The Online Library of Liberty

A project of Liberty Fund, Inc.

• Key to Abbreviations and References
• General Introduction
• Introduction
• The History of Astronomy
• The History of the Ancient Physics and the History of the Ancient Logics and Metaphysics
• The History of Astronomy
• Section I: Of the Effect of Unexpectedness, Or of Surprise
• Section II: Of Wonder, Or of the Effects of Novelty
• Section III: Of the Origin of Philosophy
• Section IV: The History of Astronomy
• Of the External Senses
• Of the Nature of That Imitation Which Takes Place In What Are Called the Imitative Arts/of the Affinity Between Music, Dancing, and Poetry
• Of the Affinity Between Music, Dancing, and Poetry
• Of the Affinity Between Certain English and Italian Verses
• Contributions to the Edinburgh Review of 1755–56 Review of Johnson’s Dictionary a Letter to the Authors of the Edinburgh Review
• Review of Johnson’s Dictionary
• A Letter to the Authors of the Edinburgh Review.
• Appendix: Passages Quoted From Rousseau
• Preface and Dedication to William Hamilton’s Poems On Several Occasions
• Account of the Life and Writings of Adam Smith, Ll.d. From the Transactions of the Royal Society of Edinburgh [read By Mr Stewart, January 21, and March 18, 1793]

The History of Astronomy

The importance of this essay to modern scholars lies mainly in the preamble and the first three sections; these contain a statement and elaboration of the chief ‘principles’ that Smith believed to ‘lead and direct philosophical enquiries’. The History of Astronomy sensu stricto, that begins only in Section IV, is of interest partly as an indication of contemporary knowledge of the subject, but mainly for the incidental remarks made by the author in pursuance of his central aim. Though acceptable to a modern historian in its main lines, it contains so many errors of detail and not a few serious omissions as to be no longer more than a museum specimen of its kind. This is not to deny its high merit for an age when systematic study of the history of the sciences was in its infancy. But by 1758 a student would have been better advised to read Jean–Étienne Montucla’s Histoire des mathématiques (written incidentally in the enlightened spirit characteristic of the young Adam Smith) which by 1802 had been revised and extended by Jérôme de Lalande. The first history of astronomy still used as an important work of reference was completed by Jean–Baptiste–Joseph Delambre in 1827.

In any attempt to assess the success of Smith’s enterprise we are met at the outset by his inconsistent and ill–defined terminology ‘philosophy is the science . . . Philosophy . . . may be regarded as one of those arts . . .’ (both in Astronomy, II.12). In fact the terms philosophy, physics, arts, sciences, and natural philosophy are used almost indiscriminately. In this of course he was not alone: Hume (Treatise of Human Nature, Introduction) speaks of ‘philosophy and the sciences’, which seems to promise a distinction more in line with modern usage; but by including Natural Religion and Criticism among the ‘sciences’ he introduced a possible source of confusion. The actual words ‘natural science’ in the sense of an ‘inquiry by reason alone into all things in the natural kingdom of God’ were first used by Thomas Hobbes in Leviathan; but ‘natural philosophy’ was preferred (though not in the restricted sense still current in the Scottish universities) throughout the seventeenth and eighteenth centuries. The first demarcation between ‘science’ and ‘art’ is attributed by the Oxford English Dictionary to Richard Kirwan: ‘Previous to the year 1780 mineralogy tho’ tolerably understood as an art could scarcely be termed a science’ (1796). James Hutton about the same time wrote that ‘philosophy must proceed in generalising those truths which are the objects of particular sciences’. In respect of the recent blossoming of the so–called ‘social sciences’ the failure of English to distinguish the species Naturwissenschaft from the genus Wissenschaft has become even more embarrassing than heretofore.

Had Smith consistently used ‘philosophy’ to include natural philosophy, leaving it to the context to indicate whether the general term or the specific application was concerned, there could, in relation to the period, be no quarrel. When he writes (Astronomy, IV.18) ‘Philosophers, long before the days of Hipparchus [c. 140 b.c.], seem to have abandoned the study of nature . . .’ and to have regarded ‘all mathematicians, among whom they counted astronomers’ with ‘supercilious and ignorant contempt’ his usage (whatever we may think of his judgement) was in general accord with ancient and medieval practice.

