding a language which will convey to present readers the beliefs and convictions of people who thought quite differently. All history therefore involves translation from the source language – here that of seventeenth-century mathematicians, philosophers and poets – into the target language – here, that of the early twenty-first century.⁴² Thus the historian properly translates ‘natural science’ into ‘science’ and ‘physiologer’ into ‘scientist’.

But perhaps there is more than an issue of translation here? In Newton’s language, it might be claimed, there is not only no single word or phrase equivalent to our word ‘revolution’, but the very concept is lacking. Newton’s culture, it could be argued, was inherently conservative and traditionalist; Newton could not have formulated the idea of a revolution even if he had wanted to. In Chapter 3 we will see that, although it may be a helpful generalization to describe Renaissance and seventeenth-century culture, in many respects, as backward looking, there are important exceptions, and it was the exceptions that made modern science possible. For the moment, though, let us just note that there is a word which, for Protestants at least, had many of the connotations of ‘revolution’, and that word is ‘reformation’. In the space of a few decades, between 1517 and 1555, Luther and Calvin had transformed the doctrines, rituals and social role of Christianity; they had made a revolution, one that gave rise to one hundred and fifty years of religious warfare. And so, before the Scientific Revolution was a revolution, it was a reformation. ‘The main Design,’ wrote Hooke in 1665, of his own efforts and those of the Royal Society, was ‘a reformation in Philosophy’.⁴³ Thomas Sprat, writing a history of the Royal Society in 1667, repeatedly compared the reformation in natural philosophy with the earlier reformation in religion.^xxxvi 44

Sprat went on to acknowledge that there were some hardliners who were so hostile to all aspects of ancient learning that they wanted to abolish Oxford and Cambridge. He compared these zealots to the men who had set out to abolish episcopacy in England but had ended up executing the king and establishing the Commonwealth:

I confess there have not bin wanting some forward Assertors of new Philosophy, who have not us’d any kind of Moderation towards them [the universities]: But have presently concluded, that nothing can be well-done in new Discoveries, unless all the Ancient Arts be first rejected, and their Nurseries abolish’d. But the rashness of these mens proceedings, has rather prejudic’d, than advanc’d, what they make shew to promote. They have come as furiously to the purging of Philosophy, as our Modern Zealots did to the reformation of Religion. And the one Party is as justly to be condem’d, as the other. Nothing will suffice either of them, but an utter Destruction, Root and Branch,^xxxvii of whatever has the face of Antiquity.⁴⁵

Thus Sprat acknowledged that some of the advocates of the new science reminded him of the regicides (the monarchy, as much as the episcopacy, had ‘the face of Antiquity’) – which is about as close as he could possibly come to calling them revolutionaries. Sprat was publishing seven years after the restoration of the monarchy and in support of a society founded under royal patronage. He needed to distance himself from any link between radicalism in science and radicalism in politics; it is all the more remarkable that he was willing to press home this comparison between some of the proponents of the new philosophy and the men who had, only a few years before, turned the world upside down.

Naturally, in 1790, Antoine Lavoisier, caught up in the midst of the French Revolution, declared that he was making a revolution in chemistry. Lavoisier, unlike Sprat, speaks our language because he was living through a revolution which transformed the language of politics, shaping the language we still speak. Many French intellectuals were already discussing the possibility of a political revolution in the years preceding 1789, and after 1776 the American Revolution presented them with a model.⁴⁶ In France, the word preceded the deed, though not by much.^xxxviii In the seventeenth century Galileo and Newton knew nothing of this language.^xxxix But they and their contemporaries were perfectly clear that they were trying to carry out a radical, systematic change: the fact that they did not have the word ‘revolution’ does not mean that they were obliged to think of knowledge as something stable and unchanging. ‘As to our work,’ wrote an anonymous member of the Royal Society in 1674, ‘we are all well agreed, or should be so, that it is not to whiten the walls of an old house, but to build a new one.’⁴⁷ Tearing down the old and starting again from scratch is what revolutions are all about.

§ 4

Just as over-scrupulous historians refuse to use the words ‘revolution’, ‘science’ and ‘scientist’ when writing about the seventeenth century, they baulk at using Butterfield’s other word, ‘modern’, because it, too, seems to them inherently anachronistic. Yet Renaissance books on warfare often included the word ‘modern’ in their titles to show that they acknowledged the revolutionary consequences of gunpowder.⁴⁸ In the Renaissance, modern music was understood to be quite different from ancient music because it was polyphonic rather than monodic – Galileo’s father, Vincenzo, wrote a Dialogue on Ancient and Modern Music.⁴⁹ Modern maps showed the Americas.^xl

