9

Laws

Nature and Nature’s Laws lay hid in Night: GOD said, Let Newton be! And all was Light.

– Alexander Pope, epitaph for Sir Isaac Newton (published 1735)

§ 1

On 10 November 1619 a young French soldier, René Descartes (1596–1650), found himself stranded in Ulm.¹ He was in the service of Duke Maximilian of Bavaria, a Catholic, and the terrible pan-European conflict afterwards known as the Thirty Years War was just beginning. There was the prospect of fighting ahead; but immediately it was winter, and there was, literally, nothing for a soldier to do. Descartes had received a good, but largely conventional, education from the Jesuits, and had spent two years at university studying law, to please his father; but in 1619 he had no reason to think he would ever pursue any other profession than soldiering. But, trapped by winter, Descartes shut himself up in a room heated by a stove and sat and sat, and thought and thought. He reached the conclusion that what was wrong with all existing systems of knowledge was that they had been cobbled together by lots of different people over long periods of time. What was needed was a complete new beginning. One person, starting from scratch, should redesign the whole of philosophy, including natural science.

Excited and exhausted, Descartes fell asleep and dreamed three dreams. In the first he was attacked by phantoms and a great wind, and he had some sort of debilitating weakness on one side of his body. He tried to get into a chapel to pray but was unable to enter. Waking from this nightmare, Descartes prayed and tried to compose his spirits. When he fell asleep again he heard a great clap of thunder and, as it seemed to him, opened his eyes to see the room filled with sparks from the fire; it was not clear to him quite when he became properly awake, but eventually the sparks disappeared and he fell asleep again. In his third dream there was a big book, a collection of poems. Opening it, he found the words Quid vitae sectabor iter? ‘What road in life shall I follow?’ A stranger entered and gave him another poem, beginning with the words est et non, ‘it is and it is not.’ Descartes tried to find this poem in the book, but the book and the stranger disappeared. (In the Corpus poet-arum of Pierre de la Brosse (1611) these two poems are to be found on facing pages.) Lying half awake, Descartes tried to interpret his dreams. The first two, he thought, were to be understood as showing that he had lived his life badly up to now, the third as mapping out his future: his path must be to commit himself to the philosophical task of establishing what is and what is not.

For the rest of his life Descartes dated his new life as a philosopher from these dreams. He began work on a series of rules for thinking which would help him establish the truth; he sold some property so that he could be self-sufficient and concentrate on his great enterprise. Fourteen years later, and living in the Protestant Netherlands, Descartes was about to publish a series of treatises in natural science when he heard of the condemnation of Galileo and decided that, since his philosophy supported Copernicanism, he dare not publish for fear of condemnation by the Catholic Church (though Descartes would have been in no danger had the Church condemned him, as in the Netherlands he was safely out of the Inquisition’s reach). In 1637 he finally published the Discourse on Method and three essays on mathematics and natural science. In 1641 he published a summary of his philosophy, The Principles.

By the time Descartes published the Discourse he had decided that the best way to introduce his philosophy was through applying scepticism absolutely to the limit. How can we know that the world is real? How can we know that we are not dreaming? How can we know that we are not being systematically deceived by some demonic demiurge? We cannot. There is only one thing we can be sure of: cogito ergo sum (‘I think, therefore I am’). From this one secure starting point Descartes sets out to prove the existence of a god who would not allow us to be systematically deceived, and then to construct an account of the natural world as consisting purely of matter in motion. The Discourse is a strange work because it is both autobiography and philosophy; Descartes teaches us how to think by telling us the steps he went through in learning how to think. And so he tells the reader not about the dreams, which were too private an experience, but about the day he spent in the stove-heated room, when his life as a philosopher began.

The problem with Descartes’ story is that it is not true. There is no reason to doubt the stove-heated room or the dreams, but all the evidence suggests that Descartes’ new philosophical life had begun exactly a year earlier, on 10 November 1618. That day, he was in Breda, in the Low Countries, serving in the army of the Protestant Maurice of Nassau. In the town someone had stuck up a poster challenging people to solve a mathematical problem. The poster was in Flemish, and so Descartes turned to the person standing next to him, also studying the poster, and asked him to translate it. This person was Isaac Beeckman, a schoolteacher and engineer, and we know about his relationship with Descartes because he kept a diary which was rediscovered in 1905 and published in four volumes between 1939 and 1953.²