In the Middle Ages the interpretation of ‘philosophy’ varied from one university to another. Roughly speaking when the trivium was enlarged under the term studia humanitatis (and in many cases the quadrivium, as such, disappeared in practice), ‘philosophy’ meant moral philosophy. Mathematics and astronomy, together with ‘natural philosophy’ (more often called ‘physics’), became mainly the concern of the Faculty of Medicine; this was especially the case in the Italian universities. But Smith’s judgement cited above follows a brief account of the epicyclic and eccentric systems of planetary motion by which ‘those philosophers (IV.9) imagined they could account for the apparently unequal velocities of all those bodies’. Who are ‘those philosophers’? It was, we are told, Apollonius (IV.8) who ‘invented’ the system and Hipparchus who ‘afterwards perfected’ it. Apollonius was a mathematician of the calibre of Eudoxus and Euclid; Hipparchus pioneered the branch of mathematics that came long afterwards to be known as spherical trigonometry and he was also among the greatest observers of all time. Most of the astronomical works of each were irretrievably lost; but to neither is any interest in ‘philosophy’ attributed—a fact at which Smith himself hints in another context (Astronomy, IV.25) where he speaks of ‘the philosophy of Aristotle, and the astronomy of Hipparchus’. The precise distinction made by the Greeks themselves will be cited in the Introduction to the essay on ‘The Ancient Physics’.

It would of course be absurd to demand precisely demarcated categories which would only stifle attempts to reveal latent relationships. But that in relation to the age of Adam Smith there are traps easily fallen into is shown by a recent comment3 that Smith referred to Isaac Newton ‘as a philosopher not scientist’. From Smith’s use of the term in this context nothing can be inferred, since the word ‘scientist’ did not exist before 1839. The use of such expressions as ‘Adam Smith’s philosophy of science’ may similarly be a source of confusion; better to risk a charge of repetitiveness and pedantry than that of circularity; each reference must be explicated on its own merits.

This caveat has an indirect bearing on the introductory sections of the Astronomy. Smith’s aim in this and the succeeding essays was to show how these histories illustrate ‘the principles which lead and direct philosophical enquiries’. Having in the first three paragraphs given the barest hint of the relevance of ‘surprise’ and ‘wonder’ to these ‘principles’ he reviews at what may seem inordinate length the influence of the sentiments of surprise and wonder on the emotions of joy, grief, panic, frenzy, etc. The modern reader, especially one unfamiliar with the pervasive significance accorded to the ‘passions’ by Smith and his contemporaries, may feel puzzled to know what all this has to do with the clearly expressed aim of the essays. Smith might have been wise to recall Bacon’s words that such observations are ‘well inquired and collected in metaphysic, but in physic they are impertinent’ (Advancement of Learning II.vii.7). But after a dozen pages the rhetorical fog lifts: the ‘surprise’ excited in the observer by the motion of a piece of iron ‘without any visible impulse, in consequence of the motion of a loadstone at some little distance from it’ and the ‘wonder’ how it came to be ‘conjoined to an event with which, according to the ordinary train of things, he could have so little suspected it to have any connection’ (II.6) establish the thesis in the clearest possible manner. The further deployment of the thesis, even if unnecessarily prolonged, displays Smith’s elegant and imaginative style at its best. Had he but set his own words ‘philosophy is the science of the connecting principles of nature’ at the beginning instead of near the end, and then avoided the trap in the ill–defined term ‘philosophy’, this section might well have ranked as the most fundamental in the whole work. Though not free from confusion, the concluding pages of this section reveal in greater emphasis Smith’s ‘principles of philosophical enquiries’. Central among these is an interpretation of causal investigation as a search for a ‘bridge’; the examples here are much more convincing. The special characteristics of this ‘bridge’ or ‘chain’ are analogy to more familiar objects, coherence, and—of special significance for the modern scholar—‘without regarding their absurdity or probability, their agreement or inconsistency with truth and reality’ (II.12). This remarkable passage is our justification for caution in speaking about what has been called ‘Smith’s philosophy of science’. For Smith himself who, as we have seen, defines ‘philosophy’ as ‘the science of the connecting principles of nature’ the term could have no clear connotation; nor could it for anyone until the term ‘science’ was restricted to what Smith is here calling ‘philosophy’. There is still no general agreement as to the range of the ‘philosophy of science’; but that it is essentially meta–science, or talk about science, would probably not be contested. Of this there could not in Smith’s time be any explicit recognition. No doubt the study of his enterprise will shed light on the nature of the problems to be talked about; but in respect of its ‘systems’ his inquiry was less about their truth than about ‘how far each of them was fitted to sooth(e) the imagination, and to render the theatre of nature a more coherent, and therefore a more magnificent spectacle, than otherwise it would have appeared to be’ (ibid.). This has certainly a modern ring about it; but a modern ‘philosophy of science’ that thus ignored the problem of truth would get rather a cold reception. It is thus less the philosophy of science than the history of the idea of the ‘philosophy of science’ that Smith’s enterprise is likely to illuminate.4

