The Invention of Science

4 Planet Earth

an utterly insignificant little blue-green planet.

– Douglas Adams, The Hitchhiker’s Guide to the Galaxy (1979)¹

§ 1

The voyages of discovery brought about an astonishing transformation in geographical knowledge from 1460 onwards. Where the known world in the first half of the fifteenth century was more or less identical to the world known to an educated Roman at the time of Christ, by the beginning of the sixteenth century it was clear that there were extensive inhabited territories that had been unknown to the Greeks and Romans. Where the conventional view had been that lands close to the equator must be uninhabitable, this had turned out to be nonsense. This expansion of the known world was carefully recorded by cartographers, and it brought about the first great triumph of experience over philosophical theory.

The subject of this chapter, though, is not the voyages of discovery as such. In the wake of Columbus’s discovery of America a silent revolution occurred, the invention of what we now call ‘the terraqueous globe’. This revolution took place in the space of a few years and encountered no (or almost no) resistance. It is of profound importance, but it is completely invisible in the standard historical literature. Thomas Kuhn once wrote:

A historian reading an out-of-date scientific text characteristically encounters passages that make no sense … It has been standard to ignore such passages or to dismiss them as the products of error, ignorance, or superstition, and that response is occasionally appropriate. More often, however, sympathetic contemplation of the troublesome passages results in a different diagnosis. The apparent textual anomalies are artifacts, products of misreading.²

My subject is a whole library of texts that at first sight make no sense. For fifty years now historians of science, inspired by Kuhn, have sought out such texts in order to demonstrate their expertise, their capacity to make sense of the apparently nonsensical, but these particular texts have been almost completely ignored. Why? Because they point to something that isn’t supposed to exist: a silent revolution. According to Kuhn, revolution always brings with it disputation and conflict;³ since there was virtually no disputation, it is all too easy to assume that there can have been no revolution. It is this very anomaly, on the other hand, which makes these texts the perfect place to embark on a new, post-Kuhnian history of science.

What shape is ‘the earth’? The answer to this question must seem obvious. Surely everyone knew that the earth is round? In the nineteenth century it was claimed, in all seriousness, that Columbus’s contemporaries thought the world was flat and expected him to sail over the edge.⁴ This story is balderdash. But the fact that everyone (or at least every properly educated person) thought that you could in principle sail around the world (and in 1519–22 Magellan did just that) does not mean that they thought it was round. Columbus, strangely, thought that the old world, known to Ptolemy, was half of a perfect sphere, but the new world, he believed, was shaped like the top half of a pear, or like a breast; he had the impression he was sailing uphill as he left the Azores behind him.⁵ The stalk, or nipple, of this other hemisphere was the location of the terrestrial paradise.⁶ ‘The earth’ (or rather the agglomerate of earth and water) bulged.

This view, that the agglomerate of earth and water was not a perfect sphere, was universally accepted in the later Middle Ages, and the new cosmography required its refutation.i i ⁷ According to Aristotle, the universe is divided between a supralunary zone, where nothing changes and movement is always in circles, and a sublunary zone. In the sublunary zone the four elements – earth, water, air and fire – which form the basis of all our daily experience of matter are to be found. These elements naturally arrange themselves in concentric circles around a common centre: earth surrounded by water, water surrounded by air, and air surrounded by fire. This arrangement, however, is not perfect, for dry land emerges from the water, and on the land all four elements interact. It is this interaction of the elements which makes living creatures possible, and without it the universe would be sterile.⁸

The concentric spheres which make up the universe, from Jodocus Trutfetter, A Textbook of Natural Philosophy (Summa in tota[m] physicen), 1514. Within the sublunary zone there are four distinct spheres: earth, water, air and fire; outside them are the planets, including the sun and the moon. The zodiac of the fixed stars is the outermost visible sphere, with three invisible spheres beyond.

This account presented Muslim and Christian philosophers with a problem which had not worried their pagan predecessors: how did it come about that the four elements did not form perfect concentric circles?⁹ They seized on this partly because it enabled them to introduce into philosophy a creator God unknown to Aristotle and Ptolemy. According to Genesis, God had gathered the waters together on the third day of creation in order to make dry land. So a simple answer was that the existence of dry land was a miracle. Since the waters of the oceans were higher than the land (higher, it was regularly said, than the highest mountains; otherwise, you would not find water springing from the ground near mountain peaks),ii ii it was easy to conclude that the oceans were held back from flooding the land, as they had in Noah’s Flood, by divine Providence. The philosophers found such an answer unsatisfactory, even though something similar was to be found in Pliny’s Natural History,¹⁰ and sought a natural explanation. If the initial separation required divine intervention, how should one characterize the relationship between earth and water since the Flood?

The problem was a simple one, and the range of possible answers was limited. In the course of a 250-year period, all the possibilities were fully explored.¹¹

The waters have been displaced from their original position, and their sphere now has a centre other than the centre of the universe. This view implies that ships sail uphill as they sail out on the ocean (we still acknowledge this traditional view when we use the term ‘the high sea’ or ‘the high seas’). It was held by Sacrobosco (c.1195–c.1256), who wrote the standard textbook on astronomy used in medieval and Renaissance universities, and after him by Brunetto Latini (1220–94), Ristoro d’Arezzo (writing in 1282), Paul of Burgos (1351–1435) and Prosdocimo di Beldomandi (d.1428). In 1320 Dante took it to be the standard view (though his text, the Quaestio de aqua et terra, was unknown until it was first published in 1508).
The earth (as distinct from the sphere of water) is no longer a sphere; rather, as the result of the growth of a boss, or tumour, it has acquired an elongated, irregular shape, so that its centre of gravity (the point around which it would hang without moving if suspended) corresponds to the centre of the universe but its geometrical centre doesn’t. The boss is what makes dry land possible. This was the view of Giles of Rome (1243–1316), who calculated that the earth’s diameter must have been stretched to almost twice its original length, and of Dante. The problem with this view was that it meant abandoning the idea that the universe was created out of nested spheres – a high price to pay, and one that few were prepared to contemplate.

If it could be argued that the earth might not be a true sphere then, equally, the waters might not be. Some suggested that the waters are not truly spherical but rather oval in shape, with the result that the oceans are deeper at the poles; this was held by Francesco di Manfredonia (d.c.1490) to be a partial explanation for the appearance of dry land. The weakness of this argument, as Francesco must have realized, was that if the waters were ovoid then there should be a belt of dry land at the equator and nowhere else; consequently, this argument was insufficient on its own, as he was obliged to recognize.

The earth is still a sphere, but it is no longer at the centre of the universe. This was the view of Robertus Anglicus (1271), but it was bound to have few supporters, as it ran against a core principle of Aristotelian philosophy: that the proper place for the earth was the centre of the universe. This difficulty, however, simply provoked the philosophers to think harder. Suppose, they argued, the earth is a sphere but its composition is not homogeneous: the action of the sun has made dry land less dense than it originally was, shifting the centre of gravity of the whole mass. Thus the centre of gravity of the earth still coincides with the centre of the universe, but its geometrical centre does not. Water, on the other hand, remains symmetrically arranged about the centre of the universe. This was the view of the fourteenth-century Parisian philosophers, of John of Jandun (1286–1328), Jean Buridan (c.1300–c.1358), Nicholas Bonet (d.1360), Nicholas Oresme (c.1320–82) and Albert of Saxony (c. 1320–90).¹² It preserved the system of nested spheres, and it had the great advantage of making water always flow downhill (which it does not in the first option above). As a modification of this view, one could argue that the centre of the universe corresponds to the centre of gravity of the aggregate of the two spheres of earth and water. This was the view of Pierre d’Ailly (1351–1420), despite the fact that he had read Ptolemy’s Geography, copies of which began to circulate in the Latin West around 1400. By 1475, in one or another variation, this was the standard belief.

These four views standardly took it for granted that the sphere of water was considerably larger than the sphere of earth. The conventional idea from 1200 to 1500 (mistakenly attributed to Aristotle) was that it was ten times larger, as it was held that each element exists in the same quantity, but a quantity of water occupies ten times the volume of the same quantity of earth while air occupies ten times the volume of water, and fire occupies ten times the volume of air.¹³ The relative size of the spheres and the extent of their displacement with regard to each other determines the size of the zone of dry land. It was commonplace to regard this as approximately one quarter of the earth/water globe, but it could extend to up to half of the earth/water globe. The first view assumed that the known world was all that there was to be known; the second implied that there was more land as yet undiscovered. This was usually held to lie in the southern hemisphere, and was sometimes thought to be inhabited.

The spheres of earth, water, air and fire from Sacrobosco’s Sphere (Sphaera mundi: Joannis de Sacro Busto sphæricum opusculum), Venice, 1501. The earth floats like an apple in a bucket. The orientation is not north/south; rather, Jerusalem, the centre of the known world, is at the top.

The distinct centres of the spheres of water (centred on A) and earth (centred on B), from Sacrobosco’s Sphere (Sphera volgare novamente tradotta), Venice, 1537. The two are marked as having relative volumes of 10:1, although, as Copernicus shows, were this to be the case, the sphere of land would not overlap with the centre of the sphere of water, which is here taken to be the centre of the universe.

The relative and absolute volumes of earth and water, again from Sphera volgare, 1537. Copernicus would have complained that the two spheres were not in fact drawn to the same scale.

It was generally acknowledged that there was a limited range of possible causes of a change in the relationship between earth and water. Either God had acted directly, heaping up and concentrating the waters in order to clear space for dry land, or the sun had acted on the earth to dry it out, or the stars had acted to pull either the waters or the earth out of position.