Descartes and Beeckman chatted away in Latin and discovered that they had interests in common. ‘Physico-mathematicians are very rare,’ wrote Beeckman in his diary a few days later; indeed, the stranger had told him that ‘he has never met anyone other than myself who pursues his studies in the way I do, combining Physics and Mathematics in an exact way. And for my part, I have never spoken with anyone apart from him who studies in this way.’³ But Beeckman was far ahead of Descartes in his thinking. He had already decided that the universe consisted of corpuscles in motion, and that the ‘laws’ of motion (Beeckman’s standard word for a law of nature is pactum: ‘covenant’)⁴ which functioned at a microscopic level must be the same as those which functioned at a macroscopic level. He was well on the way to formulating, entirely independently, Galileo’s law of fall. For two months Beeckman and Descartes worked closely together, and when Beeckman left Breda they kept up a correspondence in which Descartes told Beeckman that they were bound together ‘in a bond of friendship that will never die’. Beeckman was assured by Descartes that:

[Y]ou are the one person who has shaken me out of my nonchalance and made me remember what I had learned and almost forgotten. When my mind strayed far from serious concerns, it was you who guided it back down the correct path. If, therefore, by accident I propose something which is not contemptible, you have every right to claim it for yourself.⁵

Years later, in 1630, Beeckman did just that. In a letter to Descartes’ friend Mersenne he mentioned that some of Descartes’ ideas on music had come from him. Descartes was absolutely furious and denied any influence, but when Mersenne visited Beeckman and read his journal he discovered that, indeed, many of Descartes’ ideas had first been formulated by Beeckman. Descartes exploded again, telling Beeckman he had learnt as much from him as he had learnt from ants and worms. This was followed on 17 October 1630 by one of the longest letters Descartes ever wrote; it runs to twelve pages in print and is full of bitter invective in which Descartes explains to Beeckman that he is mentally ill and delusional.⁶

Why could Descartes not bear the simple truth, that much of what he knew he had learnt from Beeckman? Because ever since he had awoken from his dreams on the morning of 11 November 1619 he had told himself that he was building a new philosophy single-handedly, that he was starting from scratch, that he owed nothing to anyone. The truth of his intellectual dependence on Beeckman was utterly intolerable to him. And so, in the autobiographical account which opens the Discourse on Method in 1637, there is no Beeckman, there is only the famous account of the stove-heated room:

At that time I was in Germany, where I had been called by the wars that are not yet ended there. When I was returning to the army from the coronation of the Emperor, the onset of winter detained me in quarters where, finding no conversation to divert me and fortunately having no cares or passions to trouble me, I stayed all day shut up alone in a stove-heated room, where I was completely free to converse with myself about my own thoughts. Among the first that occurred to me was the thought that there is not usually so much perfection in works composed of several parts and produced by different craftsmen as in the works of one man. Thus we see that buildings undertaken and completed by a single architect are usually more attractive and better planned than those which several have tried to patch up . . .⁷

§ 2

Let us return to the table which we discussed in Chapter 3. In its case, according to an Aristotelian, the formal and final explanations are exterior to it – its form is in the mind of the carpenter and its purpose is to provide someone with a place to work. But in the case of the oak tree, the form and the purpose are in some sense inside the acorn. Efficient causes are external; formal and final causes are, in natural objects, internal; and material causes are, to begin with, external, although (like the water absorbed by the oak’s roots, or the breakfast I have just eaten) they become internal.

A mechanical explanation, by contrast, is about external causes, not internal ones. If you are an ancient atomist – Epicurus or Lucretius, for example – you do not accept that there is any internal principle causing the formation and development of the oak tree, directing it towards the realization of its potential. Atoms are merely passive lumps of matter. An oak tree is an agglomeration of atoms which has been given a certain shape by external forces, in the same way that my house is an agglomeration of bricks that has been given a certain shape. For an ancient atomist or an early modern mechanist (such as Beeckman or Descartes) causation is always external, never internal; there are only efficient or mechanical causes.ⁱ There are no formal or final causes, and the material cause never varies.

For Epicurus and Lucretius, what matters about atoms is their size, shape and movement. If this is all that atoms have, then the qualities that we perceive in the world – colour, taste, smell, sound, texture, temperature – must be by-products of size, shape and movement. Size, shape and movement must be primary, and the other qualities must be secondary. If sound is produced by vibration, then it is easy to see that sound may be the result of movement. If rubbing two sticks together produces heat, then it is possible to imagine that heat may be a form of movement. One can hypothesize that smell is produced by particles entering the nose. Primary qualities are objective; secondary qualities are subjective, in that they depend on our ways of sensing. In a world without ears there would be no sounds, only vibrations; in a world without noses there would be no smells, only particles floating in the atmosphere. Galileo’s example, to clarify this idea of the subjectivity of sensation, is a tickle: tickle me with a feather and I feel a distinct sensation, but there is nothing in the feather which corresponds to my sensation of being tickled. This distinction between objective reality and subjective sensation was made by Lucretius. Galileo, in The Assayer (1623) was the first modern author to echo it, although without mentioning Lucretius by name (for he was regarded as a dangerous atheist; but we know that Galileo owned two copies of On the Nature of Things).⁸ After Galileo the distinction was adopted by Descartes. The terminology we now use to express this distinction, between primary and secondary qualities, was introduced by Boyle in 1666 and popularized by Locke in 1689.⁹ (The Lockean language of primary and secondary qualities replaces an earlier Aristotelian language of primary and secondary qualities, the primary qualities being hot and cold, wet and dry.)