The dubious historiography and scrappy exposition of Section III—‘Of the Origin of Philosophy’—are characteristic of the ‘Age of Reason’: imaginative liveliness creates a colourful stage upon which the drama of Western culture is to take its rise. Regrettably ‘imagination’5 aided and abetted but not controlled by ‘reason’ takes command; and what was in the circumstances inevitably no more than a ‘likely story’ is presented with a degree of naïve dogmatism and assurance that would be beguiling if it had not engendered distorted attitudes in the long shadows of which we are still living. The danger of ‘conjectural history’ is thus made only too plain; justification of this rather critical assessment may most suitably wait on textual commentary.

In Section IV we are plunged rather abruptly into ‘The History of Astronomy’ proper: abruptly, since Smith has already stated that it is from Plato and Aristotle that he will ‘begin to give her history in any detail’. The highly complex and mathematically beautiful system of Eudoxus is thus made to appear fully formed like Pallas from the head of Zeus. For his purpose Smith is perhaps justified in thus proceeding; but not to emphasize the extreme unlikelihood of such a creation without a long preparation of accurate observation and critical correlation is to risk begging the whole question of the genesis of philosophical inquiry. Once launched, however, on the exposition of the ‘first regular system of Astronomy’ (Astronomy, IV.4) he moves, not indeed with complete mastery, but with a remarkable degree of precision and understanding. Since among the readers of this edition there may be some wholly unfamiliar with the rationale of this system it may be as well to give a necessarily somewhat simplified but also more concise account of it than Smith provides; to facilitate cross–reference this will be set out in a somewhat schematic form.

The celestial phenomena (appearances) were either relatively transitory (e.g. meteors) or eternal; comets, remaining visible for months, were the subjects of some controversy.

The ‘eternal’ bodies, with seven notable exceptions, were fixed in space relative to each other. The exceptions—Sun, Moon, Mercury, Venus, Mars, Jupiter, and Saturn (to give them their Latinized names)—were all called ‘planets’ or ‘wandering stars’, since their positions varied continuously both with respect to each other and to the pattern of the ‘fixed’ stars.

All the visible objects were seen to move in circles round the Earth in a time constituting a ‘day’. The various minor discrepancies among the planets were accounted for by assuming additional circular motions superimposed upon the uniform daily rotation. The ‘fixed’ stars were thus regarded as being carried round by the rotation of the ‘celestial sphere’ whose axis, since many of them periodically ‘rose’ in the east and ‘set’ in the west, was held to be variously inclined to the surface of the Earth. Contrary to the belief still held in some quarters, the ‘flat Earth’ had been generally abandoned about a century earlier, and, though reintroduced to conform to biblical cosmology, was probably never again seriously considered among men having any pretension to astronomical knowledge.