The first history to be written in terms of progress is Vasari’s history of Renaissance art, The Lives of the Artists (1550).⁵⁰ It was quickly followed by Francesco Barozzi’s 1560 translation of Proclus’s commentary on the first book of Euclid, which presented the history of mathematics in terms of a series of inventions or discoveries. Indeed, the mathematicians (who often spent time with the artists, because they taught them the geometry of perspective)^xli were already eager to claim that they, too, were making progress, and had begun to publish books with the word ‘new’ in the title, creating a fashion which spread from mathematics to the experimental sciences: New Theories of the Planets (Peuerbach, written 1454, published 1472); The New Science (Tartaglia, 1537); The New Philosophy (Gilbert, d.1603 – this is the subtitle, or perhaps the proper title, of the posthumously published Of Our Sublunar World; the layout of the title page is ambiguous); The New Astronomy (Kepler, 1609); Two New Sciences (Galileo, 1638); New Experiments Touching the Void (Pascal, 1647); New Anatomical Experiments (Pecquet, 1651); New Experiments Physico-mechanical (Boyle, 1660). The list goes on and on.⁵¹ As the great pioneer of the idea of progress, Bacon wrote The New Organon and The New Atlantis, and his book on The Wisdom of the Ancients (1609) implied a sharp contrast between the ancients and the moderns.

Given all this emphasis on newness, why did scientists not use the word ‘modern’ in the titles to their books? The answer is straightforward. In both Islam and Christianity ‘modern philosophy’ meant post-pagan philosophy.⁵² For William Gilbert, the founder of a new science of magnetism, for example, Thomas Aquinas (1225–74) was a modern philosopher.⁵³ Consequently, he had no interest in describing his own natural philosophy as ‘modern’, preferring to call it ‘new’. In philosophy, unlike warfare and music, the word ‘modern’ was unavailable because it had already been put to a different use. The same was true in architecture, where, in the fifteenth century, ‘modern architecture’ meant Gothic architecture.⁵⁴ In science this began to change only in the course of the debate on the ancients and moderns at the end of the seventeenth century. Jonathan Swift still counts Aquinas among the moderns in The Battle of the Books (1720), but in doing so he is being deliberately old-fashioned.⁵⁵ René Rapin, who was one of the first to set the ancients against the moderns, had redefined the concept of modern philosophy by calling Galileo ‘the founder of Modern Philosophy’ in 1676 – a judgement particularly surprising coming from a Jesuit, given Galileo’s condemnation by the Roman Inquisition in 1633 – but this usage had not caught on in English.⁵⁶ Nevertheless, with a definite article, ‘the modern Philosophy’, or ‘the modern way of Philosophy’, could be used, if a little awkwardly, to refer to contemporary science: Boyle was the first to use it thus, in 1666.⁵⁷ The phrase ‘modern science’ was first used by Gideon Harvey in 1699, in the course of an indiscriminate attack on both the old and the new philosophies.⁵⁸The old philosophy is, by the end of the seventeenth century, scholasticism; modern science is the science of Descartes and Newton.

Just as the word ‘modern’ was slow to establish itself in a scientific context, so too it was only towards the end of the seventeenth century that the word ‘progress’ and those with similar meanings became commonplace. The proper title of the Royal Society, founded in 1660, is The Royal Society of London for the Improving of Natural Knowledge. ‘Improving’ implies progress, so it is not surprising that the fuller title of Thomas Sprat’s History of the Royal Society was The History of the Institution, Design and Progress of the Royal Society of London for the Advancement of Experimental Philosophy – ‘experimental philosophy’ being, of course, yet another term for what we now call ‘science’, and ‘progress’ being used here somewhat ambiguously between its old meaning (a journey, and so a process of change) and its new meaning (a process of improvement); ‘advancement’ is another progress-related word. A year later Joseph Glanvill published Plus ultra: or the Progress and Advancement of Knowledge since the Days of Aristotle. By the end of the century progress was taken for granted, as in the title of Daniel Le Clerc’s The History of Physick, or, An Account of the Rise and Progress of the Art, and the Several Discoveries Therein from Age to Age (1699).⁵⁹ Before the word ‘progress’ became fashionable, Robert Boyle twice used as his epigraph a passage from Galen: ‘We must be bold and go hunting for the truth; even if we do not come right up to it, at least we will get closer to it than we are now.’⁶⁰ Boyle is using the hunt as a metaphor for progress. It is this triumph of the idea of progress, along with the redefinition of the word ‘modern’, which marks the end of the first phase of the long Scientific Revolution through which we are still living.⁶¹

In any case, there were alternatives to the language of progress which served exactly the same purpose – the languages of invention and discovery. In 1598 Brahe insisted the new geoheliocentric system of the cosmos was his own invention – he was laying claim to having invented a theory in exactly the same way as he claimed to have invented the astronomical sextant. Others had tried to steal the credit for the geoheliocentric system from him but, properly, it belonged to him alone.⁶² Galileo, when in 1610 he announced to the world what he had seen through his telescope, was compared to his fellow Florentine Amerigo Vespucci, to Christopher Columbus and to Ferdinand Magellan.⁶³ In discovering the moons of Jupiter, Galileo had discovered new worlds, just as the navigators had done. Thereafter every scientist dreamed of making comparable discoveries. Here is the first professional scientist, Robert Hooke (1635–1703), writing in a hurry. Plenty of people, in all ages, he says, have enquired ‘into the nature and causes of things’:

