C H I N A , E U R O P E , A N D
Kenneth Po...
C H A P T E R O N E32
advantaged in human capital), more productive, or otherwise heirs of many
years of slowly accruing a...
E U RO P E B E F O R E A S I A? 33
with cultivating more land. In parts of eighteenth-century Bengal, for instance,
C H A P T E R O N E34
Asia, where pasture land was so scarce, but the remarkable development of
water transport in China a...
E U RO P E B E F O R E A S I A? 35
grain trade comparable to all of Europe’s long-distance grain trading; and there
must h...
C H A P T E R O N E36
Why emphasize Europe’s probable edge in housing, rather than, say, the re-
markable supply of safe d...
E U RO P E B E F O R E A S I A? 37
to somewhere between 31.6 and 34.0, and Razzell suggests that other age-
specific mortal...
C H A P T E R O N E38
them to be generally greater than those for comparable groups of northwest
Europeans until the ninet...
E U RO P E B E F O R E A S I A? 39
ignored; it is thus reassuring to find that Chinese, who lived relatively long
lives, se...
C H A P T E R O N E40
visitors in this period often remarked on how healthy the indigenous popula-
tion was.49
For many ot...
E U RO P E B E F O R E A S I A? 41
now clear that Asians (or at least east Asians) did have some control over
marital fert...
C H A P T E R O N E42
There seems, then, little reason to think that most Europeans—even northwest
E U RO P E B E F O R E A S I A? 43
emphasis from differences in actual physical destruction to a claim that the
legacy of ...
C H A P T E R O N E44
to near-constant war-making to the high cost of British labor) highlighted by
many other scholars. M...
E U RO P E B E F O R E A S I A? 45
stayed even in various technologies and continued their own patterns of both
invention ...
C H A P T E R O N E46
of iron and steel that were of a quality at least as good as anything available in
early modern Euro...
E U RO P E B E F O R E A S I A? 47
most land in food crops, and thus had caused these clever but nonetheless
insufficient i...
C H A P T E R O N E48
absence of scientific societies and Newtonian clergymen, China (and other
societies) lacked adequate ...
E U RO P E B E F O R E A S I A? 49
Later in this chapter, I will argue that the most important innovations for
creating su...
C H A P T E R O N E50
This latter scenario may well describe the situation in certain parts of south-
east Asia, where hig...
E U RO P E B E F O R E A S I A? 51
non-farm labor.88
Before this period, many industrial laborers had worked in
C H A P T E R O N E52
have to be financed out of retained earnings—and those with higher wages will
be less able to do that...
E U RO P E B E F O R E A S I A? 53
immense wage differentials are hard to find before the mid-nineteenth century,
since dif...
C H A P T E R O N E54
So even here, the “high wage/necessity” argument faces problems. None-
theless, in this restricted b...
E U RO P E B E F O R E A S I A? 55
produced very sharp price rises, which would have greatly limited the useful-
ness of t...
C H A P T E R O N E56
not rise much between 1750 and 1850, that solution had to involve trading
partners who could bring l...
E U RO P E B E F O R E A S I A? 57
North China.111
(North China is also much further south, and thus closer to
tropical pr...
C H A P T E R O N E58
learning, and coal are part of this chapter’s technology story, as is the general
setting that made ...
E U RO P E B E F O R E A S I A? 59
nally, the much weaker property rights in the colonies and the relative indepen-
dence ...
C H A P T E R O N E60
combustion in all sorts of chemical and physical processes (from brewing to
metallurgy to dye-making...
E U RO P E B E F O R E A S I A? 61
becoming more widely available thanks to road- and canal-building; but as we
shall see ...
C H A P T E R O N E62
involved—the existence of atmospheric pressure—and had long since mas-
tered (as part of their “box ...
of 40

Com base num tratamento rigoroso de um amplo conjunto de dados - de cariz económico, social e demográfico - "A Grande Divergência" traz-nos um novo olhar sobre um tema clássico da história econômica mundial. Por que razão começou no Noroeste europeu o crescimento industrial sustentado, apesar de outras regiões do globo, tanto na Europa como na Ásia Oriental, estarem tão ou mais avançadas?