Although Descartes followed the ancient atomists in their distinction between primary and secondary qualities, he rejected their belief in empty space, the void. Matter, as far as Descartes was concerned, had only one fundamental characteristic, that it occupied space; it followed that there could be no vacuum, as this would be space with nothing occupying it. For Descartes the material world consists of divisible corpuscles. He avoids the term ‘atoms’ because the ancient atomists had insisted that atoms were not divisible and that the space between them was empty.

Matter, in Descartes’ scheme of things, can interact only by direct contact; there can be no action over a distance, and when two bodies interact they can do so only by pushing against each other, so magnetism and gravity have to be explained as the result of some sort of pushing process, not a pulling one. According to Descartes, in the case of gravity this pushing process was the result of the Earth being caught up in a vast vortex of fluid swirling round the sun. This vortex both held the planets in their orbits and pressed objects down towards the surface of the Earth. The sun was just one among many stars, each surrounded by its own vortex. Similarly, magnetism worked through little streamers of corkscrew-like matter reaching out and locking on to iron: the pull of a magnet is in fact a push, just as a corkscrew pushes a cork out of a bottle. (I may pull on the corkscrew, but the corkscrew pushes on the cork.)

In Descartes’ system there is only one type of matter which, by its interactions and conglomerations, produces the vast diversity of materials that we experience. The laws of its interaction are the three laws of nature. These are that ‘each thing, as far as is in its power, always remains in the same state; and that consequently, when it is once moved, it always continues to move’; that ‘all movement is, of itself, along straight lines’; and that ‘a body, upon coming in contact with a stronger one, loses none of its motion; but that, upon coming in contact with a weaker one, it loses as much as it transfers to that weaker body.’¹⁰

It is easy to regard Cartesianism, with its insistence that magnets, corkscrews and even what we now call gravity always push and never pull, as something of a joke, but recent work has shown that Descartes carried out some subtle and beautiful experiments, and his vortex theory was still viable well into the eighteenth century.¹¹ The crucial dispute between Cartesians and Newtonians was over the shape of the Earth: Newton predicted an oblate ellipsoid, or flattened, Earth, while Descartes had predicted a prolate ellipsoid, or egg-shaped, Earth. French expeditions to Peru and Lapland (1735–44) discovered (to the dismay of those involved) that Newton was right and Descartes wrong.^ii 12

§ 3

The modern idea of laws of nature is a by-product of Descartes’ philosophy, for Descartes was the first person to treat the laws of nature as being what knowledge of nature was all about. Galileo, Harriot and Beeckman had each independently discovered what we call the law of fall; but none had used the word ‘law’ in this context. According to the Comte de Buffon, writing in the eighteenth century, ‘Nature is the system of eternal laws established by the creator’;¹³ it follows that the core task of science is the identification of the laws of nature.ⁱⁱⁱ Buffon could, if he wished, look back to the seventeenth century and identify a whole series of laws that had been discovered during the Scientific Revolution: Stevin’s law of hydrostatics, Galileo’s law of fall, Kepler’s laws of planetary motion, Snell’s law of refraction, Boyle’s law of gases, Hooke’s law of elasticity, Huygens’ law of the pendulum, Torricelli’s law of flow, Pascal’s law of fluid dynamics, Newton’s laws of motion and law of gravity. Most, perhaps all, of these had been given the title ‘laws’ by Buffon’s day (only Newton used the word ‘laws’ when describing his own discoveries), though only a minority had already acquired eponymous labels – the rest were yet to be named after their discoverers.^iv It is hardly surprising that there is a book on the Scientific Revolution entitled Nature and Nature’s Laws, since the discovery of laws of nature is one of the Scientific Revolution’s most remarkable achievements.¹⁴ In 1703 Newton became President of the Royal Society and drew up a scheme to define its goals. ‘Natural Philosophy,’ he wrote, ‘consists in discovering the frame and operations of Nature, and reducing them, as far as may be, to general Rules or Laws, – establishing these rules by observations and experiments, and thence deducing the causes and effects of things . . .’¹⁵ Laws of nature were now what science was all about.

By contrast, the ancients had known, by our reckoning, only four physical laws: the law of the lever, the optical law of reflection, the law of buoyancy and the parallelogram law of velocities.¹⁶ Or, rather, the ancients had known four principles that we call laws. The ancients did refer to the ‘laws’ of nature when they wanted to say that nature is regular and predictable, but they never identified any particular scientific principle as a law. The Romans talked a great deal about the law of nature (lex naturae), but they usually meant the moral law.