Since the Sun and Moon are seen to make a circuit of the stellar sphere once in roughly 365 and 29 days respectively, the motion of each was regarded as being compounded of that of the stellar sphere and that of a second sphere whose axis was inclined to that of the steller; in the case of the Sun the ‘equator’ of the second sphere was called the ‘ecliptic’, and the latter’s ‘obliquity’ represents the observed progressive changes in the Sun’s altitude in the course of the year. A third sphere had to be added to account for a further minor irregularity in the observed motion. The Moon’s observed motion resisted any adequate representation; it was one of the few problems that gave Newton a headache 2,000 years later.

The motions of the remaining ‘planets’ were partially accounted for by supposing them to share the daily and (approximate) annual motion of the Sun’s two spheres—the third was peculiar to the Sun. But these five bodies—and very obviously those that were believed to be always further from the Earth than is the Sun—possessed a characteristic irregularity of apparently coming to a halt, and then roughly retracing their paths to a second point before once more proceeding in the general direction. These meaningless ‘stations’ and ‘retrogradations’ of each of these planets were ‘saved’ by the ingenious device of ‘fixing’ each planet on a sphere, the poles of whose axis were also ‘fixed’ on the surface of the surrounding sphere to whose axis their axes were inclined; and at the same time supposing them to rotate in the opposite sense, each at a characteristic rate different from that of the surrounding sphere. The process could be repeated, and the inclinations and relative rates of rotation varied, to give the closest possible approximation to the ‘appearances’.

All this is set out by Smith with only relatively minor historical inaccuracies; but he does not here make clear that the ‘constant and equable motions’ reported by reliable commentators to have been demanded by Plato were in fact uniform angular motion in perfectly circular paths. Nor, though he has his own view as to the human urge to see coherence and a continuous chain in natural phenomena, does he comment on Plato’s postulates in flat opposition to the evidence of the senses, except in respect of the daily revolution. Plato discussed these questions in several dialogues, and his final ‘vision’ of the cosmos (if he did in fact ever arrive at one) is still a matter of controversy. But his guiding principle, from which he made no fundamental departure, was that the ‘visible’ heavens have the same relation to ‘things divine’ as they really exist as do geometrical figures to those ‘truths of reason’ that they are made to represent.

In proceeding from the concentric systems of Eudoxus to the excentric (and epicyclic) systems that permanently superseded it among the Greeks, Smith missed two points of fundamental importance to his ‘principles that lead and direct’ philosophical investigation. The first was that Aristotle’s addition of twenty–two spheres had nothing to do with the ‘insufficiency’ of the spheres to represent the motions; the reason was what we should call a philosophical demand for a physical coherence: the additional spheres were so intercalated as to prevent the characteristic motion of each of the planets from being transmitted to the remainder. Another serious physical discrepancy apparently first observed by Autolycus of Pitane but not by Aristotle, was the fact that no system of spheres concentric with the Earth could conceivably account for the marked changes in the apparent size of e.g. Mars and Venus, implying variation in their distances from the Earth. The contrast between ‘astronomy’ and ‘physics’ sketched by Aristotle, well known to the Middle Ages and Renaissance through the Commentaries of Simplicius, but apparently lost sight of later until stressed by Paul Duhem in his Σώζειν τὰ φαινόμενα, will be discussed more at large in the Introduction to the Ancient Physics.

The first step towards the epicyclic (and incidentally towards the Copernican) theory of planetary motion was taken by Heracleides of Pontus, who, noting the fact that neither Mercury nor Venus is ever seen far from the Sun as the latter makes its annual circuit of the heavens, put forward the hypothesis that the circular paths of the former bodies were centred at the Sun, not the Earth. A century later, when Alexandria had replaced Athens as the centre of ‘Greek’ culture, this hypothesis was extended by Aristarchus of Samos to include all the planets, of which he regarded the Earth instead of the Sun to be one. This revolutionary hypothesis, in which the diurnal rotation of the Earth (already assumed by Heracleides) was also adopted, was summarily rejected by his contemporaries. Nevertheless, since their imaginative leaps achieved the essential basis of that of Copernicus, the omission by Smith of any mention of these two men is quite unaccountable.