But their endeavours, having been only single and scarce ever united, improved, or regulated by art, have ended only in some small inconsiderable product hardly worth naming. But though mankind have been thinking these six thousand years, and should be so six hundred thousand more, yet they are and would be much whereabouts they were at first, wholly unfit and unable to conquer the difficulties of natural knowledge. But this newfound world must be conquered by a Cortesian army, well-disciplined and regulated, though their number be but small.⁶⁴

The Royal Society was to be this ‘Cortesian army, well-disciplined and regulated, though their number be but small’. Hooke’s imagery is misleading – and he was misled by it. He was opposed not by the Aztecs but by Aristotelian philosophers. He did not need to conquer nature in order to understand it. His army did not need to be disciplined and regulated; competition (as we will see in Chapter 3) provided the only discipline it needed. But he was right about the fundamentals. He had chosen the Cortesian army as his image because he wanted to conjure up in the mind the most convulsive, irreversible transformation recorded in history; because he wanted to discover new worlds; and because he wanted his discoveries to benefit his own society, just as the conquest of the New World had enriched Cortes’s Spain. Hooke’s key terms were not ‘science’, or ‘revolution’, or ‘progress’, but it is a reasonable translation of his own terms (‘natural knowledge’, ‘newfound world’, ‘Cortesian army’) into our language to say that he was dreaming of what we call the Scientific Revolution.

He was not alone: ‘The Aristotelian Philosophy is inept for New discoveries,’ wrote Joseph Glanvill in 1661. ‘There is an America of secrets, and [an] unknown Peru of Nature,’ yet to be found:

And I doubt not but posterity will find many things, that are now but Rumors, into practical Realities. It may be some Ages hence, a voyage to the Southern unknown Tracts, yea possibly the Moon, will not be more strange then one to America. To them, that come after us, it may be as ordinary to buy a pair of wings to fly into remotest Regions; as now a pair of Boots to ride a Journey. And to conferr at the distance of the Indies by Sympathetick conveyances, may be as usual to future times, as to us in a litterary correspondence . . . Now those, that judge by the narrowness of former Principles, will smile at these Paradoxical expectations: But questionless those great Inventions, that have in these later Ages altered the face of all things; in their naked proposals, and meer suppositions, were to former times as ridiculous. To have talk’d of a new Earth [the New World of the Americas] to have been discovered, had been a Romance to Antiquity: And to sayl without sight of Stars or shoars by the guidance of a Mineral [the compass], a story more absurd, then the flight of Daedalus.⁶⁵

And of course Glanvill was right: we do fly and ‘conferr’ at a distance; we have been not only to Australia but also to the Moon.

Thomas Hobbes, writing in 1655, thought that there was no astronomy worth the name before Copernicus, no physics before Galileo, no physiology before William Harvey. ‘But since these, astronomy and natural philosophy in general have, for so little time, been extraordinarily advanced . . . Natural Philosophy is therefore but young.’⁶⁶ But it was Henry Power (one of the first Englishmen to experiment with the microscope and the barometer) who, in 1664, gave most eloquent expression to the idea that knowledge was being transformed and that the new knowledge was quite unlike the old:

And this is the Age wherein all mens Souls are in a kind of fermentation, and the spirit of Wisdom and Learning begins to mount and free it self from those drossie and terrene Impediments wherewith it hath been so long clogg’d, and from the insipid phlegm and Caput Mortuum of useless Notions, in which it has endured so violent and long a fixation.

This is the Age wherein (me-thinks) Philosophy comes in with a Spring-tide; and the Peripateticks may as well hope to stop the Current of the Tide, or (with Xerxes) to fetter the Ocean, as hinder the overflowing of free Philosophy: Me-thinks, I see how all the old Rubbish must be thrown away, and the rotten Buildings be overthrown, and carried away with so powerful an Inundation. These are the days that must lay a new Foundation of a more magnificent Philosophy, never to be overthrown: that will Empirically and Sensibly canvass the Phaenomena of Nature, deducing the Causes of things from such Originals in Nature, as we observe are producible by Art, and the infallible demonstration of Mechanicks: and certainly, this is the way, and no other, to build a true and permanent Philosophy . . .⁶⁷

In 1666 the mathematician and cryptographer John Wallis (who introduced the symbol ∞ for infinity) wrote, more circumspectly, ‘For, since that Galilæo, and (after him) Torricellio, and others, have applied Mechanick Principles to the salving of Philosophical Difficulties; Natural Philosophy is well known to have been rendered more intelligible, and to have made a much greater progress in less than an hundred years, than before for many ages.’⁶⁸