A law is an obligation (‘Thou shalt not kill’, for example) imposed on some creature (human beings, angels) capable of accepting or rejecting that obligation. The moral law applies to rational, language-using creatures, and the law of nature binds all human beings by virtue of their capacity to recognize that there are moral obligations that are common to them all. There are no laws in non-human nature because human beings are (as far as we know) the only rational, language-using beings in nature. To talk about those regularities that appear in the natural world as ‘laws’ is to speak metaphorically: this was as obvious in the first century and the seventeenth as it is now. But the metaphor is pretty straightforward. The Greeks made use of it now and again (although most of the time they liked to contrast the natural and the social), and the Romans, who were always in and out of the law courts, found it an obvious way of referring to the fact that nature is regular and predictable in its workings. It was an even more obvious metaphor for Christians, as it was easy to think of God as a legislator imposing laws on nature and to personify nature as obeying him.

So when we talk about nature’s laws we can be talking about laws governing human behaviour or laws governing nature – ‘natural law’ and ‘laws of nature’, as we now say. In classical Latin there is no distinction: lex (or ius) naturae and naturalis lex (or ius) are synonymous, and their most frequent usage is to refer to those moral laws that all human beings have (or are supposed to have) in common. So, too, at first, in the modern languages. ‘Law of nature’ is the most common term in English before 1650 (Hobbes, an extreme case, uses ‘natural law’ twice in Leviathanand ‘law of nature’ more than a hundred times) and loy naturelle the most common in French (and so, too, legge naturale in Italian, ley natural in Spanish). A linguistic distinction to separate the two kinds of law, moral and scientific, emerges with Descartes, who writes of la loy (or les loix) de la natureand never (when discussing scientific matters) of la loi naturelle. Before Descartes la loy de nature and la loy de la nature were synonymous, although the first was more common. But Descartes and his authorized translator from Latin into French never write of la loy de nature.Thus Descartes opted for the least common phrase available to him in French as a translation for lex naturae in order to give his phrase a precise reference to scientific laws, not moral ones. By a similar process in German, the rarer term, Naturgesetz, comes to mean primarily law of nature, while the commoner term, Naturrecht, continues to mean natural law.

It is surely easier to give an uncommon phrase a new meaning than a common phrase. The English, however, followed Descartes in using ‘law of nature’, not ‘natural law’ to refer to scientific laws; but this had a peculiar effect, as ‘law of nature’ was the most common term in English for the moral law. Using the same term for both was unnecessarily confusing, and over time the moral and political philosophers and the theologians largely abandoned ‘law of nature’, ceding it to the scientists, and switched to ‘natural law’, bringing themselves into line with the French, Germans and Italians. This is a striking case of French imposing itself upon English, and of the scientists determining for the first time the language of the theologians. As a result, for us moderns, laws of nature are scientific laws and natural laws are moral laws. In this respect, we are all Cartesians.

§ 4

Long before Descartes we can find references to laws of nature in a scientific context, and scholars have struggled to disentangle the origins of the concept.¹⁷ There is no doubt that it has multiple origins, nor that it takes on a wholly new importance with Descartes. I will distinguish three origins, of which the most important (in my view) has been largely neglected hitherto. First, nominalist philosophers from William of Ockham (1288–1348) on attacked the Aristotelian doctrine of forms. There is, they argued, no such thing as a form or essence, there are only particular objects. When we talk about forms we are using a label (or name, hence the term ‘nominalism’) that we have chosen to attach to certain particulars. In their view, Aristotelian forms were ghostly presences; you could never capture them, but they were always being added into the explanation. Clearly, in the case of making a table, the carpenter does have a plan: the form is an idea in his mind, and the table he makes corresponds to that form. But where is the form of the oak tree? And if you cannot locate it, how can it act in the world? If the universe is regular and predictable, this is not because there are internal forms but because God has imposed order upon it from the outside. God could have made the universe in many different ways; he has arbitrarily chosen to make it as it is, and the order it displays is an order he has chosen to impose upon it. Thus Jean Gerson (1363–1429), a nominalist, maintains that ‘the law of nature as concerns created things is what regulates their movement and action and their tendency towards their goals.’¹⁸ The term ‘law’ here implies external, divine causation, but the particular content of the law of nature is never specified, and there is surely scope for occasional exceptions to the law, if only monsters and miracles. Some modern commentators would like to argue that the invention of laws of nature could take place only in a monotheistic culture, where God could be conceived as an absolute legislator – thus, the Scientific Revolution owes everything to Christianity. It is certainly true that the arguments of the nominalists are theocentric, but as we shall see this is not true for other ways of thinking about laws of nature.

