In the preparation of this article I benefited greatly from exchanges with François Crouzet, David Kendrick, and M. M. Postan. The article reflects work going forward on a study of the world economy since the eighteenth century which has received support from the National Endowment for the Humanities.

¹ S. N. Eisenstadt has usefully summarized a great deal of historical data—mainly political, but some economic—in support of this proposition in his The Political Systems of Empires (New York: The Free Press, 1969Google Scholar, 1963, paperback edition with additional preface). I deal briefly with the economics of such systems in Politics and the Stages of Growth (Cambridge: Cambridge University Press, 1971), Chapter 2Google Scholar.

² Quoted, Wilson, Charles, Profit and Power: A Study of England and the Dutch Wars (London: Longmans, Green Co., 1957), p. 1Google Scholar.

³ See, for example, Eisenstadt, The Political Systems, p. 43.

⁴ See, notably, Perkins, Dwignt H., “Government as an Obstacle to Industrialization: The Case of Nineteenth Century China,” Journal of Economic History, XXVII (Dec. 1967), 478–492Google Scholar. A similar conventional view of the limited power and influence of the Indian merchant has also been challenged. See, for example, Narain, Brij, Indian Economic Life (Lahore: Uttar Chand Kapur & Sons, 1929)Google Scholar, Chapters III and IV; also, M. N, Pearson, “Merchants and Rulers in Mughal India,” and Spodek, Howard, “Rulers, Merchants and Other Elites in the City-States of Saurashtra, India,” papers delivered at the Association of Asian Studies 25th Annual Meeting,Chicago, Illinois,March 30–April 1, 1973Google Scholar (mimeographed).

⁵ The inherently cyclical character of what I call pre-Newtonian societies is analyzed at some length in the author's Politics and the Stages of Growth. Chapter 2.

⁶ Quoted, ibid., p. 49.

⁷ Postan, M. M., “Some Economic Evidence of Declining Population in the Later Middle Ages,” Economic History Review, 11 (1950), 221–246CrossRefGoogle Scholar.

⁸ Rostovtzeff, M., The Social and Economic History of the Roman Empire (2nd ed., Oxford: Oxford University Press, 1957), Vol. I, p. 2Google Scholar.

⁹ For relevant data, see Appendix.

¹⁰ For purposes of rough calculation, I used the total net public expenditure figures provided by B. R. Mitchell with the collaboration of Phyllis Deane in Abstract of British Historical Statistics (Cambridge: Cambridge University Press, 1962), pp. 389–391Google Scholar, and extrapolations from the national income estimates of Gregory King (1688), Joseph Massie (1759–60), and Arthur Young (1770). The outcome, to be used for only the crudest purposes of approximation, is as follows:

Jan Marczewski's data on French central government expenditure in relation to gross physical product (at current prices) fall in the same range. See “Some Aspects of the Economic Growth of France, 1660–1958” in Economic Development and Cultural Change, IX (April 1961), p. 372Google Scholar.

French GNP per capita, in a less urbanized society, was probably 20 percent below that of Britain down to the 1780's, when the gap widened. On the other hand, French GNP was absolutely much larger, given the difference in population between the two countries of about three to one, for most of the eighteenth century. The absolute levels of French expenditures were, therefore, much larger than those for Britain, despite the lower proportion they bear to national product.

¹¹ Bosher, J. F., French Finances, 1770–1795 (Cambridge: Cambridge University Press, 1970), pp. 23–24Google Scholar.

¹² Quoted, Gerschenkron, Alexander, Europe in the Russian Mirror (Cambridge: Cambridge University Press, 1970), p. 85Google Scholar. Kliuchevskii's phrase refers to Russia in the seventeenth century, under pressure from Peter's predecessors. Gerschenkron notes it as a peculiarly apt evocation of the impact of Peter's reign. For an explanation of the emergence of substantial nation states in response to the changing technology and economic exigencies of war, see Bean, Richard, “War and the Birth of the Nation State,” Journal of Economic History, XXXIII (March, 1973), 203–221Google Scholar, and subsequent discussion.

¹³ The long run calculus for policies conditioned by war and the possibility of war is, of course, quite different from the short run. For an analysis, based on this distinction examining the impact of Britain's wars from the thirteenth to the twentieth centuries, see the author's “War and Economic Change: The British Experience,” Chapter VII in The Process of Economic Growth (2nd ed., Oxford: Clarendon Press, 1960), pp. 154 ffGoogle ScholarPubMed.