But, lastly, we come to the fifth view: there are no separate spheres of earth and water, there is less water than earth, and the oceans lie in the concavities of the earth, so that earth and water make up a single aggregate sphere. This, which is the modern conception (although of course we no longer think of ‘earth’ as one of the four elements), was held by Robert Grosseteste (c.1175–1253), Andalò di Negro (1260–1334), Themo Judaei (mid-fourteenth century) and Marsilius of Inghen (1340–96). Of these four, the opinions of Robert Grosseteste and Marsilius of Inghen were made available in print to Renaissance readers (though Marsilius was read by philosophers, not astronomers); but knowledge of the existence of this notion will have been reasonably widespread throughout the fifteenth century, for others described it, though only to reject it. It implies that land could be – indeed, should be – scattered across the whole surface of the Earth, a view endorsed by Roger Bacon (1214–94), probably under the influence of Grosseteste, and by the author of The Travels of Sir John Mandeville (c.1360).¹⁴ Of all the views, this is the only one straightforwardly compatible with the existence of antipodes (that is, bodies of land directly opposite each other on the globe).

It is essential to stress that this last belief found no support in the fifteenth century. For astronomers and geographers in 1475 (the year in which Ptolemy’s Geography was first printed, though the first Latin manuscript translation was in 1406), the basic choice was between an account of the sphere of water as displaced from the centre of the universe and an account of the sphere of the element earth as displaced from the centre of the universe (but still overlapping with it). To support Columbus’s voyage you did not have to think that these theories were wrong; you simply had to agree that going west might nevertheless be a quicker route to the Indies than circumnavigating Africa or going overland. After the discovery of a new continent, however, the outmoded view of Grosseteste once again became respectable among the philosophers.

Thus in 1475 there was general agreement that the centre of the two spheres of earth and of water were no longer identical, and indeed there was now a puzzle about three other centres: Where was the geometrical centre of the universe? Did it correspond to the centre of one of the spheres and, if so, which one? And if the earth was not homogeneous, where was its centre of gravity? Finally, where was the centre of gravity of the conjoined spheres of earth and water? Where the Aristotelian universe had one centre, there were now potentially five different ways of defining the centre of the universe.

§ 2

Late-medieval and Renaissance students learnt their astronomy by studying the Sphaera, or Sphere (c.1220) of Johannes de Sacrobosco, who taught in Paris but may have been English (in which case his name was presumably originally John of Holywood).¹⁵ His textbook was first printed in 1472 and went through more than two hundred editions.¹⁶ In addition there were numerous commentators who sought to explicate the text, beginning with Michael Scot (c.1230), including Giambattista Capuano da Manfredonia (c.1475),iii iii and culminating with Christoph Clavius (1570), the leading Jesuit astronomer of the late sixteenth century. The Sphere was still the standard text from which Galileo lectured when he was a professor at the University of Padua (1592–1610); the last edition for students, in 1633, conveniently marks the demise of Ptolemaic astronomy as a living tradition. In line with the notion that the globe was made up of two non-concentric spheres, one of earth and one of water, and following the example of Ptolemy’s Almagest (which had been available in the Latin West from the twelfth century), Sacrobosco proved separately that the surface of the earth was curved (he showed how this could be made apparent to someone travelling either north–south or east–west), and that the surface of the water was curved. (This was evident because a lookout on top of a ship’s mast could see further than someone standing on the deck.) Modern commentators assume that Sacrobosco had proved that the Earth is round;¹⁷ he had done nothing of the sort, and the medieval commentators did not claim that he had, for neither he nor they believed that the two spheres shared a common centre.

It should now be apparent that when medieval philosophers talked of ‘the earth’ they normally meant the sphere of the element earth which, where it showed above the ocean, constituted dry land; this sphere floated in an ocean of ocean, itself a larger sphere. The term ‘the earth’ was, however, inherently ambiguous. We find John of Wallingford (d.1258), for example, distinguishing in the space of two sentences between a) the earth, meaning dry land; b) the earth, meaning the element earth, whose centre is the centre of the universe; and c) the whole globe, i.e. the agglomeration of earth and water.¹⁸ The third usage (which looked back to Cicero’s Dream of Scipio) was distinctly unphilosophical for anyone who accepted the dominant two-spheres theory, so unphilosophical that it is difficult to find examples of terra being used in this sense in the later Middle Ages or early Renaissance, except by Latinizing humanists such as Petrarch.¹⁹ To all intents and purposes, the notion that the earth/water assemblage was to be thought of as a single globe or sphere disappeared around 1400. Even before 1400 it had never been the dominant view. The earth/water agglomeration was no longer round.

All these late-medieval discussions took place within the context of a geographical knowledge which corresponded to that of the ancients. No one believed that the earth was flat (it consisted of a portion of a sphere), but the habitable earth could be represented fairly accurately on a flat surface. This habitable earth had a centre, which was generally taken to be Jerusalem. However, there was another centre: measuring from west to east, from the Fortunate Isles (the Canaries) to the Pillars of Hercules (which marked the limits beyond which it was impossible to travel), there existed a notional location on the equator called Arim, or Arin, believed to be 10 degrees east of Baghdad. For the Arabs, and for astronomers relying on Arabic sources, Arim represented the degree zero of longitude and latitude.²⁰ It was universally accepted that dry land was confined to one hemisphere, the rest being covered by ocean. Of the dry land, the furthest northern and southern parts were uninhabitable because they were too cold or too hot, and so the habitable portion of the earth represented approximately one half of the whole of the dry land, one sixth of the surface of the whole agglomeration of earth and water.

As Dante pointed out in 1320, there was an obvious problem here, for the arguments of the philosophers and the maps of the geographers did not match up. If the philosophers were right and the habitable earth was a sphere floating on the surface of a larger globe of water, then a map should show the habitable earth as a circle. In fact, maps showed it as shaped like a cloak spread out on the ground; but the known world was referred to as the orbis terrarum, the circle of lands, as if it had the required form. Dante, unlike the philosophers, took his geography seriously, but no philosopher could have found his abandonment of the fundamental principle that the universe was made out of spheres entirely satisfactory.

Where the Aristotelian, idealized scheme of concentric spheres was symmetrical on every axis, the medieval elaborations (with the exception of the fifth) were each symmetrical around one axis only. Moreover, this axis was not the north–south axis of the poles but an axis through Jerusalem and the geometrical centre of the universe. Had late-medieval philosophers tried to imagine (which, of course, very few of them did) an Earth spinning in space on a north–south axis, then many of them would have been sure that the centre of gravity of the Earth (of either the sphere of earth or the sphere of water) was not on the north–south axis; such a spinning globe would have a natural tendency to wobble. The exception was the Parisian philosophers, for whom the centre of gravity of both the earth and the water remained coincident with the centre of the universe. Entirely logically, the only medieval philosopher of significance to take seriously the theory of the diurnal rotation of the Earth was a Parisian, Nicholas of Oresme (1320–82). Crucially, Oresme, unlike other philosophers who accepted (as he did) that there were two spheres of earth and water with separate geometrical centres, did not accept that the sphere of water was in itself bigger than the sphere of earth. He claims that if the two spheres had the same centre, water would inevitably cover the whole surface of the earth – except perhaps for a few mountain tops. And he describes the sphere of water as like a cloak or hood covering the earth. The result is that he has, as the illustrations accompanying his Livre du ciel et du monde (1377) show, a conception of the Earth as in effect a single globe, capable of rotating on its axis (but, since it is wrapped in the sphere of water, incapable of having antipodes).iv iv As it happens, Oresme’s text was never published, and cannot have circulated widely because it was written in French.²¹

Thus the two-spheres theory of the world was shared by nearly all philosophers, astronomers and cartographers (despite the difficulties it was known to present) until the late fifteenth century, and the rediscovery of Ptolemy’s Geography was integrated into it without too much difficulty.²² The Portuguese explorers reached the equator in 1474/5 (it is not difficult to tell when you have reached the equator: the Pole Star disappears from view), discovering a new heaven and new stars, but they found no uninhabitable zone: this required some minor rethinking, but little more.²³ It was true that Ptolemy in the Geography (unlike the Almagest) treated earth and water as a single sphere, and this was obviously bound to be of interest. After the translation of Ptolemy’s Geography there is a record of a terrestrial globe being made in 1443 ‘according to Ptolemy’s description’.²⁴ Columbus read Ptolemy and was convinced that earth and water formed one sphere; he produced a small globe to illustrate his planned voyage. At the same time he chose to reject Ptolemy’s account of the extent of the habitable world, preferring that of Marinus of Tyre (c.100–150), who had claimed that it extended more than halfway around the globe – a view difficult to reconcile with the two-spheres theory. But there was as yet no general crisis for the two-spheres theory: the geographers summoned by Ferdinand and Isabella to advise on Columbus’s plans had no hesitation in dismissing them out of hand.²⁵

The map of the world from Ptolemy’s Geography, printed in Rome in 1490. The same plates had previously been used in two earlier editions (Bologna, 1477; Rome, 1478), and they are consequently the earliest printed illustrations for the Geography.

That crisis began with Columbus’s landfall in 1492. In 1493 Peter Martyr described Columbus as returning from ‘the Western Antipodes’. In a notarial certificate drawn up by Valentim Fernandes in 1503 Pedro Álvares Cabral’s discovery of Brazil in 1500 is described as the discovery of ‘the land of the Antipodes’.²⁶ (He was right: Brazil is antipodal to the eastern extremity of the world known to the ancients.) But the decisive event was the publication in 1503 of the first letter written (or supposedly written) by Vespucci, entitled Mundus novus, which went through twenty-nine editions in the space of four years.²⁷ (It was Vespucci’s second letter which introduced the word ‘discovery’ to a European audience; his first had already destroyed the medieval cosmography.) Vespucci’s claim was that he had encountered a vast new landmass which formed no part of the previously known world – he had found a New World. Moreover, it was clear that this land mass, although it was only one quarter of the way around the globe from his starting point, was halfway around the globe from other parts of the known world. And Vespucci had sailed 50 degrees south of the equator: this was not just the equatorial antipodes that some exponents of the two-spheres theory had envisaged. Antipodes had become a reality, and there was no longer any way of fitting the Earth’s land masses into one hemisphere.