Published on: Mar 4, 2016
Published in: Science      

Transcripts -

  • 1. THE GREAT DIVERGENCE C H I N A , E U R O P E , A N D T H E MAK I N G OF T H E M O D E R N WOR L D E C O N O M Y Kenneth Pomeranz P R I N C E T O N U N I V E R S I T Y P R E S S P R I N C E T O N A N D O X F O R D
  • 3. ONE EUROPE BEFORE ASIA? POPULATION, CAPITAL ACCUMULATION, AND TECHNOLOGY IN EXPLANATIONS OF EUROPEAN DEVELOPMENT T HERE IS no consensus on how Europe became uniquely wealthy by the mid-nineteenth century. However, Eric Jones’s European Miracle probably comes closest to enunciating the current “mainstream” posi- tion. Jones’s argument is eclectic, and many Europeanists would reject or question many of his claims; but several of his general propositions nonethe- less command wide assent. For our purposes, the most important of these general statements—one also found in any number of other works—is that industrialization was not the point at which European economic history de- parted from other Old World trajectories; instead, it represents the full flower- ing of differences that had been more subtly building for centuries. In fact, many scholars simply take this for granted; since Jones explicitly argues for the proposition, his work serves as a useful point of departure. According to Jones, “Europeans”1 were already uniquely wealthy before industrialization. In particular, they had vastly more capital at their disposal, especially livestock,2 which they accumulated by “holding back population growth a little below its maximum.” This in turn allowed Europeans to “hold their consumption levels a little above those of Asia.”3 Moreover, their capital stock was less liable to destruction, because Europe suffered fewer natural disasters and began sooner than other places to build with fire-resistant brick and stone.4 Thus, less of Europe’s annual surplus above subsistence was needed to offset depreciation, and its advantage in capital stock grew steadily with time, even before the Industrial Revolution. But in fact there is little evidence to suggest a quantitative advantage in western Europe’s capital stock before 1800 or a set of durable circumstances— demographic or otherwise—that gave Europe a significant edge in capital ac- cumulation. Nor is it likely that Europeans were significantly healthier (i.e., 1 It is not always clear whom Jones includes in this term; in some cases it embraces the whole continent, in others just western or even northwest Europe. 2 Jones 1981: 4–5. 3 Ibid., 14 4 Ibid., 22–35, 40–41.
  • 4. C H A P T E R O N E32 advantaged in human capital), more productive, or otherwise heirs of many years of slowly accruing advantages over the more developed parts of Asia. When we turn to comparisons of the technology embodied in the capital stock, we do find some important European advantages emerging during the two or three centuries before the Industrial Revolution; but we also still find areas of European backwardness. Europe’s disadvantages were concentrated in areas of agriculture, land management, and the inefficient use of certain land-intensive products (especially fuel wood). As it worked out, some of the areas in which Europe had an edge turned out to be important for truly revolu- tionary developments, while the particular areas in which other societies had better techniques did not. But even Europe’s technological leadership in vari- ous sectors would not have allowed a breakthrough to self-sustaining growth without other changes that made it much freer than other societies of its land base. This was partially a result of catching up in some of the land-saving technologies in which it lagged, a process that was greatly facilitated by knowledge gained through overseas empire, and partly a matter of serendipity, which located crucial resources (especially forest-saving coal) in particularly fortunate places. It was also partly due to global conjunctures. Those global conjunctures, in turn, were shaped by a combination of European efforts (many of them violent), epidemiological luck, and some essentially indepen- dent developments. (One example of the latter is China’s switch to a silver- based economy, which helped keep New World mines profitable and sustain Europe’s colonial presence during the long period before other products were developed.) These global conjunctures allowed western Europeans access to vast amounts of additional land-intensive resources. Moreover, they could obtain these resources without needing to further strain a European ecology that was already hard-pressed before the great nineteenth-century boom in population and per capita resource