Hooke, Glanvill, Hobbes, Power and Wallis were participants in this transformation; but their understanding of what was taking place was shared by well-informed bystanders. In 1666 Bishop Samuel Parker hailed the recent triumph of ‘the Mechanical and Experimental Philosophie’ over the philosophies of Aristotle and Plato, and claimed that:

we may rationally expect a greater Improvement of Natural Philosophie from the Royal Society, (if they pursue their design) then it has had in all former ages; for they having discarded all particular Hypotheses, and wholly addicted themselves to exact Experiments and Observations, they may not only furnish the World with a compleat History of Nature, (which is the most useful part of Physiologie [natural science]) but also laye firm and solid foundations to erect Hypotheses upon.⁶⁹

Parker (with good reason) thought the great improvement in knowledge was just about to happen, now that the right method of enquiry had been established. Only two years later the poet John Dryden (also with good reason) took the view that it was already well under way:

Is it not evident, in these last hundred years (when the Study of Philosophy has been the business of all the Virtuosi in Christendome) that almost a new Nature has been reveal’d to us? that more errours of the [Aristotelian] School have been detected, more useful Experiments in Philosophy have been made, more Noble Secrets in Opticks, Medicine, Anatomy, Astronomy, discover’d, than in all those credulous and doting Ages from Aristotle to us? so true it is that nothing spreads more fast than Science, when rightly and generally cultivated.⁷⁰

Dryden’s chronology is right: ‘these last hundred years’ takes us back, almost exactly, to the nova of 1572. His vocabulary is exemplary: he uses Virtuosi to mean scientists and ‘Science’ to mean, well, science.^xlii He sees that the new science relies on new standards of evidence. He acknowledges the possibility of relativism (how many new natures might there be?) while insisting that the new science is not just some sort of local fashion but an irreversible transformation in our knowledge of nature.⁷¹

§ 5

One could go on accumulating evidence of this sort for the validity of the idea of the Scientific Revolution, but plenty of scholars would remain unpersuaded and unpersuadable. The anxiety which now troubles historians when they read the words ‘scientific’, ‘revolution’, ‘modern’ and (worst of all) ‘progress’ in studies of seventeenth-century natural science is not just a fear of anachronistic language; it is a symptom of a much larger intellectual crisis which has expressed itself in a general retreat from grand narratives of every sort.^xliii The problem, it is claimed, with grand narratives is that they privilege one perspective over another; the alternative is a relativism which holds that all perspectives are equally valid.

The most influential arguments in favour of relativism flow from the philosophy of Ludwig Wittgenstein (1889–1951).^xliv Wittgenstein taught in Cambridge on and off from 1929 to 1947 – he left the year before Butterfield lectured on the Scientific Revolution – but it would never have occurred to Butterfield that he needed to consult Wittgenstein, or indeed any other philosopher, to learn how to think about science. It was not until the late 1950s, following the publication of the Philosophical Investigations in 1953, that arguments drawn from Wittgenstein began to transform the history and philosophy of science; their influence can already be seen, for example, in Thomas Kuhn’s The Structure of Scientific Revolutions.⁷² Thereafter it became common to claim that Wittgenstein had shown that rationality was entirely culturally relative: our science may be different from that of the ancient Romans, but we have no grounds for claiming that it is better, for their world was utterly unlike ours. There is no common standard by which the two can be compared. Truth, according to Wittgenstein’s doctrine that meaning is use,⁷³ is what we choose to make it; it requires a social consensus but not any correspondence between what we say and how the world is.⁷⁴

This first wave of relativism was later supplemented by other, profoundly different intellectual traditions: the linguistic philosophy of J. L. Austin, the post-structuralism of Michel Foucault, the postmodernism of Jacques Derrida and the pragmatism of Richard Rorty. The phrase ‘the linguistic turn’ is often used to refer to all these different traditions, because they share a common sense of how, as Wittgenstein put it, ‘the limits of my language mean the limits of my world.’^xlv As we shall see in a moment, much of the argument about the Scientific Revolution stems from the ramifications of this view.