Secondly, in the mathematical disciplines, lex was often used as a synonym for regula, or ‘rule’, to refer either to natural regularities which could not be shown to be strictly necessary – in other words, where there was no full philosophical (causal) explanation – or to axioms. Thus Roger Bacon refers to the law of reflection (the angle of reflection equals the angle of incidence) and Copernicus’s disciple Rheticus claimed that Copernicus had discovered ‘the laws of astronomy’ (Copernicus himself made no such claim). Ramus, as we have seen, writes of the ‘laws’ of Ptolemy and of Euclid.¹⁹ The term ‘law’ implies unbroken regularity, with no exceptions, but nothing is conveyed about causation. These laws have a specifiable content.

Both traditions come together in Paris in the work of Jean Fernel (1497–1558), who started his career as an astronomer and mathematician and then turned to medicine, inventing the term ‘physiology’. According to Fernel, there are eternal, immutable laws which govern the universe: these are ordained by God, and without them there would be no order in the universe. The laws of medicine fit within this wider structure of laws, and the fundamental law of medicine is the ancient Hippocratic principle that opposites cure opposites – a fever is cured by cooling the body, for example. This looks to us like a principle, maxim, or rule of thumb, not a law, because it lacks specificity.²⁰

Neither the nominalist nor the mathematical usages are particularly common, and we cannot show any direct influence from either of them on seventeenth-century usages. Galileo makes only three references to the laws of nature, on each occasion when he is arguing against the theological objections to Copernicanism; there are no laws of nature in his more properly scientific works.²¹ The first person to place the idea of a universal law at the heart of the attempt to understand nature, and to give some specific content to that idea, was Descartes, first in his correspondence in 1630, next in The World (completed by 1633 but published only posthumously: Descartes gave up all hope of publication when he heard of the condemnation of Galileo), and then in The Principles of Philosophy (Latin 1644, French 1651; in the earlier Discourse of Method Descartes uses the phrase ‘principles of nature’ instead of ‘laws of nature’).^v Descartes, as we have seen, propounded three laws, but his laws, the first laws that matter, do not appear on modern lists of scientific discoveries: Descartes’ first two laws of nature became Newton’s first law of motion, and the third was refuted by Newton’s laws.

More to the point, a modern list of laws would have deeply puzzled Descartes. His three laws were intended to be the only laws. From them, it should be possible to derive a complete system of knowledge embracing every aspect of the natural world, just as one can deduce the whole of Euclidean geometry from five axioms. He had no intention of seeing laws proliferate and multiply. Of course, as he worked out the implications of his laws, he drew a series of subsidiary conclusions. There were, for example, seven rules (regulae) which enabled one to predict what would happen in collisions between bodies travelling along the same straight line (bodies in a vacuum, although Descartes held there was no such thing; the rules for bodies travelling in a plenum were beyond even his capacity): these rules are never called ‘laws’. In the course of half a century Descartes’ term ‘laws of nature’ established itself as central to the language of science, but at the same time its meaning mutated, so that it soon ceased to resemble Descartes’ original conception.

Where did Descartes’ conception of a law of nature come from? Well, Lucretius had a concept of a law of nature, although he does not use the phrase lex naturae; instead he uses (three times) the phrase foedus naturae. A foedus is a league, or compact, but it is often used as synonymous with lex, and Renaissance commentators on Lucretius interpreted him as talking about the laws of nature.²² Bacon writes of ‘the law of nature and the mutual contracts of things’: he is paraphrasing Lucretius. For Lucretius the attraction of iron by a magnet takes place according to a law of nature, and species breed true, dogs producing dogs and cats cats, according to a law of nature. It seems absolutely certain that Descartes had Lucretius in mind when formulating his laws of nature because he picks up on a phrase he uses, quantum in se est (a phrase very difficult to translate but, roughly, ‘as much as in it lies’), in his first law of motion. Lucretius uses the phrase four times in On the Nature of Things, twice in discussions of the way in which atoms naturally fall downwards, ‘as far as in them lies’, through the void, passages which prefigure Descartes’ conception of inertia. The same phrase is then employed by Newton in his definition of inertia; he evidently took the phrase from Descartes, and discovered only later that it originated in Lucretius.²³

So far, in tracing the idea of a law of nature, we have been following an established line of argument. But to understand where Descartes’ preoccupation with laws of nature comes from we must consider a text that has not previously been discussed in this context. We must turn to the longest and most philosophical of Montaigne’s essays, ‘An Apology for Raymond Sebond’, first published in 1580. The passage originally contained a single direct quotation from Lucretius, but two more were added in 1588, and we will see in a moment that it is directly inspired by Lucretius on foedus naturae. Here is an abbreviated version, in which, for simplicity’s sake, I omit Montaigne’s later additions to the 1580 text:

Nothing of ours can be compared or associated with the Nature of God, in any way whatsoever, without smudging and staining it [i.e. the Nature of God] with a degree of imperfection . . .