Thus what was disturbing about these antipodes was not that they implied that some people were ‘upside down’ compared to other people – you had to be fairly unsophisticated to have difficulties with this idea – but that the two-spheres theory could accommodate antipodes only as a limit case, along the boundary between the northern and southern hemispheres, and only then if the sphere of water was shrunk so that its diameter was almost the same as that of the sphere of earth.²⁸ Vespucci’s claim required a major reconsideration of the supposed relationship between the elements of water and earth. Up to this moment it had been possible to believe both that the spheres of the earth and of the ocean were round, and that the zone of dry land (the orbis terrarum, the habitable world) had, as the Bible put it, four corners.²⁹ Now these corners became, in John Donne’s phrase, ‘the round earth’s imagin’d corners’.³⁰

Clavius’s representation, in his commentary on Sacrobosco (1570; but taken here from the revised edition of 1581) of the standard account of the relationship between water and earth, which he rejects. The dots mark the two geometrical centres, that of the sphere of water (below) and that of the sphere of earth (above). Since discussion of whether there was one sphere of earth/water or two spheres was inseparable from discussion of whether there were antipodes (which could not exist on the two-spheres model, except perhaps in a brief band where the two spheres met if they were of similar size), Clavius’s illustration also includes (non-existent) antipodes, which are underwater. Since the antipodes are known to exist, this traditional model must be wrong.

The first people really to come to grips with this were Martin Waldseemüller and Matthias Ringmann, as they worked on their world map of 1507 and the accompanying Introduction to Cosmography.v v Struggling to think through the implications of Vespucci’s claim, they needed a way of referring to what we call the Earth, or the world – the single globe formed of land and sea. They called it omnem terrae ambitum, the whole circumference of the Earth, of which, they explained, Ptolemy knew only a quarter.

Other early world maps present themselves as illustrations of the orbis terrarum. In classical Latin, from which the phrase derives, an orbis is usually a flat disc, but sometimes it is an orb or globe. When Cicero writes of the orbis he sometimes means the habitable dry land, a disk rising above the waves, and sometimes the whole globe of land and ocean. This ambiguity was carried through into the Renaissance. Thus Ortelius’s 1570 atlas was entitled Theatrum orbis terrarum, the theatre of the sphere of lands. The frontispiece makes clear that the orbis is a globe, but the plural terrae implies a collection of maps of different countries. Mercator, exceptionally, used the phrase orbis terrae – in 1569 the word terra is beginning to mean Earth, or world (as in Planet Earth); one word has been substituted for Waldseemüller and Ringmann’s clumsy phrase. By 1606 Ortelius’s Theatrum could be translated into English as The Theatre of the Whole World. Only later, in 1629, was a satisfactory technical term invented to identify unambiguously this new entity: it was called ‘the terraqueous globe’.³¹

We can trace in detail the progress of this new concept in the years after the publication of Waldseemüller’s and Ringmann’s Introduction to Cosmography in 1507. The first sign of change is to be found in a physics textbook published in Erfurt in 1514. The author, Jodocus Trutfetter, presents the one-sphere theory first, although he then goes on to explain the view that the sea is higher than the land; he notes that the most recent cosmographers have claimed that there are inhabited antipodes at the eastern and western extremes of the world, although he balances this by explaining that Augustine had denied the possibility of antipodes. If the text is cautious, the accompanying illustration is not: it shows only three sublunar spheres, of earth, air and fire. Evidently, earth and water now make one sphere.vi vi ³²

In 1515 Joachim Vadianus, a man of many talents (he was the Poet Laureate of the Habsburg empire), published in Vienna a little pamphlet, Habes lector, or Dear Reader (reprinted half a dozen times), in which he suggested, in the light of the discovery of America, that, contrary to the standard interpretation of Aristotle, habitable land was scattered almost randomly across the surface of the globe, and that earth and water were so intermingled as to form one sphere.³³ The geometrical centre of the globe and its centre of gravity were, he asserted, one and the same. As for Augustine’s fear that to admit the existence of antipodes would be to acknowledge that there were human beings who were not descended from Adam, he had a simple answer: one could travel overland from Spain to India, almost halfway around the globe, and there was no reason to think that any inhabited land was set at a vast distance from the rest (the implication being that America was close to Asia). Three years later, again in Vienna, George Tannstetter (also known as George Collimitius), who was in close collaboration with Vadianus, published an edition of Sacrobosco’s Sphere which contains the first illustration of the ‘modern’ conception of the globe as made up of interlocking land and sea.³⁴

The first sophisticated illustration of the earth and water as making a single sphere where the two elements interlock: from Joannes de Sacro Bosco, Opusculum de sphaera (1518), edited by Tanstetter. There are now three sublunary spheres, not four.

Clavius’s representation, from his commentary on Sacrobosco (1570, here from the 1581 edition) of the relationship between earth, water, air and fire. Earth and water make one sphere, surrounded by three levels of the atmosphere (the weather being generated in the middle level) – only the outermost of these levels is a perfect sphere, beyond which is the sphere of fire.

In 1531 Jacob Ziegler published in Basle an elaborate commentary on Book II of Pliny’s Natural History. In it he interpreted Pliny’s account of how the waters are higher than the earth in terms of the medieval two-spheres theory, only to conclude, bluntly, that modern discoveries had shown this view of the globe to be fallacious, as land was not confined to only one hemisphere of the globe.³⁵ In the same year as Ziegler’s book, there appeared at Wittenberg an edition of Sacrobosco with an introduction by the leading Lutheran theologian and educator, Melanchthon.³⁶ Melanchthon’s introduction praised astronomy as the study of God’s handiwork, but it also went on to provide an elaborate defence of astrology. This edition was repeatedly reprinted, and widely pirated (in Catholic countries the introduction was often printed without the name of the author, since all texts written by Protestants were banned; in earlier copies Melanchthon’s name is often blotted out on the title page). A crucial new illustration showing the earth/water globe was copied from an edition of the Sphere produced by Peter Apian in 1526, and, through the influence of the Wittenberg edition, it became the new standard; it was even copied in the much-reprinted commentary on Sacrobosco produced by Christoph Clavius, the first edition of which appeared in 1570.³⁷

In 1538 the Wittenberg presses produced a new, elaborate version of the Melanchthon edition which included ‘volvelles’, paper instruments or illustrations with circular moving parts.³⁸ In this edition (which also went on to be frequently reprinted and copied) the conventional chapter headings into which Sacrobosco’s text had been divided were revised. Where earlier editions had had a chapter proving the earth was a sphere, and another proving that the waters were a sphere, this new edition presented a whole section as being about water and earth making up one globe. The text itself had not been changed (as it was, for example, in an edition for use in schools which appeared in Leiden in 1639), but the new heading, Terram cum aqua globum constituere, transformed its meaning.³⁹ From 1538 the new understanding of earth and water as making up a single sphere became an orthodoxy among both Protestant and Catholic astronomers.

Peter Apian’s new illustration to show that the earth is round, later copied by Melanchthon and Clavius, from Sacrobosco, Sphaera … per Petrum Apianum … recognita ac emendata (1526).

In 1475 the two-spheres theory of the world was universally held by philosophers and astronomers; by 1550 every expert had abandoned it.⁴⁰ That did not mean, however, that certain aspects of the old theory could not be preserved within the new. One might think that the adoption of the theory of the terraqueous globe automatically meant acknowledging that the seas are lower than the dry land, but the contrary view seemed to be clearly established by scripture and by innumerable respectable authorities. So the Jesuit Mario Bettini (1582–1657) argued that when God had turned the separate spheres of earth and water into one sphere by opening up cavities in the earth to absorb the bulk of the water, it had been necessary to compensate for the fact that (since water is, by definition, lighter than earth) the centre of gravity of the new terraqueous globe was in danger of not coinciding with the centre of the universe; consequently, the waters bulged outwards so that their weight was equal to that of the earth they had displaced. Gaspar Schott (1608–66, also a Jesuit) accepted this argument as the explanation for the origin of most rivers. Their headwaters, he thought (as this illustration seeks to show), lie below the highest point of the sea (high sea level: F), but above the shoreline (low sea level: BC). It was, he held, an open question whether there were rivers that originated at a point above high sea level (E). Thus the doctrine that the seas are higher than the land survived well into the second half of the seventeenth century.vii vii ⁴¹ Obviously, the notion that the height of a mountain could be measured from sea level could establish itself only after this view had been abandoned. Still, this was not the old two-spheres theory, and it was now axiomatic that earth and water had a single centre, which was both the geometrical and the gravitational centre of the globe. I can find only two people who, after the publication of Waldseemüller’s map, sought to defend the old theory against its attackers: the new reality was incompatible with the old theories.