Though no motion of the Earth was acceptable to astronomers until the time of Copernicus, and even then but tardily, the concept of epicyclic motion (i.e. the circular motion of a body about another body itself describing a circle about a third) rapidly achieved a dominating influence and received a definitive form in the Almagest of Ptolemy (c.a.d. 150). Stripped down to the barest essentials this system was based on the following postulates:

• (i) The Earth is the ‘centre’ of the world.
• (ii) The Sun moves at a uniform rate on a circle (the ‘eccentric’) whose centre is somewhat distant from the Earth.
• (iii) The remaining planets (except the Moon) move on circles (epicycles) whose centres move on larger circles (‘deferents’) centred at the eccentric; but the planets themselves are represented as moving at a uniform rate round a separate point (‘equant’) on the side of the eccentric remote from the Earth.
• (iv) The Moon’s motion is especially anomalous.

The eccentric and epicycle had been elaborated by earlier astronomers, notably Hipparchus (c. 170 b.c.), but the equant point, concerned not with the shape but with the rate of planetary movement, was the creation of Ptolemy himself. Since their concern was to provide a mathematical model for forecasting celestial events, the Alexandrian (Hellenistic) astronomers took no account of the existence of ‘spheres’. The later Islamic astronomers, strongly influenced by Aristotelian and later ‘physics’, devised means of harmonizing epicyclic and eccentric motion with concentric celestial spheres. This mode of thought achieved its ultimate refinement in the theory of Georg (of) Peurbach. The so–called ‘Copernican Revolution’ was in fact a retrogression to ‘ancient’ principles buttressed by superior mathematical technique and the less ‘parochial’ world–view characteristic of the Renaissance. Far from being technically ‘modern’, the system of Copernicus was in some respects retrograde in the pejorative sense; this judgement does not detract from the dedication and intellectual courage of the man himself.

By one of those paradoxes that the history of science displays from time to time, Tycho Brahe, ‘the great restorer of the science of the heavens’ as Smith describes him, spent his life and fortune (aided by royal patronage on a lavish scale) in assembling the data enabling Ioannes Kepler to demolish both his own extension of the system of Heracleides and the details of the Copernican system. Tycho’s model, postulating a heliocentric system of all the planets, the Sun and Moon alone describing circles about the Earth, was mathematically equivalent to that of Copernicus, at the same time avoiding any affront to the physical prejudices of the age, still predominantly Aristotelian. Endowed with a spirit in which intense religious feeling, high poetic fancy, and unswerving intellectual integrity were combined to a degree probably unsurpassed in any man before or since, Kepler made the first and final break with the Platonic postulates of ‘equable circular motion’ for celestial bodies. It is the Sun, not the Earth, around which the planets describe the only discoverable simple curve—not a circle, but an ellipse; and it is the Sun that determines, in a degree corresponding to the harmonics of the diatonic scale, the speed with which they move in the paths appointed by God. Stripped of the overtones that Kepler himself regarded as his supreme act of praise to the living God, his three6 ‘laws’ are the basis of the modern astronomy of the solar system.

Within the limits of the available knowledge Smith’s account of the revolution in astronomical thought effected by Copernicus, Tycho Brahe, and Kepler displays remarkable understanding; there is however one misleading feature in his exposition—the statements (Astronomy, IV.29,32) that the Copernican system has no need of epicycles. It is indeed true that each of these statements is made in the context of the apparent shape of the planetary motions, but not many paragraphs later it is made clear that in order to rid his system of the ‘incoherence’ of the equant point (IV.53) Copernicus had in fact been compelled to employ a number of epicycles. One of Kepler’s earliest discoveries was that the motion of the Earth demanded just such an equant point: it is of course a mathematical dodge to represent the hitherto ‘unthinkable’ fact that the planets move faster when near the Sun than when more remote. Smith’s account is further notable for having stressed the possibly decisive nature of Galileo’s telescopic observations—the ‘rough’ surface of the Moon, the satellites of Jupiter, sunspots, and the phases of Venus—all phenomena that could ‘appeal to a wide audience’, thus enlisting a wider support for the Copernican hypothesis than Copernicus’s own dry mathematical exposition would have done. Smith’s claim that the latter ‘was adopted . . . by astronomers only’ (IV.36), though qualified on the next page, gives a misleading impression of the situation. This and some relatively minor points are more conveniently dealt with in footnotes to the text.