We wish to make God subordinate to our human understanding with its vain and feeble probabilities; yet it is he who has made both us and all we know. ‘Since nothing can be made from nothing: God could not construct the world without matter.’ What! Has God placed in our hands the keys to the ultimate principles of his power? Did he bind himself not to venture beyond the limits of human knowledge? . . . You only see – if you see that much – the order and government of this little cave in which you dwell; beyond his Godhead has an infinite jurisdiction. The tiny bit that we know is nothing compared with ALL:

omnia cum coelo terraque marique

Nil sunt ad summam summaï totius omnem.

[Lucretius: ‘The entire heavens, sea and land are nothing compared with the greatest ALL of all’]

The laws you cite are by-laws [une loy municipale]: you have no conception of the Law of the Universe [l’universelle; i.e. la loi universelle]. You are subject to limits: restrict yourself to them, not God . . . [Montaigne enumerates various miracles] a material body cannot pass through a solid wall; a man cannot stay alive in a furnace . . . It is for you that he made these laws [regles]; it is you who are restricted by them. God, if he pleases, can be free from all of them: he has made Christians witnesses to that fact . . .

That Reason of yours never attains more likelihood or better foundations than when it succeeds in persuading you that there are many worlds . . . it seems unlikely, therefore, that God made only this one universe and no other like it . . . Now if there are several worlds, as Epicurus and almost the whole of philosophy have opined, how do we know whether the principles and laws which apply to this world apply equally to the others?²⁴

Montaigne’s thinking here derives from his reading of Lucretius. In his own copy, against one of the four passages where Lucretius discusses the foedera naturae, the laws of nature, he wrote, summarizing Lucretius, ‘The order and uniformity of the conduct of nature makes evident the uniformity of her principles.’²⁵ It is against this position that he appears to be arguing here. Whether he is sincere, it is a little hard to say: having emphasized his belief in miracles, he proceeds to make the concept of a miracle entirely subjective only a few paragraphs later. What is important for our immediate purposes is the way in which his discussion echoes through the later literature on the law of nature, for of course Montaigne was read by all the educated.

Here is Walter Charleton more or less paraphrasing Montaigne in 1654:

[B]y the Law of Nature, every Body in the Universe is consigned to its peculiar Place, i. e. such a canton of space, as is exactly respondent to its Dimensions: so that whether a Body quiesce, or be moved, we alwayes understand the Place wherein it is Extense, to be one and the same, i. e. equal to its Dimensions.

We say, By the Lay [sic] of Nature; because, if we convert to the Omnipotence of its Author, and consider that the Creator did not circumscribe his own Energy by those fundamental Constitutions, which his Wisedom imposed upon the Creature: we must wind up the nerves of our Mind to a higher key of Conception, and let our Reason learn of our Faith to admit the possibility of a Body existent without Extension, and the Extension of a Body consistent without the Body it self; as in the sacred mystery of our Saviours Apparition to his Apostles, after his Resurrection . . . the dores being shut [compare Montaigne: ‘a material body cannot pass through a solid wall’]. Not that we can comprehend the manner of either, i. e. the Existence of a Body without Extension, and of Extension without a Body; for our narrow intellectuals, which cannot take the altitude of the smallest effect in Nature, must be confest an incompetent measure of supernaturals: but that, whoever allowes the power of God to have formed a Body out of no praeexistent matter [compare Montaigne: ‘nothing can be made from nothing’] cannot deny the same power to extend to the reduction of the same Body to nothing of matter again.²⁶

And here is Boyle distinguishing, following Montaigne, between the universal laws and the municipal laws of nature (the term ‘municipal laws’ is, as he recognizes, an odd one to use in English; he only uses it, I feel sure, because he has Montaigne in mind):

[W]e may sometimes usefully distinguish between the Laws of Nature, more properly so call’d, and the Custom of Nature, or, if you please, between the Fundamental and General Constitutions among Bodily Things, and the Municipal Laws, (if I may so call them,) that belong to this or that particular sort of Bodies. As, to resume and somewhat vary our Instance drawn from Water; when this falls to the Ground, it may be said to do so by virtue of the Custom of Nature, it being almost constantly usual for that Liquor to tend downwards, and actually to fall down, if it be not externally hinder’d. But when Water ascends by Suction in a Pump, or other Instrument, that Motion, being contrary to that which is wonted, is made in virtue of a more Catholick Law of Nature, by which ’tis provided, that a greater Pressure, which in our case the Water suffers from the weight of the Incumbent Air, should surmount a lesser, such as is here the Gravity of the Water, that ascends in the Pump or Pipe.²⁷

Descartes also certainly read Montaigne, and he drew from him an astonishing idea: a proper law of nature would be universal in the sense that it would not just be true for this universe, it would be true for any possible universe. Nowadays we have a less stringent test: laws of nature are true for every time and every place in our universe.²⁸ If we take this to be a central feature of laws of nature, then it is very difficult to see how an Aristotelian could have any conception of them. In Aristotelian physics different laws apply in the sublunary and supralunary spheres.²⁹ In the one there is change and natural movement is vertical, while in the other there is no change and natural movement is circular. There are no physical laws common to both spheres. In the sublunary sphere it might seem easy to formulate some general laws: all living creatures die; children take after their parents. But the phoenix does not die, and monstrous births do not resemble their parents. Aristotelians therefore recognize that, in the sublunary sphere, there are no regularities which do not have exceptions; in the supralunary sphere, all is regularity without exception; there are no regularities which apply in both spheres. Consequently, there are no Aristotelian laws of nature.