Within history of science, one post-Wittgensteinian tradition has been particularly important: it is often called Science and Technology Studies.⁷⁵ This movement originated with Barry Barnes and David Bloor at the Science Studies Unit of the University of Edinburgh (founded in 1964), both of whom were deeply influenced by Wittgenstein (Bloor, for example, wrote Wittgenstein: A Social Theory of Knowledge (1983)). Barnes and Bloor proposed what they called ‘the strong programme’. What makes the strong programme strong is the conviction that the content of science, and not just the ways in which science is organized, or the values and aspirations of scientists, can be explained sociologically. Its essence lies in the principle of symmetry: the same sorts of explanation must, according to this principle, be given for all types of knowledge claims, whether they are successful or not.^xlvi Thus if I meet someone who claims the earth is flat, I will seek a psychological and/or a sociological explanation for their peculiar belief; when I meet someone who claims that the Earth is a sphere floating through space and orbiting the sun, I must look for exactly the same sorts of explanation for this belief too. The strong programme insists that it is illegitimate to say that the explanation for the second belief is that it is right, or even that people believe it because they have good evidence for it. It thus systematically excludes from consideration the very feature which makes scientific arguments distinctive: their appeal to superior evidence. No follower of Wittgenstein can accept the notion of ‘evidence’ uncritically – indeed, some would claim they cannot accept it at all. Bertrand Russell first met Wittgenstein in 1911. In the brief obituary he wrote forty years later a memory from their early encounters imposed itself upon him:

Quite at first I was in doubt as to whether he was a man of genius or a crank, but I very soon decided in favour of the former alternative. Some of his early views made the decision difficult. He maintained, for example, at one time that all existential propositions are meaningless. This was in a lecture room, and I invited him to consider the proposition: ‘There is no hippopotamus in this room at present.’ When he refused to believe this, I looked under all the desks without finding one; but he remained unconvinced.⁷⁶

No one should be surprised if histories and philosophies of science which start from Wittgenstein seem to miss the very thing that science is all about.^xlvii

Barnes and Bloor are sociologists, and it is perfectly understandable that they insist that they and their fellow sociologists ought to stick to sociological explanations. But they go well beyond that. A relativistic view, that science is not a way of coming to grips with reality, is not the conclusion these scholars draw from their research; it is the assumption (following their interpretation of Wittgenstein) they build into it. In order to justify this, proponents of this position insist that evidence is never discovered: it is always ‘constructed’ within a particular social community. To say one body of evidence is superior to another is to adopt the viewpoint of one community and reject that of another. The success of a scientific research programme thus depends not on its ability to generate new knowledge but on its ability to mobilize the support of a community. As Wittgenstein puts it, ‘At the end of reasons comes persuasion. (Think of what happens when missionaries convert natives.)’⁷⁷

These scholars present science as being about rhetoric, persuasion and authority because the symmetry principle obliges them to assume that that is all it can be about. In doing so they go directly against the views of the early scientists themselves. Thus an influential article is entitled ‘Totius in verba: Rhetoric and Authority in the Early Royal Society’, though the motto of the Royal Society was nullius in verba (‘take no man’s word for it’; or alternatively, perhaps, ‘words don’t count’) – the claim of the Society’s founders being that they were escaping from forms of knowledge based on rhetoric and authority.^xlviii A form of history which presents itself as acutely sensitive to the language of people in the past proceeds by dismissing out of hand what they said, over and over again, about themselves. Anachronism, driven in disgrace out of the back door, re-enters in triumph through the front.

It may be hard to believe, but proponents of the strong programme have acquired a dominant position within the history of science. The most striking example of this approach in action is Steven Shapin and Simon Schaffer’s Leviathan and the Air-pump (1985), generally acknowledged as the most influential work in the discipline since Thomas Kuhn’s Structure of Scientific Revolutions.^xlix The new history of science offered, in Steven Shapin’s phrase, a social history of truth.^l Scientific method, it was now argued, kept changing, so that there was no such thing as the scientific method: a famous book by Paul Feyerabend was entitled Against Method,^li its catchphrase ‘Anything goes’; it was followed by Farewell to Reason.⁷⁸ Some philosophers and nearly all anthropologists agreed: standards of rationality were, they insisted, local and highly variable.⁷⁹

But we must reject the Wittgensteinian notion that truth is simply consensus, a notion incompatible with an understanding of one of the fundamental things science does, which is to show that a consensus view must be abandoned when it is at odds with the evidence.^lii The classic text here is Galileo’s ‘Letter to Christina of Lorraine’ (1615) in defence of Copernicanism. He starts by saying that there are certain matters on which the philosophers have all been agreed, but that he has discovered with his telescope facts that are entirely at odds with their beliefs; consequently, they need to change their views.⁸⁰ What seemed to be true can no longer be regarded as true. What Galileo is engaged in here (may even be said to be inventing) is what Shapin and Schaffer call ‘the empiricist language-game’, according to which facts are ‘discovered rather than invented’.⁸¹That’s true; the Wittgensteinian move then consists, in the view of his followers, in maintaining that there are no grounds for thinking that this game has greater validity than any other, and Galileo then becomes no more reasonable than the philosophers he is opposing.^liii And at this point Wittgensteinian history of science places itself directly at odds with Galileo’s own account of what he is actually doing, and history of science is in a direct conflict with science.^liv