¹⁴ Alexander Gerschenkron, Europe in the Russian Mirror, especially pp. 86–8.

¹⁵ In the case of Great Britain, Mulhall evidently added imports, British exports, and re-exports (but not specie exports) to derive his figures. Presumably, he followed a similar procedure in the other cases, although precious metals evidently do enter into his calculations for the Spanish colonies, Spain, and Portugal. Where his data can be checked, they are accurate approximations of the best sources available in the late nineteenth century. But they must, of course, be regarded as approximations. The British and French data, for example, accord with the official series, although it may be that the French figure for 1720 is too low; the French value figures must be deflated by about sixty percent over the eighteenth century to match roughly the British official value (volume) data. Moreover, 1780 is an awkward year, involving as it does the damping and distortion of trade caused by the American War of Independence and related conflicts. The British and American figures for 1780 are, in particular, abnormally low.

¹⁶ See, especially, Fisher, H. E. S., “Anglo-Portuguese Trade, 1700–1770,” Economic History Review, 2nd series, XVI (1963)Google Scholar, republished in Minchinton, W. E. (ed.), The Growth of English Overseas Trade (London: Methuen & Co. Ltd., 1969), especially pp. 158–63Google Scholar.

¹⁷ Johnson, Emory R., et al., History of Domestic and Foreign Commerce of the United States (published by the Carnegie Institution of Washington, D.C., 1915), Vol. I, p. 89, Table 3Google Scholar.

¹⁸ Crouzet, François, “Wars, Blockade, and Economic Change in Europe, 1792–1815,” Journal of Economic History, XXIV (Dec. 1964), 568–9Google Scholar.

¹⁹ Phyllis Deane and Cole, W. A., British Economic Growth, 1688–1959 (Cambridge: Cambridge University Press, 1969), p. 7Google Scholar. The basis for these calculations is not clear from the text, but they appear to derive from estimates of population increase in the major cities and towns.

²⁰ J. -C. Toutain, for example, has re-examined the rather unsatisfactory French data on rural-urban population in the eighteenth century (“La population de la France de 1700 à 1959,” Cahiers de L'Institut de Science Économique Appliquée, Suppl. No. 133, January 1963, pp. 48–57)Google Scholar. He accepts the traditional measurement for urban population as those living in concentrations of 2,000 or over. His figures (Table 15, pp. 54–5) show a rise in the following ranges for the urban population during the eighteenth century: 1700, 2.9–3.3 million or 15–17 percent; 1801, 4.5–6.4 million or 16–23 percent. I have arbitrarily taken the higher figure in both cases. If the English and Welsh figure for urban population were calculated also on the basis of communities larger than 2,000, it would approximate 40 percent for 1801, according to the data of Williams, R. Price, “On the Increase of Population in England and Wales,” Journal of Statistical Society, XLIII (1880), 466–467Google Scholar.

²¹ The argument here is that Adam Smith's famous perception about the relation between the widening of the market and the division of labor is not a sufficient explanation for technological change. The issue remains important, because Allyn Young perpetuated Smith's incomplete view of technological change in his famous article, “Increasing Returns and Economic Progress,” Economic Journal, XXXVIII (1928), 527–542Google Scholar, and Nicholas Kaldor has reinforced it forty-four years later (“The Irrelevance of Equilibrium Economics,” Economic Journal, LXXXII (1972), 1237–1255)Google Scholar. For further discussion, see below, pp. 574–578.

²² See, especially, Thomas, P. J., Mercantilism and the East India Trade (London: Frank Cass & Co. Ltd., 1963 edition)Google Scholar, for an account of the economic and political debates which marked the struggle to inhibit Indian imports and the loopholes which permitted the cotton textile industry to find its feet in Britain. The famous pamphlet of 1701, Considerations upon the East India Trade (probably written by Henry Martyn), had predicted that the East India trade, in products produced by cheap labor, would “be the cause of the inventions of Arts, and Mills, and Engines, to save the Labour of Hands in other Manufactures,” even under the free trade conditions he advocated. But the inhibitions imposed on Indian cotton manufactures heightened still further the incentive to learn to produce their equivalent by machine methods.

²³ A demand-supply approach to fundamental science, invention, and innovation is elaborated in the author's The Process of Economic Growth, Chapters II and IV.

²⁴ Lilley, Samuel, “The Development of Scientific Instruments in the Seventeenth Century,” Chapter VI in The History of Science: Origins and Results of the Scientific Revolution: A Symposium (Glencoe, Illinois: The Free Press, 1951), pp. 74–5Google Scholar. See also Lilley's, Men, Machines and History (New York: International Publishers, 1966)Google Scholar.