Descartes, however, is not after universality in the limited sense in which we would understand the term but in the stronger sense introduced by Montaigne when he asks what laws would apply in other universes, assuming such exist. In The Principles of Philosophy (1644) Descartes insists that he is not describing the laws that govern our universe but a set of laws such that, if one began with pure chaos, a universe indistinguishable from ours would evolve. This is not, Descartes assures us, how our universe began: God made it, God ordered it, as we all know. But it enables us to establish the laws that would need to apply in any possible universe. Descartes gets himself into a bit of a tangle here. He wants, like the nominalists, to insist that God freely determined what the laws of nature and even of mathematics are: they seem necessary to us, but they are not necessary to him; at the same time he wants to argue that any rational God would have to opt for these laws if he wanted to create an orderly, coherent universe. As Newton’s disciple Roger Cotes complained:

He who thinks to find the true principles of physics and the laws of natural things by the force alone of his own mind, and the internal light of his reason [i.e. Descartes]; must either suppose that the world exists by necessity, and by the same necessity follows the laws proposed; or if the order of nature was established by the will of GOD, that himself, a miserable reptile, can tell what was fittest to be done.³⁰

How did Descartes get into this tangle? Because he was trying to establish laws that were, in Montaigne’s terms, truly universal, laws which would work both for a universe created by an omnipotent God and for an Epicurean universe created out of chaos by the random concatenation of atoms – hence his failure to use the term ‘laws’ for local effects.^vi

Descartes’ conception of the laws of nature was deeply influential. In the Principia, Newton, like Descartes, has only three laws. He took the view that Kepler’s principles of planetary motion (which Kepler had never called laws) were, as propounded by Kepler, merely statistical regularities; they acquired law-like status only when they were shown, along with Galileo’s law of fall, to derive of necessity from a genuinely universal principle, that of gravity.³¹ (Newton evidently hesitated over whether to call gravity a law, as it does not correspond to the three Cartesian laws; he does call it a law in the Optics, but not in the Principia.) Boyle, too, evidently thought there was only a small number of ‘more Catholic laws’, and these were the laws of nature properly so called.

But Bacon had advocated a different approach. Beneath a single supreme law (he called it the summa lex, the fundamental law, but he never worked out what it was), he had sought other subordinate laws (which he sometimes thinks of as ‘clauses’ within the overarching law), for, after all, even Montaigne allowed for the existence of municipal laws: the law of heat is an example Bacon gives, which would define the essence of heat in all its various manifestations; while Lucretius had discussed the law of magnetism. This approach opened the way to a multiplication of laws: Boyle’s hypothesis regarding gases (which he never called a law) could now be counted as one. We see this looser attitude already at work in Walter Charleton, where there are lots of other laws apart from the three ‘General Laws of Nature, whereby she produceth All Effects’, such as ‘the Laws of Rarity and Density’ and ‘the setled and unalterable Laws of Magnetical Attraction’.³² It is this more easygoing Lucretian, Baconian, Charltonian approach which eventually became that of the Royal Society and of eighteenth-century science, in contradistinction to the far bolder one of Montaigne and Descartes.³³

§ 5

Descartes and his followers, who were the first to emphasize the idea of laws of nature, were faced with a series of theological difficulties; despite this, they argued that their approach was easier to reconcile with Christianity than was Aristotelianism, as Aristotle had believed the universe to be eternal and had not believed in personal immortality. There were four particular pinch points.

First, how can the soul be made to fit into a mechanistic universe? Descartes made a strict separation between mind and matter: mind was immaterial and immortal, so that mind’s relationship to the sensory world of space and time was inherently problematic. Descartes solved this problem as best he could by claiming that mind acted on the body through the pineal gland. As a result, mind became ‘the ghost in the machine’.³⁴

Second, what is the role of God in the making of the universe? Descartes was willing to envision a universe where God set up the initial conditions and then left the machine to assemble and run itself. Others, though, argued that it was quite clear that general laws of the sort described by Descartes could never produce the perfect design that one could find in the paw of a dog. Descartes never compared the universe as a whole to a man-made machine because he had no intention of saying that the universe had been deliberately designed and constructed in the way that a man-made machine is. Robert Boyle, on the other hand, insisted that was precisely how one had to think of the universe: following Kepler, he compared the universe to a clock, and so compared God to a clockmaker. Descartes’ universe is an automaton, but (at least, potentially) a self-fabricating automaton. The Cartesian universe is not made for man;³⁵ the Boylean universe is. We are entitled to feel at home in Boyle’s universe, even if it is a mechanical apparatus; it is not clear that an immortal, immaterial soul has any business feeling at home in Descartes’ universe.