The confused state of astronomy during the first half of the seventeenth century was just such as to give point to Smith’s ‘principle’ that discovery is the fruit of a search for a ‘connecting chain of intermediate objects to link together . . . discordant qualities’ (IV.60)—in this case the immensity of the celestial bodies and the hardly conceivable speeds with which they are hurled round the Sun. The ‘gap’ left in the ‘imagination’ by a purely mathematical model, however subtle and however accurately representative of the facts, received expression in the full title of Kepler’s Astronomia Nova. The ‘physical or if you will metaphysical’ element in his system was supplied by a supposed magnetic ‘radiation’ emitted by the Sun as it rotated, thus maintaining the revolutions of the planets at varying speeds. ‘That doctrine,’ wrote Smith, ‘like almost all those of the philosophy in fashion during his time, bestowed a name upon this invisible chain, called it an immaterial virtue, but afforded no determinate idea of what was its nature.’ (Astronomy, IV.60.) In an age dominated by Newton’s proper rejection of ‘occult causes’ such a reaction was inevitable. But it is not the whole story. Kepler’s ‘magnetic virtue’ was more than a name; in fact magnetism was not, in the distinction made by Newton, an ‘occult’ but a ‘manifest’ quality. The fact that it is a different ‘manifest’ quality—gravitation—that was later shown to be the controlling factor between Sun and planets does not detract from Kepler’s recognition that a ‘chain’ must exist. In his second letter to Richard Bentley, Newton emphasized that ‘the cause of gravity is what I do not pretend to know’. Smith and his clear–sighted contemporaries failed to realize that the greatest creative advances in the search for the ‘invisible chain’ have seldom been free from the wildest guesses.

The ‘first who attempted to ascertain, precisely, wherein this invisible chain consisted, and to afford the imagination a train of intermediate events, . . .’ was, Smith justly states, Descartes (Astronomy, IV.61). The details of the Cartesian system fortunately do not concern us. But Smith shows remarkable sagacity in emphasizing that it was he (and not, as is still occasionally stated, Galileo) who stated three propositions that jointly imply ‘Newton’s’ First Law of Motion; that his notion of God’s conservation of the quantity of motion in the universe (IV. 61) made a notable advance towards Newton’s Second Law; and that he was ‘among the first of the moderns, who . . . took away the boundaries of the Universe’. Not surprisingly Smith nowhere shows any knowledge of the wide–ranging mathematical speculation of the fifteenth–century Cardinal Nicholas of Cues (whom Kepler called ‘divine’), nor of the limited publication of Thomas Digges’s theory of stellar distribution in depth; but his omission of any reference to the ill–supported but widely publicized ‘plurality of worlds’ affirmed by Giordano Bruno is less easy to excuse.

His lengthy treatment of Descartes in a history of astronomy, Smith claims, is justified less by his theory of the heavens that by the time Smith was writing was almost entirely abandoned, than by his demonstration that a coherent ‘system of the world’ could be based on simple mechanical principles applicable to both celestial and terrestrial bodies. This was a radical departure from the ‘natural philosophy’ still dominant in the schools: Samuel Pepys was so ‘vexed’ to discover that his younger brother, John’s, knowledge of ‘physiques’ was based on Descartes instead of Aristotle that he decided to find out ‘what it is that he has studied since his going to the University’. So far as ‘physiques’ were concerned both Samuel and John were wasting their time; for in the same year a young sizar of Trinity College in the same university of Cambridge was also giving less than satisfaction in his undergraduate studies. But within three years he was to think of ‘extending gravity to the orbe of the Moon’. Cambridge was slow to appreciate the tremendous revolution that the young Lucasian Professor of Mathematics proceeded to hatch within its walls; but a few years after its publication (1687—under the imprimatur of Samuel Pepys P.R.S.!) the elements of Newton’s Philosophiae Naturalis Principia Mathematica were being introduced to the students of the University of Edinburgh by David Gregory.