One morning in August 1578 a debate broke out at the breakfast table of the Duke of Savoy, Emanuele Filiberto, as to why rivers run to the sea. An Averroist philosopher who was present, Antonio Berga, insisted that, as the sea was higher than the land, it could not simply be because water naturally flows downhill. Berga went on to appeal to the old orthodoxies: the sphere of water is ten times greater than the sphere of earth, the two spheres do not have the same geometrical centre, and the oceans are higher than the land. Berga’s views were disputed by Giovanni Battista Benedetti, who was officially the duke’s mathematician and philosopher, and, since the honour of both men was now at stake, the dispute rumbled on after the meal was over. Benedetti told Berga to read Piccolomini, and he put some of his own arguments on paper for the duke to read; Berga published a refutation of Piccolomini, and implicitly of Benedetti; and Benedetti responded, cruelly mocking Berga (who showed his lack of expertise by confusing the Antarctic and the Arctic) and calling him ‘half Huguenot’ in his philosophy (this was tit-for-tat, as Berga had dismissed the new theories as philosophical heresies).⁴² Berga, it must be stressed, made no attempt to claim that he had support among contemporary philosophers for his antiquated views: if there were others who thought as he did they were too sensible to entrust their arguments to print. For to preserve the old orthodoxy it would have been necessary to insist that the world’s land masses were confined to one hemisphere.⁴³ Berga side-stepped this issue and, as far as I can tell, only one person was so foolish as to explicitly present this argument.viii viii

Schott’s illustration from Anatomia physico-hydrostatica fontium ac fluminum (1663) to show how the surface of the ocean curves upwards and how water from the ocean travels underground through fissures in the earth to emerge as springs and rivers. The fact that the ocean is higher than the land explains why water can spring out of the ground above the level of the shore, although Schott acknowledges that the relative heights of the tops of mountains and the ocean have not been established.

Still, one would have expected there to be a range of alternative theories proposed to account for the new evidence. One could, for example, argue that far from there being one earthly sphere floating in the ocean it was now apparent that there were two. This view was expressed by those (echoed by Copernicus) who described the New World as altera orbis terrarum, another sphere (or circle) of land masses. It was put forward in all seriousness in 1535 by Oviedo (Gonzalo Fernández de Oviedo y Valdés), writing the official Spanish history of the discovery of the New World.⁴⁴ But for Copernicus this was merely a turn of phrase, for it was evident that you could not have two spheres of earth and at the same time place the element earth at the centre of the universe. A universe in which there were two earths within one sphere of water was no longer an Aristotelian universe. Altera orbis terrarum was a catchy phrase which could not be turned into a viable theory. So the two-spheres theory was abandoned, even while efforts were made by some conservative thinkers to preserve the traditional claim that the seas are higher than the land.

One author, however, was not so easily defeated. In his Universae naturae theatrum of 1596, Jean Bodin argued that the new continents were simply vast plates floating on a bottomless ocean. He held that the element earth is heavier than the element water, but that (following Aristotelian orthodoxy) heavier objects can, if they have the correct shape, float on lighter objects. The floating continents would displace their own weight in the water (according to Archimedes’ principle) but, in a striking non sequitur, only one seventh of their bulk would be below the waves. To make matters worse, Bodin clung to the traditional belief that the ocean bulges up above the land, higher than the highest mountain tops, although it was hardly compatible with his account of the continents as floating high above the waves. Bodin was sure that one could have floating land masses; he believed there were reliable reports of islands that sneakily changed their position during the night – but the big continents, he thought, remained in one place. Thus Bodin proposed, not a terraqueous globe but an aquaterreous one in which (as one annotator summed up his thesis in the margin of the text) terram aquis supernatare, the earth floats on the surface of the waters.⁴⁵

Bodin’s motives for this strange argument are complex. In the first place, he was clear that land was not confined to one hemisphere, so the old two-spheres theory would not do. Secondly, he had read in Copernicus a demonstration that if the earth were one tenth the size of the waters it would be entirely immersed if any part of it overlapped with the centre of the sphere of water. So he decided that the only solution, if one wanted to retain the right ratio between water and land, was to break up the land and scatter it across the surface of the waters. In doing so, he completely abandoned two principles which had been fundamental to Aristotle: that the element earth is a sphere, and that the element earth is at the centre of the universe. Yet he came closer, he believed, to the Old Testament account of creation.

So peculiar was Bodin’s theory that Gaspar Schott, writing two generations later, simply could not understand it.⁴⁶ He interpreted Bodin, quite wrongly, as advocating a very large sphere of earth floating in a sphere of water, thus retaining the core principles of a traditional Aristotelian argument. He drew an elaborate diagram to explicate what he took to be Bodin’s theory, although his drawing is quite unlike Bodin’s own. Schott’s complete incomprehension suggests that it would have been difficult for Bodin to persuade other scholars that his views made sense. Anyone who closely examined them would have been forced to conclude that his account of how heavier-than-water bodies might float was riddled with inconsistencies because Archimedes and Aristotle were simply incompatible, and it is very difficult to see how a stable theory could have been generated which was based on Bodin’s concept of floating continents.

What are we to make, then, of the peculiar story of the almost silent demise of the two-spheres theory? There had been good evidence against it long before Vespucci reached the New World. Giles of Rome and Dante had pointed out that if the theory was correct the land emerging from the waters should have a circular shape, and it did not. Dante, entirely sensibly, said that one should establish whether something was the case (an sit) before determining why it was the case (propter quid); in his view, the evidence falsified the two-spheres theory, even if that theory was an elegant reinterpretation of Aristotle.ix ix Moreover, the early, embattled proponents of what would later be known as the terraqueous-globe theory, Andalò di Negro and Themo Judaei, had pointed to the circular shape of the shadow of the earth during eclipses of the moon (a phenomenon already known to Aristotle) as proof that there was only one terraqueous sphere, not two overlapping spheres. Water, they insisted, was not simply transparent: a sphere of water would cast a shadow, and no such shadow was to be seen.⁴⁷ Copernicus recycled this argument in On the Revolutions (1543).

Jean Bodin’s illustration to show his new theory of the relationship between earth and water, from the Universae naturae theatrum (1596). The middle image shows the standard late-medieval conception of a sphere of earth one-tenth the size of the sphere of water; the top image shows that such a sphere of earth will not overlap with the centre of the universe; and the bottom image shows Bodin’s own conception of a series of flat plates of earth floating on the oceans.

Schott’s version of Bodin’s new theory of the relationship between land and water, from his Anatomia physico-hydrostatica (1663).

In the fourteenth century, evidence, good evidence, against the two-spheres theory had been presented, and it had been brushed aside. In the early sixteenth century the voyages of Vespucci provided further evidence against that theory, and it was decisive. Was the quality of the evidence different? It was. There are two important features to Vespucci’s voyages (for all that modern scholars debate how many voyages he made, and whether he wrote the accounts of his voyages that were published in his name). First, there was no disputing the importance of the discoveries in the New World, for the simple reason that they had become matters of state, the concern of kings. How could scholars ignore what governments took seriously? Second, and even more importantly, these discoveries were new. When Andalò di Negro invoked the shadow of the Earth as seen in eclipses of the moon, or Dante invoked the shape of dry land in the known world, they were appealing to information that had long been available. It was easy to assume that these arguments had already been taken into account, somehow, somewhere, by the advocates of the two-spheres theory, for in a manuscript culture no one can hope to have every relevant text to hand. But it was evident that Vespucci’s information was quite simply unprecedented: it needed to be addressed here and now.

The invention of discovery, acting in combination with the printing press, transformed the balance between evidence and theory, tilting it away from the reinterpretation of old arguments and towards the acquisition and interpretation of new evidence. As far as the two-spheres theory was concerned, the voyages of Vespucci were deadly. The new facts were killer facts. As it happens, this is the first occasion since the establishment of universities in the thirteenth century on which a philosophical theory was destroyed by a fact.x x Astonishing as it may seem, there is no previous occasion on which new empirical evidence determined the outcome of a long-standing debate between philosophers. Aristotle, for example, had argued that the nerves are all connected to the heart; Galen had shown they were connected to the brain; but Aristotelian philosophers, both ancient and medieval, had continued to follow Aristotle’s teaching, as if Galen did not exist.xi xi In 1507 the relationship between theory and evidence changed, and changed for ever.

§ 3

In 1543 Copernicus published On the Revolutions, in which he argued that, far from standing still at the centre of the universe, the Earth orbits the sun once a year and turns every twenty-four hours on its axis.⁴⁸ Copernicus was a canon of the Cathedral of Warmia in Polish Prussia and had studied extensively in Italy (astronomy at Bologna and medicine at Padua). He begins his great work by running through a set of conventional arguments drawn from Sacrobosco: the heavens are spherical; the earth is spherical; the waters are spherical. In the last sentence of Book 1, Chapter 2, Copernicus rejects the argument (taken from Pliny and the Bible) that the waters are higher than the land. Then in Chapter 3 he stresses the importance of the discovery of America: earth and water make one globe in which the centre of gravity and the geometrical centre coincide. The waters cannot be, as many medieval philosophers had claimed, ten times as extensive as the earth, for if they were, and the earth is round and rises above the surface of the water, then simple geometry shows that no part of the earth will coincide with the centre of the universe. There really are antipodes and antichthones; ‘Indeed geometrical reasoning about the location of America compels us to believe that it is diametrically opposite the Ganges district of India’ (a calculation rather different from that made by Vadianus, who had placed India and Africa as each other’s antipodes). Thus Copernicus argued for a spherical Earth – appealing to the evidence of the shape of the Earth’s shadow cast on the moon during eclipses to confirm that the Earth was to all intents and purposes a perfect sphere, the occasional mountain and valley notwithstanding – a crucial first step towards arguing that the Earth rotates on a north–south axis.

By 1543 the broad outline of Copernicus’s argument for the Earth as a single globe was conventional. But we know that Copernicus had first formulated his views by 1514, for at that date at least one copy of his preliminary sketch, the Commentariolus, or Little Commentary, was in existence.⁴⁹ He gives us two accounts of the development of his thinking, one at the beginning of the Little Commentary and the other at the beginning of On the Revolutions. From them we learn that he had long been dissatisfied with conventional astronomical theories, that he had engaged in a systematic programme of reading in an attempt to identify alternatives, that the idea that the Earth moved had at first seemed to him absurd but that he had persisted with it, determined to see if it could provide the basis for a new account of the movements of the heavens.