When Shapin and Schaffer refer to ‘the empiricist language-game’ as if it were just one amongst any number of equally valid language-games, they assume that there is no reality outside the language-games of Galileo and his opponents because it is the language-games themselves which define what is to count as real; they assume that ‘the limits of my language mean the limits of my world.’^lv This cannot be true in any absolute sense; Galileo’s telescope transformed the world of astronomers before they had any new words for what they could now see – before they even had the word ‘telescope’. In writing about his discoveries Galileo was not obliged to write in a way that others found puzzling or incomprehensible: it was what he said that caused consternation, not how he said it. Yet, although the philosophers understood him perfectly well, some of them continued to insist that what he and other astronomers claimed to see couldn’t possibly be there. Galileo’s world and their world had different limits, although they understood each other perfectly. The limits were not set by their language but by their priorities, by their sense of what was negotiable and what was not.^lvi

The telescope may seem a special case. Of course our world changes when we introduce a new technology, or go somewhere we have never been before. But every day we experience things for which we have no words, and in such circumstances we are often lost for words, or find ourselves saying that something is ‘beyond words’. Only later do we sometimes find the words (love, grief, jealousy, despair) for what we have been feeling all along. ‘It did not occur to him,’ writes Tolstoy of Andrei, ‘that he was in love with Miss Rostov.’ And the whole, wonderful point of some experiences – music, sex, laughter – is that there are not and never will be any adequate words to describe them. This does not mean they do not exist.

But even though it is by no means always the case that ‘the limits of my language are the limits of my world’, we must acknowledge that our language often determines the limits of what we can argue about and understand with precision. Clouds were only named early in the nineteenth century – ‘cirrus’ and ‘nimbus’ may sound old because they are Latin, but the Romans had no names for the different types of cloud.⁸² Of course, long before there was a language for clouds, people experienced them more or less as we do: one only has to look at seventeenth-century Dutch seascapes to see all the different types of cloud accurately represented, even though the painters had no names for them. Robert Hooke evidently saw clouds perfectly clearly when he asked, ‘What is the reason of the various Figure of the Clouds, un dulated, hairy, crisped, coyled, confus’d, and the like?’⁸³ But he is well aware that describing clouds is at the limit of his linguistic capacities. The naming of clouds was a great event in the history of meteorology, after which much more serious discussion and understanding were possible.

When we are studying ideas, linguistic change is the key to finding out what people understood that their predecessors did not. A decade before Galileo’s telescopic discoveries, William Gilbert, the first great experimental scientist of the new age, had acknowledged: ‘Sometimes therefore we use new and unusual words, not that by means of foolish veils of vocabularies we should cover over the facts [rebus] with shades and mists (as Alchemists are wont to do) but that hidden things which have no name, never having been hitherto perceived, may be plainly and correctly enunciated.’^lvii 84 His book begins with a glossary to help the reader make sense of these new words. Then a few months after Galileo discovered what we call the moons of Jupiter (Galileo does not call them moons, but first stars and then planets), Johannes Kepler invented a new word for these new objects: they were ‘satellites’.^lviii Thus historians who take language seriously need to search out the emergence of new languages, which must represent transformations in what people can think and how they can conceptualize their world.^lix

It is important here to distinguish this claim from the argument with which this chapter started. The historian has always to learn the language which people in the past used, and must always be alert to changes in that language; that does not mean they need always use that language when writing about the past. Kepler’s word ‘satellite’ acknowledges that Galileo had discovered a new type of entity, but it is perfectly sensible for us to say that what Galileo had discovered were the moons of Jupiter (a terminology used neither by Galileo nor Kepler – the earliest usage I can find is 1665 – and thus, strictly speaking, anachronistic), particularly as, for us, stars (Galileo’s term) are fixed, and satellites (Kepler’s term) are usually human-made objects launched into space.

Recent history of science, for all its talk of languages and discourses, has not been nearly attentive enough to the emergence in the seventeenth century of a new language for doing natural science, a language discussed in Part Three of this book. Indeed, so invisible has this new language been that the very same scholars who refuse to use the word ‘scientist’ for anybody before the second half of the nineteenth century happily talk about ‘facts’, ‘hypotheses’ and ‘theories’ as if these were transcultural concepts. This book seeks to remedy this peculiar lapse.^lx We can state one of its core premises quite simply: a revolution in ideas requires a revolution in language. It is thus simple to test the claim that there was a Scientific Revolution in the seventeenth century by looking for the revolution in language that must have accompanied it. The revolution in language is indeed the best evidence that there really was a revolution in science.

There are some features of linguistic change that are worth bearing in mind as we go on. Obviously (as we have already seen in the cases of ‘arts’ and ‘sciences’), over time the meanings of words change. But often words do not simply change their meaning, rather they acquire new meanings, meanings which are sometimes apparently unrelated to their original meaning. We have seen how the word ‘revolution’ is now used in so many different senses that one source of confusion as to whether there is or is not a Scientific Revolution is a failure to distinguish these senses one from another. When I visit the local branch of my bank, I don’t think of this vast business as a tree, but ‘branch’ here is a dead metaphor. So too is the word ‘volume’ when used in the context of measurement: first in French, and then much later in English, ‘volume’ began to be used to refer not to a book but to the space occupied by a three-dimensional object. If I measure the volume of a sphere, the language I am using is a dead metaphor.