²⁵ See, notably, Musson, A. E. and Robinson, Eric, Science and Technology in the Industrial Revolution (Manchester: Manchester University Press, 1969)Google Scholar.

²⁶ “Why Was Science Backward in the Middle Ages?” Chapter II in The History of Science, p. 31.

²⁷ Gillispie, Charles E., “The Natural History of Industry,” Isis, XLVIII (1957), 398–407Google Scholar, republished in Musson, A. E. (ed.), Science, Technology and Economic Growth in the Eighteenth Century (London: Methuen & Co. Ltd., 1972), p. 126Google Scholar.

²⁸ For the author's view of the sectoral pattern of investment in modern growing economies, see The Process of Economic Growth, especially pp. 96–103, as well as The Stages of Economic Growth (2nd ed., Cambridge: Cambridge University Press, 1971), especially pp. ix–xiv; 12–16; 174–6; 184–6Google ScholarPubMed.

²⁹ The sensitivity of the British patent series for the eighteenth century to both wars and cyclical fluctuations tends to confirm that the supply curve of invention was elastic in the short period with respect to the level of demand in the private sector, although there may be some ambiguity as between the act of invention and the filing of patents. See, for example, Ashton, T. S., “Some Statistics of the Industrial Revolution in Britain,” The Manchester School, XVI (1948), pp. 214–34CrossRefGoogle Scholar, partially reprinted (including the patent series) as Chapter 3 in A. E. Musson (ed.), Science, Technology and Economic Growth in the Eighteenth Century.

³⁰ In one of the few efforts explicitly to assess the motives of inventors, of which I am aware, Shelby T. McCloy has this to say about French inventors of the eighteenth century (French Inventions of the Eighteenth Century (Lexington, Kentucky: University of Kentucky Press, 1952) p. 189)Google Scholar: “The question of the motives or incentives of the inventors is much more difficult to solve. In our own century, due in no small degree to the writings of Karl Marx, a reader might easily jump to the conclusion that the economic motive was paramount. With not a few of the inventors it must have been. It is probable, in fact, that most of them hoped to realize some financial benefit from their inventions. This is far from saying that the hope of economic gain was the paramount motive, or, indeed, that it was the original driving force. Few inventors benefited appreciably from their inventions; a much greater number squandered their inheritance and savings on their inventive activity. The largest return to most of them was a government pension, usually modest. Some received no reward whatever. As a matter of fact, some inventors were so indifferent to monetary returns that they renounced claim to economic exploitation of their inventions. In this category were Berthollet, Berthelot, Camus, and Saint-Sauveur. With difficulty the friends of Conté persuaded him not to do likewise, and only the consideration of the other members of his family moved him. Vaucanson bequeathed his collection of machines, on which he had spent much of his earnings, to the king for public display. With these men patriotism and humanitarianism burned brightly. Even more brightly burned the desire for achievement and fame; this was the dominant motive of the French inventors. Economic returns were of secondary consideration, and humanitarianism and patriotism were seldom absent.”

³¹ Shelby T. McCloy (ibid., pp. 186–8) notes that the largest group of French inventors of the eighteenth century were men and women trained through apprenticeship, who had received some instruction in the sciences, were drawn to the large cities from towns, villages and rural districts, and were from the French middle class.

³² Hessen, B., “The Social and Economic Roots of Newton's ‘Principia,’” in Science at the Crossroads, papers presented to the International Congress on the History of Science and Technology,held in London,June 29–July 3, 1931Google Scholar, by the Delegates of the U.S.S.R., especially pp. 167, 176, 182–3, and 191.

³³ Clark, G. N., Science and Social Welfare in the Age of Newton (Oxford: Clarendon Press, 1937), especially Chapter III, pp. 60–91Google Scholar. Clark (p. 86) distinguishes five groups of influences which “worked upon science from the outside: those from economic life, from war, from medicine, from the arts, and from religion.”

³⁴ de Santillana, Giorgio (ed.), Galileo Galilei: Dialogue on the Great World Systems (Abridged text edition, Chicago: The University of Chicago Press, 1955), p. xvGoogle Scholar.