Those few commentators who have grasped that Copernicus’s doctrine that the earth and the seas form one sphere was relatively new have quite correctly concluded that there was a fundamental obstacle which Copernicus had to overcome before he could envisage a rotating Earth: he had to envisage the Earth as spherical (to push at the limits of possibility, as symmetrical on a north–south axis, or, as an absolute minimum, having its centre of gravity on its north–south axis).⁵⁰ Edward Rosen has argued that the geographical information in Book I, Chapter 3, of On the Revolutions (such as the claim that America is the antipodes of the Ganges) is based on Waldseemüller’s map of 1507, the book that accompanied it, and another map by John Ruysch published in the same year.⁵¹ If so, it seems Copernicus came to view the Earth as a spherical globe some time between 1507 and 1543. But when exactly?

Here we have no useful source of information other than the Little Commentary. It begins with a number of axioms. The second is ‘centrum terrae non esse centrum mundi, sed tantum gravitatis et orbis Lunaris’ (‘the centre of the earth is not the centre of the universe [because the sun, not the Earth, is at the centre of the universe], but only the centre of gravity and of the lunar sphere’). As we saw in Chapter 3, the late-medieval view was that the earth overlapped with the centre of the universe but that there were at least three relevant centres of gravity: the centre of the earth, towards which solid objects fell; the centre of the sphere of water, towards which water descended; and the centre of gravity (that is to say, the point of balance or equilibrium) of the two spheres. One of these three centres was held to be the centre of the universe. Centrum terrae esse centrum gravitatis simply cuts through this debate in the fewest possible words, rejecting the arguments of the Parisian School and demonstrating that Copernicus already subscribed by 1514 to the argument that Vadianus was to be the first to publish (in 1515), and which Copernicus was to repeat in On the Revolutions: the geometrical and gravitational centre of the Earth are one and the same.

Secondly, Copernicus describes the rotation of the Earth as follows: ‘Alius telluris motus est quotidianae revolutionis et hic sibi maxime proprius in polis suis secundum ordinem signorum hoc est ad orientem labilis, per quem totus mundus praecipiti voragine circumagi videtur, sic quidem terra cum circumfluis aqua et vicino aere volvitur.’ Rosen translates this as: ‘The second motion, which is peculiar to the earth, is the daily rotation on the poles in the order of the signs, this is, from west to east. On account of this rotation the entire universe appears to revolve with enormous speed. Thus does the earth rotate together with circumjacent waters and encircling atmosphere.’

We need to be a little more precise: terra cum circumfluis aqua et vicino aere volvitur means ‘the earth rotates, together with the water and the neighbouring air which flow around it.’⁵² On the traditional view (explicitly rejected by Copernicus in On the Revolutions), the earth floats like an apple in a larger sphere of water.⁵³ But here the water is compared to the neighbouring air – both lie on the surface of the land and flow around it and across it. Prefigured here, then, is the claim later made in On the Revolutions, that ‘finally, I think it is clear that land and water together press upon a single center of gravity; that the earth has no other center of magnitude; in that, since earth is heavier, its gaps are filled with water; and that consequently there is little water in comparison with land, even though more water perhaps appears on the surface.’

So if we look carefully at the text of the Little Commentary we find, in telegraphic form, what will become the argument of On the Revolutions.⁵⁴ Three conclusions follow from this. First, the Little Commentary cannot have been written before 1507. There is independent evidence to support this view, for in 1508 Lawrence Corvinus wrote a poem in which he implies that Copernicus believed at that time that the sun moved in the heavens; in other words, he had not yet adopted heliocentrism, even though he had already formulated ‘wonderful [new] principles’.⁵⁵ Second, Copernicus was one of the first since the fourteenth century to reject the two-spheres, several-centres theory of the earth, which helps to explain the emphasis he places upon this argument in On the Revolutions, despite the fact that by 1543 he was knocking on an open door. Indeed, other Copernicans must have found Copernicus’s emphasis on this point hard to understand, so quickly had it become uncontentious. Thomas Digges, when he translated the key parts of Book I into English, dropped the discussion of the roundness of the earth altogether, for he simply took it for granted that the Earth is a ‘ball of earth and water’.⁵⁶

With this chronology in mind, we can now address an important question: was Copernicus’s adoption of the terraqueous-globe theory the key event which led to his switch from geocentrism to heliocentrism? It has been suggested that Copernicus originally considered a geoheliocentric theory, that is to say a theory in which the sun goes round the earth and the planets go round the sun – the theory later advocated by Tycho Brahe.⁵⁷ I doubt this, because Copernicus seems to have assumed that the correct theory must already have been formulated: he needed to read until he found it. He was not looking for a brand-new theory; he did not yet have a conception of knowledge as progressive. Still, if Copernicus did consider geoheliocentrism, it seems clear he quickly abandoned it, presumably when he recognized that such a theory was incompatible with belief in physical spheres carrying the planets, as the orbit of Mars around the sun would intersect with that of the sun around the earth. As soon as he turned to consider a more radical theory, heliocentrism (more radical in that it involved a moving Earth, but more conservative in that it was compatible with belief in physical spheres, and in that it had already been formulated by ancient philosophers), he will have realized that he had to determine the shape of the earth/water aggregate, because his Earth had to be capable of rotating on its axis and flying through space.

Sacrobosco’s theory, that the waters had been displaced from the centre of the earth, would have had to be dismissed out of hand, for how could these waters turn evenly around the centre of the earth if that was not their centre? The Parisian view, that the centre of gravity of the earth corresponded to the centre of the sphere of water, will have seemed at first sight like a viable option. But Copernicus was a competent mathematician. He would quickly have realized, as he pointed out in On the Revolutions, that if, as was generally assumed, the sphere of water was ten times as large as the sphere of land, then the sphere of land would not overlap at all with the centre of the sphere of water, so the centres of gravity of the earth and of the water could not be made to coincide. Even if he shrank the sphere of water considerably, it would be difficult to get the centre of gravity of the sphere of earth to correspond with the centre of the sphere of water, unless one assumed that dry land was radically different from elementary earth – and a large part of the sphere of earth would have to be made up of theoretically ‘dry’ land, even though it was below the level of the waters. Pierre d’Ailly, and after him Gregor Reisch (1496), had tried to overcome this difficulty by treating earth and water as a single aggregate when identifying a centre of gravity which could coincide with the centre of the universe: the result was a theory which claimed that for some purposes ‘the Earth’ was to be thought of as consisting of two spheres; for other purposes it was to be thought of as consisting of one sphere.⁵⁸ Either way, there could be antipodes, but only along the margin between the two spheres.

A copy of the first edition of Copernicus (from Lehigh University), with a contemporary annotation. The reader is working out the logic of Copernicus’s claim that the traditional account of the relationship between earth and water is internally contradictory, as the volume of water cannot be ten times the volume of earth if the sphere of earth is to overlap with the centre of the sphere of water – which it must do if the earth is still to be at the centre of the universe, despite no longer having its own centre coincide with it. Exactly the same point caught the attention of Bodin in the Theatrum. (I am indebted to Noel Malcolm for painstakingly transcribing this nearly illegible annotation.)

Copernicus tells us that he engaged in a systematic programme of reading as he struggled with the formulation of his new astronomy.⁵⁹ Michael Shank has suggested that in the course of this reading Copernicus obtained a copy of the compendium of astronomical texts published by the Giunta press in Venice in 1508. There he would have found Grosseteste’s brief exposition of the one-sphere theory. But he would also have found a commentary on Sacrobosco by Giambattista Capuano (first published in 1499) which is the only pre-Copernican work to discuss how one might formulate an astronomical theory based on the concept of a moving earth.⁶⁰ Crucially, Capuano discusses not only the familiar idea (expounded by Oresme) that the earth rather than the heavens might rotate daily, but also the possibility that it might move through the heavens on an annual path comparable to the path normally assigned to the sun. If this text indeed fell into Copernicus’s hands (and Copernicus had been studying in Padua between 1501 and 1503, when Capuano was lecturing on astronomy, so he may have already heard his lectures, or read an earlier printed edition), then we can be sure that he read it with great care. Capuano formulated a series of objections to a moving earth which were to become classical – for example, if you throw an object straight upwards in a moving boat it will fall behind the boat.⁶¹ If the earth rotated, he argued, we would all simply be drowned, as every day our bit of earth would turn under the waves – as it must, according to the two-spheres theory. If one argued that earth, water and air all rotated together, so that they were all stationary with regard to each other, then why are there always violent winds blowing at the tops of mountains? Capuano believed these winds were caused by the movement of the spheres being transmitted to the upper atmosphere. Copernicus’s careful formulation in the Little Commentary, that the earth rotates together with the neighbouring air, seems almost designed to leave scope for a failure of the upper atmosphere to rotate along with the earth, thus providing an alternative explanation for the winds on mountain peaks. Reading Capuano would have left Copernicus in no doubt that he needed an account of what sort of body the Earth is, along with an account of what happens when objects fall on a moving Earth. (Copernicus’s account explains that falling objects move with the moving Earth, but he does not extend this to claim that a falling object on a moving ship would move with the ship.)

If we imagine that Copernicus had reached this point in his thinking soon after 1508, then the geographical discoveries of Amerigo Vespucci and the maps and commentaries of Waldseemüller and Ringmann will have been crucial for him in developing his heliocentric theory, for they provided a definitive solution to the problem of the Earth’s shape. It is evident from the text of On the Revolutions that the concept of the terraqueous globe was of fundamental importance to him; this was surely the last building block in the construction of the new theory.⁶² Without Vespucci there would have been no Copernicanism, for Copernicanism required a modern theory of the Earth.