When we write of ‘laws of nature’ the word ‘laws’, too, is being used in a metaphorical sense. What are laws of nature? To understand the sort of contexts in which the phrase is used it may help to explore its origins; such an exploration may, in the end, help us understand why there is no good answer to the question ‘What are laws of nature?’ other than an account of how we use the phrase (in this case, as Wittgenstein said, meaning is indeed use). Thus in the United Kingdom we have an unwritten constitution. What is an unwritten constitution? Any decent answer will be full of puzzles and paradoxes but will need to include an account of how the idea that states have constitutions originates with Bolingbroke in 1735, and that the idea of an unwritten constitution distinguishes the United Kingdom from the United States and France, the first countries to have written constitutions. Just as the idea of an unwritten constitution contains apparently irresolvable puzzles once written constitutions become the norm (How do we know what the unwritten constitution is? Whence comes its authority?), so crucial concepts that we use in discussing science (‘discovery’, or ‘laws of nature’) are inherently problematic, at least for us: the only way to understand them is to recover their history.⁸⁵ My argument is that during the seventeenth century the idea of natural science underwent a fundamental revision, and by the end of the century the idea that had taken shape was basically the one that we still have. I don’t claim that idea was consistent or coherent; I claim that it was successful, that it provided a template for the discovery of new knowledge and new technologies.^lxi

§ 6

Much of this chapter has been concerned with the language of science, as much of the book will be, but the book’s argument is also and equally about what Leonardo called ‘the test of experience’. The first generation of historians and philosophers to study the Scientific Revolution down-played the importance of new evidence and new experiments, insisting that what really counted was what Butterfield called ‘a transposition in the mind of the scientist himself’. The foundations of modern science were, the philosopher Edwin Burtt had insisted back in 1924, metaphysical.⁸⁶According to Koyré, ‘[I]t is thought, pure unadulterated thought, and not experience or sense-perception . . . that gives the basis for the “new science” of Galileo Galilei.’⁸⁷ Thus what Koyré took to be the key concept that made possible the new science, the concept of in ertia, was, he held, constructed by Galileo thinking about everyday experience, by a mere thought experiment. This, it seems to me, is to mistake effect for cause, to get the whole story of the new science upside down and back to front.^lxii The Scientific Revolution is precisely about new experiences and new sense-perceptions. It should be obvious that if all that was required for the Scientific Revolution was new thinking, then it would be impossible to explain why it did not take place long before the seventeenth century.^lxiii

But for thirty years now a second generation of historians and philosophers of science has been attacking the claim that the Scientific Revolution vastly improved humankind’s ability to understand nature; adopting a relativist perspective, they have been unwilling to acknowledge that Newton was superior to Aristotle or Oresme, even if only in the sense that his theories made possible better predictions and new types of intervention. Their arguments have convinced almost all anthropologists, nearly every professional historian and many philosophers. But they are wrong. Thanks to the Scientific Revolution, we have a much more reliable type of knowledge than ancient and medieval philosophers ever had, and we call it science. For the first generation, the new science was all in the mind; for the second, it was simply a language-game. These two debates, about thinking and knowing, interlock because both generations have downplayed the idea that the new science was grounded in a new type of engagement with sensory reality. Both thus missed its essential characteristic: that it systematically employed the test of experience.

For the new scientists of the second half of the seventeenth century were in a quite different position from their classical, Arab and medieval predecessors. They had the printing press (a fifteenth-century invention whose impact grew through the seventeenth century), which created new types of intellectual community and transformed access to information; they had a family of instruments (telescopes, microscopes, barometers), all made from glass, that acted as agents of change; they had a new preoccupation with the test of experience, which had now given rise to the experimental method; they had a new, critical attitude to established authority; and they had a new language, the language which we now speak, which made it much easier to think new thoughts. Mutually supporting and interlocking, these diverse elements made possible the Scientific Revolution.