Third, how do laws of nature work as causes? It is arguable that 2 + 2 will equal 4 in any universe; and, surely, levers and balances would work in the same way in any universe. But need the angle of reflection be equal to the angle of incidence in any universe? Might there be a universe in which Descartes’ third law of nature actually holds? If laws of nature are something less than mathematical truths and something more than perceived regularities, then it would seem clear that they exist only because God has chosen to make them apply. This is voluntarism, and it seems to follow naturally from the idea of laws of nature. There is a puzzle here, because the standard alternative to voluntarism is rationalism, and a rationalist would hold that the laws of nature, like the laws of mathematics, exist because they are necessary. On most questions Descartes is a rationalist, but as far as the laws of nature are concerned he appears to want to have it both ways.

A related question is: What is God’s role in causation? Has he simply set general rules in place, or does he step in on each and every occasion to ensure the rule is applied? On my keyboard, if I press the shift key I cannot then type a lower-case letter. There’s no choice involved: the letter has to be upper case. A choice made by the manufacturer when the computer was designed determines this, and nothing can change it now. On the other hand, almost every time I type the letter ‘Q’ I follow it with the letter ‘U’; but there is no causal connection between the ‘Q’ and the ‘U’, I simply choose to make one follow the other. The claim that, just as I choose to make ‘U’ follow ‘Q’, God acts to create what look like causal connections on each and every occasion – that there are, strictly speaking, no causal connections but only temporal coincidences – is called occasionalism. It was adopted by Malebranche and other followers of Descartes, and Newton sometimes speaks as if every act of gravitational attraction is directly willed by God. You cannot be an occasionalist without being a voluntarist, and every voluntarist has taken at least the first step on the road to occasionalism.

Some historians of science want to argue that you cannot have laws of nature without voluntarism, and you cannot have voluntarism without an omnipotent creator God.³⁶ Consequently, the Greeks and the Romans were incapable of formulating the idea of laws of nature, and without them they could not develop modern science. This would surely have puzzled Descartes and Newton, who found in Lucretius (in Descartes’ case) an inspiration for their own ideas, or (in Newton’s case) a prefiguration of them. The idea of an omnipotent creator God may help in formulating a theory of laws of nature, but it seems wrong to claim that it is a necessary precondition.

This brings us to our fourth and final problem: Does God override the laws of nature? Boyle was happy to argue that God makes miracles happen, and that in doing so he breaks his own laws. But Galileo described nature as inexorable and immutable, and it is very hard to understand how there can be exceptions of any sort to Descartes’ laws.³⁷ French Cartesians faced an effective censorship, and so had to be careful what they said: Descartes’ Meditations of 1641 were placed on the Catholic Church’s Index of Prohibited Books in 1663 because Descartes’ corpuscular philosophy (since, like Lucretian atomism, it denied that there was such a thing as substance or form) was held to be incompatible with the Catholic doctrine of transubstantiation (which declared that during the Mass there was a transformation in the substance of the bread and wine, even though they retained their original outward appearance).³⁸ In Protestant countries censorship was less rigorous, although there were still limits as to what could be published. Thus some disciples of Newton were prepared to follow the logic of natural law through to its conclusion and maintain that everything that happens happens in accordance with the laws of nature.³⁹ William Whiston (a pupil of Newton who was, like Newton, an Arian – that is, he denied that Christ had existed from all time and, consequently, denied the doctrine of the Trinity), for example, argued in 1696 that the Flood had been caused by the Earth passing through the tail of a comet.⁴⁰ Similarly, there must be natural explanations for the parting of the Red Sea, or the plagues of Egypt; what is evidence of divine providence is that God arranged for these exceptional events to coincide with the need for them.

Protestants had long argued that the modern miracles reported by Catholics were simply (where they were not frauds) misunderstandings of natural events; the same arguments were now applied to the Bible itself. It was obviously safer to apply such theories to the Old Testament than to the New but, by implication, Christ’s miracles, too, even the Resurrection, were to be understood as natural events, which wonderfully coincided with the need for an appearance of divine intervention. So, too, when God answers petitionary prayer, it was argued, he does not alter the course of events in order to answer the prayer, but, as an omniscient God, he knows in advance that the prayer will be followed by an event which looks like an answer. Miracles and answers to prayer thus become entirely subjective experiences; objectively, there is nothing there, apart from a coincidence. Montaigne had already asked, ‘How many things are there which we call miraculous or contrary to Nature? All men and nations do that according to the measure of their ignorance.’⁴¹