Can we test the claim that Copernicanism required a modern theory of the Earth? At first, it would seem impossible: all we have to go on is two texts by Copernicus. But there are three other early presentations of the claim that the Earth moves: Copernicus’s disciple Rheticus’s First Narration (1540), the first appearance in print of an account of Copernicus’s theories; Celio Calcagnini’s short treatise arguing that the Earth rotates on its axis (pre-1541, and thus pre-Copernicus); and a text by Rheticus (1542/3) dealing with Biblical arguments against the motion of the Earth. Although these are all too late for there to be any need to demonstrate at length that the Earth is one sphere, we could expect to find the modern theory of the Earth clearly referred to in each of them when they discuss the movement of the Earth – and we do. Each of them thinks it necessary to stress that the Earth is a perfectly round ball, globe or sphere.⁶³

§ 4

What are the implications of claiming that the Earth is a planet? Copernicus did not discuss the question; but his successors were bound to. In the summer of 1583 a strange little Italian gave a series of lectures in Oxford.⁶⁴ We know him as Giordano Bruno, but he liked to invent long names and titles for himself, names, it was said, longer than his body. The opening words of this letter of his provoked laughter:

Philotheus Jordanus Brunus Nolanus, doctor of a more sophisticated theology, professor of a more pure and innocent wisdom, known to the best academies of Europe, a proven and honoured philosopher, a stranger only among barbarians and knaves, the awakener of sleeping spirits, the tamer of presumptuous and stubborn ignorance, who professes a general love of humanity in all his actions, who prefers as company neither Briton nor Italian, male nor female, bishop nor king, robe nor armour, friar nor layman, but only those whose conversation is more peaceable, more civil, more faithful, and more valuable, who respects not the anointed head, the signed forehead, the washed hands, or the circumcised penis, but rather the spirit and culture of mind (which can be read in the face of a real person); whom the propagators of stupidity and the small-time hypocrites detest, whom the sober and studious love, and whom the most noble minds acclaim, to the most excellent and illustrious vice-chancellor of the University of Oxford, many greetings.⁶⁵

When he walked to the lectern he rolled up his sleeves, as if he were a juggler about to perform a trick. As he spoke he bobbed and dipped like a dabchick or little grebe. He lectured, as all academics did, in Latin, but he spoke Latin with a Neapolitan pronunciation; the dons of Oxford (who found their own English pronunciation of Latin civilized and sophisticated) laughed at him for saying chentrum, chirculus and circumferenchia (which, as it happens, is now the approved pronunciation). But mostly they took exception to his Copernicanism. Twenty years later, George Abbott, who would eventually become Archbishop of Canterbury, remembered it as if it were yesterday: ‘he undertooke among very many other matters to set on foote the opinion of Copernicus, that the earth did goe round, and the heavens did stand still; wheras in truth it was his owne head which rather did run round, & his braines did not stand stil.’⁶⁶

It was forty years since Copernicus had published On the Revolutions. His new astronomy had certain evident advantages over the established astronomy of Ptolemy. According to Plato and Aristotle, all movement in the heavens should be circular and unchanging and, as we have seen, in the Renaissance there were still philosophers (such as Girolamo Fracastoro (1477–1553), the first to think seriously about contagious diseases) trying to construct a simple model of the universe which consisted of spheres nested around a common centre. But, try as they might, the philosophers could not get such models to fit what actually happens in the heavens. What Ptolemy had managed to achieve was a system that accurately predicted movements in the heavens. The Ptolemaic system, like those of Plato and Aristotle, claimed that the moon, the sun and all the planets circled around the earth. But in order to predict accurately the movement of these heavenly bodies it employed a complex system of deferents (circles), epicycles (circles on circles), eccentrics (circles rotating around a displaced centre) and equants. The equant was a device for speeding up and slowing down the movement of a body in the heavens by measuring its movement not from the centre of a circle but from another point. By this means the movement could be described (or misdescribed) as constant; it was thus a method of cheating on the fundamental principle insisted on by the philosophers that heavenly movement should be circular and unchanging. (For strict Aristotelians, even the epicycle was a cheat, as they wanted all the circular movements to have a common centre.)

Copernicus proposed to abolish the equant, and to eliminate an epicycle for each planet further from the sun than the Earth by showing how the movement of the Earth created an apparent movement in the sky equivalent to an epicycle. Copernicus also claimed that his system was preferable because it specified more tightly the characteristics of the system as a whole. Ptolemaic philosophers had never been sure, for example, whether Venus or the sun was closer to the earth (the right answer, in our terms, being that sometimes it is one, and sometimes the other, but this was an unacceptable answer within the Ptolemaic system), while Copernicus’s system placed the heavenly bodies in a fixed order.⁶⁷

It used to be thought that Copernicus initiated an intellectual revolution – indeed Thomas Kuhn called his first book The Copernican Revolution (1957). But in this Kuhn was mistaken. Throughout Europe astronomers took a keen interest in what Copernicus had to say, but, with only a very few exceptions, they took it for granted that his account of a moving Earth was simply wrong. If the earth moved, we would be aware of it; you would feel the wind in your face. If you dropped an object from a tall tower, it would fall towards the west. If you fired a cannon to the west, the ball would go further than if you fired it to the east. Since none of these things happened, all the leading astronomers – Erasmus Reinhold (1511–53), Michael Maestlin (1550–1631), Tycho Brahe (1546–1601), Christoph Clavius (1538–1612) and Giovanni Magini (1555–1617) – were confident that Copernicus was wrong. Still, they were fascinated by the simplicity of his techniques for calculation, and thrilled at the idea that it might be possible to junk the equant. In an extraordinary labour of love, every surviving copy of the first (1543) and the second (1566) editions of On the Revolutions has now been studied to identify the marginal comments written by its first readers, with the result that we can tell very reliably what they liked and what they disliked, what they found credible and what they found incredible.⁶⁸ They liked Copernicanism as a mathematical device; they had no time for it as scientific truth. They read it as the prefatory letter (now known to have been written by Osiander, and added without Copernicus’s permission) encouraged them to read it, as a purely hypothetical construction.

In 1583 there were, as far as we know, only three competent astronomers in the whole of Europe who accepted Copernicus’s claim that the Earth travelled around the sun: in Germany, Christoph Rothmann (who did not publish, and eventually abandoned Copernicanism); in Italy, Giovanni Benedetti (who published a few sentences on the question in 1585); and, in England, Thomas Digges (who had published in support of Copernicanism in 1576).xii xii So it must have simply astonished the dons of Oxford to hear this peculiar Italian, as he dipped and dodged, chucked and chirred, defending Copernicanism as the literal truth.

We do not know how far Bruno got in his exposition of Copernicanism. He was stopped after he had given three lectures; he was accused of merely reciting passages from the Renaissance Platonist philosopher Ficino (who had written in praise of the sun), while giving the impression that the words were his own. This is quite possible – Bruno does similar things in his published texts and, as we have seen, the concept of plagiarism was a novel one.xiii xiii But we know what Bruno wanted to say because, after he was driven out of Oxford, he took refuge with the French ambassador in London, and there he set about writing a series of works, of which the most famous is The Ash Wednesday Supper, in defence of his position.⁶⁹ In the course of eighteen months Bruno published six books in London, all of them written in Italian.xiv xiv Before and after his time in England, Bruno published only in Latin (with the solitary exception of a play, Il candelaio, published in Paris in 1582), so his choice of Italian, when his books must have mainly been sold to Englishmen (though some will have been carried to the great book fair in Frankfurt), seems odd. But Italian was the language of Dante and of Petrarch. Educated Englishmen could read it; by using it, Bruno signalled that he was addressing himself to poets and courtiers, not to professors of mathematics or philosophy.

The English were hostile to foreigners and to Catholics. If you were too obviously foreign, as Bruno was, you risked being beaten up in the street. Bruno hardly dared venture outdoors. In the dialogues he wrote he describes himself as mixing with the elite of English society, but he later claimed this was fiction not fact.⁷⁰ Still, his books must have sold, or his printer would have stopped printing them. Bruno himself was penniless, and astonished to see that the dons of Oxford wore great, jewelled rings on their fingers – we can be sure there were none on his – so he cannot have been providing his printer with a subsidy.

These books mark a true revolution. Copernicus had described a spherical universe with the sun at its centre. He had acknowledged that it might be possible to conceive of an infinite universe, but he had refused to pursue that line of thought, saying, ‘Let us therefore leave the question whether the universe is finite or infinite to be discussed by the natural philosophers’ (Copernicus himself being a mathematician, not a philosopher).⁷¹ Bruno seized on Copernicanism to argue for an infinite and eternal universe. The stars, he said, were suns, and the sun a star: here he was following not Copernicus but Aristarchus of Samos (310–230 BCE). Thus there could be other inhabited planets in the universe; even the sun and the stars might be inhabited, for they could not be equally hot all over, and there might be creatures, quite different from ourselves, who thrive on heat. Moreover, there was nothing to show that the other planets were different from the Earth. Bruno argued that the moon and the planets could be presumed to have continents and oceans, and that they shone, not by their own light (as was generally assumed; even the moon was assumed to be translucent at least), but solely by reflected light.⁷² Thus, seen from the moon, the Earth would look like a gigantic moon; seen from even further away, it would be a bright star in the sky. The Earth, Bruno thought, would shine brightly because the seas would reflect more light than the land. (Here, as Galileo later showed, he was wrong – which is why when astronomers, after the discovery of the telescope, began to make maps of the moon they named the dark patches, not the light patches, seas.) Thus Bruno imagined an infinite universe, with numberless stars and planets, all possibly inhabited by extraterrestrial life forms.⁷³ Since Bruno did not believe that Christ was the saviour of mankind (he was a sort of pantheist), he did not have to worry about how the Christian drama of sin and salvation was played out in this infinity of worlds.