§ 7

In 1748 Denis Diderot, the great Enlightenment philosopher, published, anonymously, an erotic novel entitled The Indiscreet Jewels (the word ‘jewel’ here is a euphemism for the vagina). It was, as he and his publisher must confidently have expected, promptly banned and immediately successful. Chapter 29 is subtitled ‘Perhaps the best, and the least read, of this story’ – least read because, exceptionally, there is no sex in it. It describes how the protagonist (the sultan Mangogul, a flattering representation of Louis XV) has a dream in which he flies on the back of a hippogriff to a vast edifice suspended in the clouds. There, a great crowd of misshapen individuals are gathered around an old man in a pulpit made of cobwebs; he says nothing, but blows bubbles. All are naked, except for a few bits and pieces of cloth here and there on their bodies – these are fragments of the robe of Socrates, and we discover that we are in the temple of philosophy. Suddenly:

I saw in the distance a child who was walking towards us with slow but sure steps. He had a small head, a thin body, weak arms, and short legs, but his limbs grew larger and larger as he advanced. Throughout his successive spurts of growth, he appeared to me in many different guises; I saw him point a long telescope skyward, assess the rate of a falling body with the help of a pendulum, ascertain the weight of the air with a tube full of mercury, and break up light with a prism in his hand. He became a huge colossus; his head touched the skies, his feet were lost in the abyss, and his arms stretched from one pole to the other. With his right hand he shook a torch whose rays spread out in all directions lighting the depths of the waters and penetrating into the bowels of the earth.

The colossus strikes the temple, which collapses, and Mangogul stirs.⁸⁸

‘What,’ Mangogul asks just before he awakes, ‘is this gigantic figure?’ The answer may seem obvious: Diderot is writing about the transformation in knowledge that we now call the Scientific Revolution. As we shall see, Galileo had pointed his telescope skyward, Mersenne (following Galileo’s example) had accurately measured the rate of falling bodies, Pascal had weighed the air, Newton had broken up light with a prism. The new science destroyed the old one taught by the philosophers. But Diderot’s name for this newly born colossus is not ‘Science’, as we might expect. The word ‘Science’ in French was not and is not nearly specific enough to refer to the new sciences of Galileo and Newton, for, as we have seen, there were and are all sorts of sciences, including now the social sciences. Even ‘natural science’ would not do, for the philosophers had always claimed to be experts in natural science, and so ‘natural science’ would no more serve to distinguish the new science from the old than would ‘natural philosophy’. Instead, Plato, who has conveniently turned up to explain what is going on, says, ‘Recognize Experience, for it is she.’^lxivAnd yet surely there is nothing new about experience? Isn’t experience something that all human beings have in common? How then can ‘Experience’ be the right name for the new sciences?

In answering this question I will keep returning to the problem to which Diderot alerts us when he names his colossus ‘Experience’ – the difficulty of finding an adequate language with which to describe the new science, a problem which is not just ours when we seek to understand it but rather was a crucial difficulty for those who invented it, and also for those, like Diderot, who wrote in praise of it. Indeed, I shall argue that the new science would not have been possible without the construction of a new language with which to think, a language necessarily cobbled together out of available words and phrases. That language was pioneered in English, where, for example, ‘experience’ and ‘experiment’ began to diverge in meaning during the seventeenth century. (Diderot, who began his career by translating books from English into French, was well acquainted with this new language.) Diderot’s expérience can thus be translated into English, not quite exactly, as ‘experiment’ (a word which still does not exist in French), and it is immediately apparent that ‘experiment’ might be a helpful word in describing the new science, a more helpful word, perhaps, than ‘experience’, though we have already seen that Leonardo identified experience as the key to reliable knowledge. We can mark the beginning of this process of language construction precisely: it starts with a new word, which inaugurated a wider transformation in the role experience had to play, the word ‘discovery’, which is a word with equivalents in every European language.

In the pages which follow we will see how experience, in the form of observations and experiments directed at making discoveries, came to be something new in the seventeenth century; how this new enterprise of discovery made possible the invention of science; and how this new science began to transform the world, a process which resulted in the modern technologies on which our lives depend. They tell the story of the birth of science, of its infancy, and of its extraordinary transformation into the great colossus under whose shadow we all now live. But Diderot’s peculiar chapter also serves as a warning: with its dream framing, its monsters and its allegories, its linguistic slipperiness, it conveys a sense of difficulty. What would a history of experience – of this new sort of experience – be like?

It might seem far easier for us to answer that question than it was for Diderot, for he was still caught up in the triumph of Newtonianism (which came later in France than in England), while we have all the advantages of hindsight. But Diderot had one great advantage over us: graduating from the Sorbonne in 1732, he had been educated in the world of Aristotelian philosophy. He knew how shocking the destruction of that world had been, for he had experienced it at first hand. From a bird’s-eye view – the historian’s view – the Scientific Revolution is a long, slow process, beginning with Tycho Brahe and ending with Newton. But for the individuals caught up in it – for Galileo, Hooke, Boyle and their colleagues – it represents a series of sudden, urgent transformations. In 1735 Diderot, educated in the old ways, still planned to become a Catholic priest; by 1748, only a little more than a decade later, he was an atheist and a materialist, already at work on the great Encyclopaedia, the first volume of which appeared in 1751. The destruction of the temple of philosophy was not, for him, an historical event; it was a personal experience, the moment when he had awakened from a nightmare.