³⁵ The British patent series down to 1755 is from Hoffmann, Walther, British Industry, 1700–1950, translated by W. H. Chaloner and W. O. Henderson (Oxford: Basil Blackwell, 1955)Google Scholar, Table 54, Part A, column 64, opposite p. 330, converted back to absolute numbers from his index, with 1913 = 100; from 1756 Ashton's figures are used (A. E. Musson, ed., Science, Technology and Economic Growth, p. 119). Ashton comments on the patent series as follows (p. 118): “It may also be objected that many patents were taken out by men whose hopes outran their ingenuity or practical sense, and that the high figures of the booms represent not solid progress but the mere blowing of bubbles. A glance at the names of the patentees in each of the years of high activity suggests, however, that there is something more in it than that. The list includes, for 1769, Arkwright, Watt, and Wedgwood; for 1783, Cort, Onions and Bramah; for 1792, Wilkinson, Cartwright, and Curr; for 1801–2, the Early of Dundonald, Trevithick, and Symington; for 1813, Horrocks; for 1818, Brunel and Mushet; and for 1824–5, Maudslay, Roberts, and Biddle. One could write a fairly complete history of technology for this period without mention of any other names than these.”

Bowden, Witt (Industrial Society in England Towards the End of the Eighteenth Century, New York: The Macmillan Company, 1925, p. 12)Google Scholar sets out decadal figures for patents back to 1660. The annual averages suggest the first decade of the eighteenth century was abnormally depressed, perhaps by the impact of the war of Spanish Succession: 1660's, 3; 1670's, 5; 1680's, 5; 1690's, 10.

The French data are from Shelby T. McCloy, French Inventions, pp. 192–3. McCloy drew his figures (unfortunately incomplete for the second half or the century) from the account of inventions approved by the Academy of Sciences, down to 1754, edited by Gallon; down to 1773, The Mémoires of the Academy of Sciences are used; and the records of the Institute of France, after its organization in 1794. He notes that there are discrepancies between Gallon's figures and the later Mémoires, but the discrepancies are not great. The figures for 1796–8 are obviously damped by the effects of war and political instability. Those for 1789–92 better represent the lift in French inventiveness (and industrial activity) in the prewar years, and they represent inventions from Paris alone (ibid., p. 193 n). McCloy concludes (p. 104): “Not only did the second half of the century see a larger number of inventions than the first half, but in this latter period there were many more inventions of real significance.” McCloy demonstrates well the effect of wars in the first half of the eighteenth century in damping the number of inventions approved.

³⁶ Gregory King estimated average annual income in 1688 as £ 8 Is. 4d. for Holland; £7 18s. for England; £ 6 3s. for France. For discussion of relative income per capita levels in eighteenth-century Europe, see The Cambridge Economic History of Europe, Vol. VI, Habakkuk, H. J. and Postan, M., editors (Cambridge: Cambridge University Press, 1966)Google Scholar, Chapter I, “The Growth of National Incomes,” by W. A. Cole and Phyllis Deane, especially pp. 3–6.

³⁷ Shelby T. McCloy traces out the extensive efforts of the French government to acquire British (and other) advanced manufacturing technology, as well as to generate French textile inventions, in French Inventions, pp. 90–102 and 178–85. On the very considerable expansion of French cotton production during the eighteenth century, see Pierre Léon, Congrès et Colloques, p. 178.

³⁸ See, notably, Force, J. Clayburn La, “Royal Textile Factories in Spain, 1700–1800,” Joubnal of Economic History, XXIV (Sept. 1964), 337–363CrossRefGoogle Scholar.

³⁹ Lynch, J., “The Iberian States and the Italian States, 1763–93,” in Vol. VIII of The New Cambridge Modern History: The American and French Revolutions (Cambridge: Cambridge University Press, 1965), p. 370Google Scholar. See, also, Vives, Jaime Vicens, An Economic History of Spain (Princeton: Princeton University Press, 1969), especially pp. 524–39Google Scholar, on Spanish industrial growth in the eighteenth century. Vicens Vives (p. 538) states that at the end of the century the Catalonian cotton cloth industry totalled more than 3,000 establishments and some 100,000 workers.

⁴⁰ See, notably, F. Crouzet, “England and France in the Eighteenth Century: A Comparative Analysis of Two Economic Growths,” Chapter 7 in R. M. Hartwell (ed.), The Causes of the Industrial Revolution in England.

⁴¹ H. J. Habakkuk, “Population, Commerce and Economic Ideas,” p. 42.