Bruno was not the first to imagine an infinite universe with extraterrestrial life. Nicholas of Cusa, in his On Learned Ignorance (1440), had argued that only an infinite universe was appropriate for an infinite God. Nicholas thought the earth was a heavenly body which from a distance would shine like a star, an idea which caught Montaigne’s attention.⁷⁴ But Nicholas assumed that the earth and the sun were similar bodies. A habitable world was, Nicholas thought, hidden behind the shining visible surface of the sun; as for the earth, it, like the sun, was surrounded by a fiery mantle which was invisible to us, and which you would see only if you viewed the earth from outer space. Thus Nicholas made the earth into a heavenly body, but simultaneously he made the sun into a terrestrial one.xv xv Bruno, by contrast, was the first to distinguish stars and planets as we do now, making the sun a star and the planets, including the Earth, dark bodies shining by reflected light.

Bruno tried to resolve the standard arguments against Copernicanism by adopting the principles of the relativity of location and of movement; in his universe (unlike in those of Aristotle and Ptolemy) there was no up or down, no centre or periphery, no left and right and no way of telling if one was moving or stationary except by comparison with other objects.xvi xvi Oresme and Copernicus had adopted the principle of the relativity of movement when considering two bodies, the sun and the earth – the movement of the sun that we perceive can equally be caused by the sun moving or the Earth turning – but they had not extended the argument to the more complicated circumstances considered by Bruno. Thus, Bruno argued, you can be in the cabin of a ship sailing across a calm sea and be quite incapable of telling whether you are moving or stationary; and if you throw something straight up in the air, it falls back into your hand, it doesn’t drift backwards towards the stern of the ship as the ship moves on.⁷⁵ And Copernicus’s universe had a centre; he could not imagine (or at least could not acknowledge the possibility of) a universe in which location was purely relative. Bruno also made some radical and ill-judged alterations to the Copernican system, designed in part to eliminate basic objections to it (such as that Mars and Venus should greatly change in size if they are sometimes quite near and sometimes very far from the Earth).⁷⁶

In 1585 Bruno’s host, the French ambassador, was withdrawn from England, and Bruno had no choice but to leave with him. He wandered around Europe (carrying with him his copy of Copernicus, which is now in the Biblioteca Casanatense in Rome), and in 1592 he was arrested in Venice and handed over to the Roman Inquisition. After eight years of solitary confinement in the dark, and after prolonged torture, he was burnt alive in one of the main squares of Rome, the Campo de’ Fiori, on 17 February 1600. He had refused to recant his heresies, including his belief in other inhabited worlds.xvii xvii His books were banned throughout Catholic Europe.

Bruno is important to our story not because he was brave (though he was), or brilliant (though he was), but because he was, on occasion, right. Bruno’s revisions to, and misunderstandings of, Copernicus were misconceived. The infinite and eternal universe theory has been replaced, in the course of the last fifty years, by the Big Bang theory (so recent that it was named only in 1949).⁷⁷ But we now know that the sun is a star, that other stars have planets, and there is every reason to think that there is life elsewhere in the universe. We are not at the centre of the universe: rather, the Earth is just another planet. Bruno would find himself more at home in our universe than would Cardinal Bellarmine, the man who played the key role in his trial, as he played the key role in the Catholic Church’s condemnation of Copernicanism in 1616. On crucial points Bruno was right before anyone else: he was the first to say in print that the preface to On the Revolutions was not by Copernicus, and he was the first modern to insist that the planets shine by reflected light.xviii xviii

§ 5

It is worth comparing Bruno to Thomas Digges. In 1576, a few years before Bruno’s Oxford lecture, Digges had published a sixth edition of his father Leonard’s perpetual almanac, A Prognostication Everlasting. (The book had first been published in 1555, and it went through thirteen editions that we know of, the last in 1619.)⁷⁸ The primary purpose of the Prognostication was to enable its readers to predict the weather by a combination of astrology (the locations of the planets) and meteorology (phenomena in the atmosphere, such as rainbows and clouds). But the Prognostication also allowed you to determine when to let blood, purge (induce diarrhoea) and bathe (modern readers will find it odd to see bathing listed as if it were a medical intervention; Digges, father and son, recommend that one should not bathe when the moon is in Taurus, Virgo or Capricorn: these are earth signs, and hence at odds with water); how to tell the time from the rising of a star or from the moon; how to calculate sunrise, sunset, high tide and low tide, and the length of the day for any date. It was an eminently practical work – it provided a compass rose, for example, which you could copy on an enlarged scale, and a plan for a device for locating the planets in the heavens, which you could use as a blueprint, or (by adding a plumb line and a magnetic compass) you could turn the book itself into a paper instrument. Leonard also supplied some information that had no practical purpose: he showed the relative sizes of the sun, the planets, the earth and the moon, he explained how an eclipse of the moon could occur, and he gave the dimensions of the heavens: from the earth (which he, of course, assumed was at the centre of the universe) to the sphere of the fixed stars was, he said, 358,463 miles – and a half. To this successful work Thomas now added a translation (with a few revisions and additions of his own) of what he took to be the key sections of Book I of Copernicus’s On the Revolutions.

Few copies of the Prognostication, in any of its editions, survive. It was a cheap publication, aimed at minor gentlemen and farmers, the sort of thing that is used for lighting fires once it becomes obviously outdated. If most almanacs were designed to last only for a year, even a perpetual almanac would soon become grubby and dog-eared. By the 1640s, if it survived that long, the print and layout of most copies would have looked hopelessly old-fashioned: the first eight editions were printed throughout in blackletter; then there were three in which the bulk of the book was printed in a humanist typeface but the translation of Copernicus remained in blackletter, perhaps to signify its intellectually serious content; the whole text was given a modern appearance only in 1605. As maritime compasses became cheaper and more widely available, instructions on how to make your own would have become increasingly irrelevant. By the eighteenth century astrology itself was generally regarded as outdated. Pages of tables and designs for instruments were probably often torn out for ready reference, leaving mutilated copies. Most copies will have been thrown away long before it occurred to anyone that the book was worth preserving simply because it was old and rare. Nobody published a proper study of the 1576 edition until 1934.⁷⁹

And then, overnight, this edition became an object not only of great rarity (there are plenty of rare, ephemeral pamphlets) but also of great value. Every auctioneer, every librarian, was on the lookout for it. For it was now recognized that Thomas Digges had not only included in it the first substantial defence of Copernicanism by an Englishman, or in English,⁸⁰ he had also included an illustration of the cosmos which showed the stars arranged not in a sphere but stretching out to the limits of the page and beyond – the first illustration of an apparently infinite universe. This illustration spreads over two pages, and it appears to have been added as an afterthought as the book went through the press. Binders were never quite sure what to do with it – to incorporate it as a fold-out page, or as a two-page spread. Often it must have been damaged, torn, or either left as a loose page or omitted altogether. Of the first edition, only seven copies are known to survive, and not one has come on the market since the importance of the volume was recognized. The very wealthiest collectors have had to make do with copies of the later editions.

The 1576 edition of the Prognostication is a little puzzle in which we find the whole problem of the early modern history of science in miniature. It represents an intellectual breakthrough: Digges was the first competent astronomer explicitly to propose an infinite universe. (Nicholas of Cusa had argued that an omnipotent God would surely make an infinite universe, but this was a philosophical, not an astronomical argument.)⁸¹ Moreover, Digges was not an insignificant figure in the new astronomy. In 1573 he had published a study of the nova which had appeared the previous year.⁸² And yet at the very same time he was happily engaged in using the new astronomy to predict the weather and to decide when doctors should bleed their patients. He published his new, Copernican account of the cosmos alongside his father’s old, Ptolemaic account. He knew the Copernican system could work only if the cosmos was much bigger than the Ptolemaics had imagined, but he did not correct his father’s figures for the dimensions of the universe. His father had provided an illustration of the Ptolemaic cosmos in which the outermost sphere was labelled ‘Here the learned do appoint the habitacle of God and the Elect.’ Thomas’s illustration, modelled on his father’s, also mingles astronomy and theology: its outermost zone (now an infinite space rather than a sphere) is also labelled ‘the habitacle of the Elect’. How can the old and the new, the past and the future, rational science and superstition, live side by side here without any sign of discomfort? This question requires several answers.

The first answer is that Copernicus himself was less of a revolutionary than is commonly supposed. In all his published work, Copernicus made no mention of astrology – but nor is there anything to suggest that he would have disputed the standard view, that astronomy existed to make astrology possible.⁸³ Copernicus’s universe is different from Ptolemy’s in that the sun, not the Earth, lies at (or rather, to be exact, very close to) its centre. But it is in other respects exactly like Ptolemy’s: it is made up of a series of spheres, nested one within the other. It is finite in size.xix xix All movement within it (outside the immediate vicinity of the Earth) is determined by the fundamental principle that heavenly movement is circular and therefore unchanging. Ptolemy, Copernicus thought, had betrayed this principle not (as strict Aristotelians thought) by adding epicycles to deferents in order to explain why the planets sometimes appear to move backwards in the sky, but by introducing the equant in order to speed them up and slow them down. Copernicus achieved the same effect by different means.

Historians of astronomy trade insults on the question of whether there are equants in Copernicus or not; the answer is that there are no equants, but there are mechanisms designed to replicate equants.⁸⁴ Historians of Arabic astronomy point out that the mechanisms used by Copernicus had already been invented by the Arabs, and argue that Copernicus borrowed them without acknowledgement, rather than inventing them from scratch, though no one has yet identified a printed book or a manuscript describing the key mechanism to which he is likely to have had access.xx xx ⁸⁵

Copernicus’s own diagram showing the heliocentric cosmos from the original manuscript of On the Revolutions (1543). The moon is not drawn but is mentioned in the inscription. The sphere of fixed stars is the outer ring.