Despite the lack of any break in the narrative, it seems most probable that it was at this point (Astronomy, IV.67) that Smith’s original manuscript ended and the remainder was added at some later date (above, 7–8).

About Smith’s account of the Newtonian system, which, despite his doubts, stands least in need of correction at the present day, little need be said. It is clearly written and includes all the ‘verifications’ available by the middle of the eighteenth century. It is doubtful whether he had ever studied the Principia at that time. Voltaire’s Elemens de la philosophie de Neuton had been published in London by 1737, and, if this section was in fact written some years after the rest of the essay, Colin Maclaurin’s Account of Sir Isaac Newton’s Philosophical Discoveries would have been available to him after 1748; of course he may have been sufficiently well grounded in the qualitative aspects before leaving Glasgow. The only disconcerting feature of his account, taken as a contribution to the ‘principles of philosophical investigation’, is the facile manner in which he accepts gravitation as an adequate explanation of the mutually determined motions of the celestial bodies, simply on the grounds that it has always been ‘familiar’ to men on the Earth. Taken in conjunction with his remarks (Astronomy, IV.61) in hailing Descartes as having been the first to attempt to ‘ascertain, precisely, wherein this invisible chain consisted’, this must be regarded as a serious deficiency. It betrays a strange lack of awareness of the fact that what he saw as ‘so familiar a principle of connection, which completely removed all the difficulties the imagination had hitherto felt in attending to them [sc. planetary motions]’ (IV.67), many continental ‘philosophers’, notably Leibniz, regarded as either a miracle or a blasphemy. The root of their objections was that celestial gravitation, unlike the ‘familiar’ form, must be held to act instantaneously across immense distances. Moreover, since the planets showed no sign of slowing down as a result of external resistance, there could be no material medium to transmit the gravitational influence. Such an ‘action at a distance’ must be regarded as either an inexplicable miracle or an ‘occult’ property of matter itself. Neither ‘solution’ was acceptable: not the former, since it removed the question entirely from the realm of natural philosophy; nor the latter, since it reintroduced the ‘specific occult qualities’ postulated by the Aristotelians, which as Newton himself later remarked ‘put a stop to the improvement of natural philosophy’ (Opticks, Q.30). This fundamental dilemma, and much else of a more technical nature, was ventilated in the famous Leibniz–Clarke Correspondence first published in 1707. Newton, on whose behalf (and at the instigation of Princess Caroline) Clarke replied to Leibniz, showed his recognition of the difficulties by adding to the second edition of the Principia (1713) the famous General Scholium containing the even more famous (and misunderstood) phrase ‘Hypotheses non fingo’, and by his letters to the Master of Trinity, Richard Bentley, in one of which he explicitly denied that gravity is ‘essential and inherent to matter’. Newton was fully aware of the lack of finality in his ‘System of the World’ and returned to the question several times; but since Smith was apparently unaware of this, it would be inappropriate to enter into the inevitably long and difficult discussion here.

[3 ]H. F. Thomson, ‘Adam Smith’s Philosophy of Science’, Quarterly Journal of Economics, lxxix (1965), 218.

[4 ]For a further elaboration, see the present writer’s ‘Adam Smith and the History of Ideas’ in Essays on Adam Smith. The essay was designed to be read in conjunction with this introduction.

[5 ]On Smith’s attitude to the ‘faculty’ of imagination see below, 20.

[6 ]Really four: the first, the demonstration that the planets’ orbits, including the Earth’s, are coplanar with the Sun is unaccountably omitted from the ‘text–books’. Kepler himself never set out the laws in any systematic form.