⁴² The quality of French inventiveness is incontestable, as is the weakness of innovation, relative to Britain. Thus, Peter Mathias (“Who Unbound Prometheus? Science and Technological Change, 1600–1800,” Chapter 1 in A. E. Musson (ed.), Science, Technology and Economic Growth, p. 81) says: “The French record of scientific growth and invention in the eighteenth century was a formidable one. Berthollet first revealed to the world the bleaching possibilities of chlorine, first isolated as a gas in 1774 by a Swedish chemist Scheele, which was followed by energetic efforts to promote its manufacture in France. A similar sequence followed with Leblanc making soda from salt and sulphuric acid. Very sophisticated work was done in the production of dyestuffs in France; with varnishes, enamels, and many other techniques and materials. Yet the difference in the rate of industrial growth based on these advances in chemistry between France and Britain in the period 1780 to 1850 was remarkable. Almost all the theoretical work on structures, stresses, and the mechanics of design in civil engineering was French. This did not appear to have much relationship to the speed of development, or even innovations in these fields, as far as economic progress was concerned. The same was true of power engineering and hydrodynamics.” For a systematic survey, see Shelby T. McCloy, French Inventions, which covers the full range of French inventiveness in the eighteenth century as well as (Chapter XII) government encouragement of inventions before and after the first French patent law of 1791.

⁴³ So far as the economy is concerned and the pool of those from which inventors and innovators might be drawn, it is the absolute size of urban populations that matters most. So far as politics is concerned, the proportion of the urban-rural breakdown can matter significantly. It may be that Britain s precocious urbanization helped tip the balance in the civil war of the seventeenth century just as France's disproportionate rural population shaped French politics down through the first half of the twentieth century, at least.

⁴⁴ Hull, Charles Henry (ed.), The Economic Writings of Sir William Petty, Vol. I (Cambridge: The University Press, 1899), pp. 263–4Google Scholar.

⁴⁵ We are left with two interesting puzzles. Neither Holland nor the American colonies suffered the political and social inhibitions of France vis-à-vis Britain. For most of the eighteenth century, income per capita was probably higher in Holland than Britain, while the American colonies expanded their population at an astonishing rate, with high and probably rising per capita incomes, quite possibly above the British level. Why did the industrial breakthrough not occur in one or the other area? The answer may lie in the fact that entrepreneurship in Holland was concentrated on holding its ground in international commerce, in all its aspects, shifting, in fact, towards finance rather than industry when the rise of Britain and France constrained its commercial possibilities; while the yields from agriculture and international commerce in the North American colonies were so attractive as to prevent a buildup of industrial inventiveness and entrepreneurship. British colonial regulations, of course, were designed to deter manufacturing development in North America in most sectors; however, as American economic history from 1783 to 1806 suggests, it was probably a marginal factor. In the vocabulary of The Process of Economic Growth, the yields related to industrial inventiveness and entrepreneurship were deficient in Holland and North America rather than the underlying propensities.

⁴⁶ This view, of an endemic process of modernization throughout Europe and the Atlantic, should be contrasted with that presented by Phyllis Deane and W. A. Cole, British Economic Growth, 1688–1959, especially pp. 82–97. Briefly put, their hypothesis is that relatively high agricultural prices at home and unfavorable British terms of trade, expanding the money value of British imports, supplied the expanding market for the domestic and foreign sale of British manufactures. I believe this is too narrow and parochial a view of the process at work throughout the world trading area. Moreover, the postwar behavior of British imports and exports, set out in the Appendix, does not conform to the Deane and Cole hypothesis when examined on a year-to-year basis; that is, British exports move promptly to higher than prewar levels and do not await a prior rise in British imports. The data suggest that endogenous processes of expansion were at work in the major British markets, a judgment confirmed by direct historical evidence on their eighteenth-century experience.

⁴⁷ Charles C. Gillispie, “The Natural History of Industry,” in A. E. Musson (ed.), Science, Technology and Economic Growth, p. 125.

⁴⁸ Burnet, I. D., “An Interpretation of Take-Off,” The Economic Record (September 1972), pp. 424–8Google Scholar; Young, A. A., “Increasing Returns and Economic Progress,” The Economic Journal, XXXVIII (1928), 534 ffGoogle Scholar.

⁴⁹ I owe the suggestion of extending Burnet's downward sloping supply curve into a phase of constant or diminishing returns to my colleague, Professor David Kendrick. For a discussion of factors leading to deceleration of output (and price decline) in a leading sector, see The Process of Economic Growth, pp. 96–103.

⁵⁰ Nicholas Kaldor, “The Irrelevance of Equilibrium.”

⁵¹ A. A. Young, “Increasing Returns,” p. 535.