For the first two generations of astronomers reading Copernicus the crucial point about his book was not that it advocated heliocentrism but that it took the principle of circular movement more seriously and applied it more systematically than Ptolemy had. One consequence of Copernicus’s mathematical model was that it was easier to make calculations using his system than Ptolemy’s, and many astronomers proceeded to publish Copernican tables of planetary locations even though they thought that Copernicanism was not a plausible description of how the cosmos is organized. (Just as everyone happily uses the underground map for London, even though it distorts the distances between the stations; its great advantage is that it makes it easy to work out which route to follow and where to change, while a spatially accurate map would be much harder to read.)

But Digges was not a conventional reader of Copernicus, for he understood that Copernicus really did intend to be taken literally when he described the Earth as moving and the sun as stationary. In his version of Book I of On the Revolutions the arguments that could be adduced against the movement of the Earth are given a more prominent position. According to the perfectly standard figures offered by Leonard Digges, the circumference of the Earth measures 21,600 miles, which means that, if Copernicus is right and the Earth rotates once a day on its axis, this movement alone requires us to travel at 900 miles an hour, quite apart from the additional movement required for the Earth to travel in a vast circle around the sun once a year. If we are flying along at 900 miles an hour, it was argued (and remember that those doing the arguing had never travelled faster than they would on a galloping horse, about 30 miles an hour), then we ought to be able to feel the movement; the wind should be rushing through our hair. Birds, when they take off from trees, should be swept away towards the west. If you drop something from the top of a tower it should fall considerably to the west of the base of the tower. Digges insists these arguments are mistaken (and he thus may have influenced Bruno’s discussion of the relativity of motion). If you climb to the top of the mast of a moving ship, Digges argues, and lower a plumb line, the plumb line will descend vertically to the bottom of the mast; it will not stream out backwards until the plumb ends up in the water behind the ship. This is a slightly different (and less convincing) experiment to the one later imagined (or performed) by Galileo, where you drop an object from the top of a mast, but it serves the same purpose of establishing that the concept of verticality is relative: a plumb line or a falling body on a moving ship will make a line vertical to the deck of the moving ship, not vertical to a fixed point on the surface of the Earth. Galileo also demonstrated that if you throw an object straight up in the air when on a moving ship, it falls not far behind you but straight back in your hand: this was a direct refutation of an argument from Giambattisa Capuano, who may be the source of all these moving-ship experiments, some real, and some merely thought experiments. Thus Digges had not simply translated Copernicus but strengthened his argument at its most vulnerable point.⁸⁶

After the discovery of his illustration of the cosmos, Digges was given the credit for being the first to portray the stars as arranged not in a sphere but scattered over the outer margins of the page until they disappeared, and he certainly thought the stars went on for ever. But Digges’s universe has a centre, so it is not really infinite, for an infinite universe could have no centre. He thinks each star is larger than the whole solar system; they have to be a truly astonishing distance away or there would be some measurable change in their relative positions as the Earth moves on its vast orbit around the sun, and so they must be absolutely gigantic if they are to remain visible.⁸⁷ It follows that Digges does not think of the sun as a star, or of the stars as suns. Moreover, his universe is shaped by his theology. The space occupied by the stars is heaven, the habitation of God, the angels and the elect. The solar system is the zone of sin and damnation. This sinful world is, Digges tells us, a dark star – ‘this litle darcke starre wherein we live’.⁸⁸

In fact, Digges’s image of the universe – its boundless extent, its identification of the stars with heaven and of the Earth with hell (hence, perhaps, Mephistopheles’ famous line in Marlowe’s Doctor Faustus (1592): ‘Why this is hell, nor am I out of it’), its description of the Earth as a dark star – comes from a popular poem commonly read by English schoolboys at the time, The Zodiake of Life (Latin, 1536) by Marcello Palingenio Stellato.⁸⁹ Digges knew the eleventh book of the poem off by heart ‘& takes mutch delight to repeate it often’.⁹⁰ What Digges had done was put the sun rather than the earth at the centre of Stellato’s universe.

Digges’s image of the Copernican cosmos, with the stars extending outwards to the edge of the page, symbolizing a universe without bounds (from the Prognostication – this is from the Linda Hall Library copy of the 1596 edition, but the illustration first appears in 1576).

Stellato had been posthumously condemned by the Inquisition for denying the divinity of Christ (heretical works he had written had been found among his papers after his death), and his body had been dug up and burnt, but Protestant Europe knew nothing of his rejection of Christianity (although there are plenty of indications of it in the Zodiake), and his anticlericalism and his determinism made it possible to read him as, if not actually a Protestant, then at least as sympathetic to Protestant views.⁹¹ The fact that the Zodiake had been placed on the Index only confirmed this reading. For his English publishers, and presumably for Digges, he was ‘the most Christian poet’ (1561), ‘the godly and zealous poet’ (1565), ‘the excellent and Christian poet’ (1576), although Bruno, astutely, read him as a kindred spirit. It never occurred to Digges that the Earth might shine like a star, or that the planets are other Earths. The sun and the Earth are unique, and the universe has a centre.

Stellato and Digges were not the only ones to think of the Earth as a dark star.⁹² In 1585 Giovanni Battista Benedetti published a collection of essays in which he dealt, among much else, with issues in contemporary cosmology. Like Digges, Benedetti was a realist Copernican. But he was more radical than Digges. Noticing that the path of the moon is in effect an epicycle around the path of the Earth, and that the planets appear to travel through epicycles, Benedetti proposed a remarkable hypothesis: what we think of as planets, he suggested, are merely the shining moons circling dark planets. These hidden planets (they are ‘cloaked’, to adopt the terminology of Star Trek) are Earth-like and presumably carry life. Benedetti’s proposal was based on the assumption that the moon and the Earth are made of quite different types of substance, the moon being much more reflective than the Earth, although less reflective in the darker patches, where much of the sun’s light is absorbed rather than being reflected. Benedetti held that the universe is spherical but that it is surrounded by unbounded empty space.⁹³

Digges and Benedetti had not read Bruno, so they had not encountered his theory that, from a distance, the Earth would be indistinguishable from a star. A great proto-scientist, William Gilbert (1544–1603), the founder of modern studies of magnetism and electricity, however, had read Bruno, and adopted his arguments wholesale. Gilbert copied from Digges his illustration of an unbounded universe. But Gilbert understood that, seen from the moon, the Earth would shine like a vast moon; and that, seen from further away, it would shine like a star (here, he directly argued against Benedetti). The moon, he thought, had continents and oceans, just like the Earth. Like Bruno, he thought the oceans would be brighter than the land. He saw no reason why the other planets should not be just like the Earth.⁹⁴

Gilbert drew, before the invention of the telescope, the first map of the moon, and as a result discovered its libration, the fact that it appears to turn slightly, up and down and from side to side, as it faces the Earth. This confirmed his conviction that the planets float freely in space. Moreover, Gilbert was the first to break completely with the notion that movement in the heavens must be circular: his planets trace complicated paths through the void; such a path could explain why the moon appears to wobble in the sky. Gilbert’s On the Universe was never finished (he died in 1603, but the section on cosmology appears to date from the early 1590s), and it was not published until 1651. Bacon read it in manuscript, but had no time for it: Gilbert’s preoccupation with magnetism seemed to him an irrational obsession, and as a result he had ‘built a ship out of a shell’.⁹⁵

§ 6

Digges, Bruno, Benedetti and Gilbert were members of the tiny group of realist Copernicans. They were bold pioneers of the new philosophy. Yet it would be wrong to think that they shared any common understanding of what natural science is or how it should be conducted. Digges was a proper mathematician. He taught surveying, navigation, cartography and military engineering. He experimented with mirrors and lenses; some think he had a secret telescope. He tried to measure the distance from the Earth of the supernova of 1572 and established that it was in the heavens – thus refuting that central claim of Aristotelian philosophy, that there was never any change in the heavens. (Digges thought it was a miraculous event and provided advice to the English government on what it might portend.)⁹⁶

Benedetti was a comparable figure to Digges: he was an adviser on mathematical and engineering questions to Duke Emanuele Filiberto of Turin, and he published on perspective, the construction of sundials (which itself involves questions of perspective, since one must project the path of the sun on to a flat surface), calendar reform, the physics of falling bodies and the question of earth and water. But his cosmological arguments are purely speculative and philosophical.

Gilbert was a physician (he was briefly the personal physician to first Elizabeth I and then James I) who chose to embark on a programme of experimental enquiry into the workings of magnets, and he evidently had close links to the experts who made compasses and taught navigation. His study of the moon’s libration shows that he was looking for new observations with which to solve cosmological issues.

There is an old-fashioned way of writing the history of early modern science in which Copernicus, Digges, Benedetti and Gilbert are presented as scientists, although none of them used the word of themselves. The assumption is that they were engaged in an activity which is continuous with modern science; indeed, they were all Copernicans, and the publication of On the Revolutions is often (wrongly) taken to mark the beginning of modern science. Not Bruno, however, despite his Copernicanism. Bruno read Copernicus, lectured and wrote about him; often he was right where Copernicus was wrong. But he had no interest in measurement or experiment; he thought Copernicus was excessively preoccupied with mathematical problems. Copernicus, Digges and Benedetti called themselves mathematicians; Bruno and Gilbert called themselves philosophers. Copernicus and Digges wrote books on astronomy; Benedetti on physica (natural science); Gilbert on physiologia (the study of nature). None of them was a scientist, because science, as we understand the term, did not yet exist. Newton, however, was a scientist – who can doubt it? Sometime between the 1600s and the 1680s, science was invented.