use, and without having to reallocate vast amounts of their own labor to the various labor-intensive activities that would have been necessary to manage their own land for higher yield and greater ecological sustainability. Without these “external” factors, Europe’s inventions alone might have been not much more revolutionary in their impact on economy and society than the marginal technological improvements that continued to occur in eighteenth-century China, India, and elsewhere. Agriculture, Transport, and Livestock Capital Europe did indeed have more livestock per person than most other settled societies, and within a European system of farming that livestock constituted such valuable capital equipment that more farm animals usually meant more prosperity. And in a few places in Asia a shortage of livestock did interfere
  • 5. E U RO P E B E F O R E A S I A? 33 with cultivating more land. In parts of eighteenth-century Bengal, for instance, landless laborers were unable to take advantage of empty, fertile land because they lacked access to plow animals; but this was less because of an absolute shortage of livestock than because landlords, fearing the loss of their labor force, took care to monopolize the necessary animals.5 The very fact that un- used land was still plentiful makes it unlikely that Malthusian pressures were to blame for people not having livestock. In some other Asian societies, human populations had reached densities at which they restricted the availability of livestock; but nothing in those cases indicates that a shortage of farm animals inhibited agricultural production. Indeed, had a shortage of animals been a crucial problem, it is hard to see why at least larger, wealthier farmers would not have raised and used more of them; yet for the period in which we have reasonable data, there is no observable difference between large and small North China farms in animal power used per acre.6 Moreover, what by European standards was a tiny number of ani- mals sufficed to do all the work needed to keep virtually all usable land under cultivation. Moreover, in this region—with a crop mix and ecology more like Europe’s than that of the rice-growing south—my best estimate is that even with relatively few draft animals, late eighteenth-century Chinese placed con- siderably more—and higher-quality—manure on the soil than did their Euro- pean contemporaries.7 The resulting yields supported an exceptionally dense population for a dry-farming region,8 at living standards that, as we shall soon see, were probably comparable to that of western Europe. Meanwhile, in the rice regions of Asia, even smaller numbers of draft animals coincided with the highest agricultural yields in the world; rice farming simply does not require as much animal power, and post-harvest operations also require much less power than does making wheat flour.9 Subtropical and tropical regions else- where, such as Meso-America, also supported dense populations with few or even no plow animals. If even with more animals European farming was not exceptionally productive, it is hard to see this as a crucial advantage. Of course, plow animals can also pull other loads. The huge preponderance of land transport in preindustrial Europe probably results in part from the availability of so many farm animals, who had to be fed everyday but were only needed part-time for farming. Did Europe then have a crucial advantage in capital equipment for land transportation? Perhaps so, compared to east 5 Van Schendel 1991: 42; Marshall 1987: 7, 23. 6 Huang 1985: 145. 7 For calculations, see appendix B. 8 The population figures in Huang (1985: 322) for Shandong, for instance, give us 400 people per square mile circa 1750—supported without net food imports—versus roughly 160 even for the Netherlands (based on McEvedy and Jones 1978: 62–63), with the help of substantial food imports. 9 Bray 1984: 48, 198–200 (comparison with Europe); Palat 1995: 60 (on milling).
  • 6. C H A P T E R O N E34 Asia, where pasture land was so scarce, but the remarkable development of water transport in China and Japan surely offset this and represented an at least equally valuable form of capital in transport; east Asia’s overall advantage in transport was noted at the time by Adam Smith.10 And in parts of Asia where, as in Europe, there was lots of meadow and grassland, rural transport was probably just as highly developed. The enormous bullock trains of north India, sometimes including 10,000 beasts,11 are a powerful, if anecdotal, example. Quantitative estimates are fraught with many uncertainties, but what we can piece together suggests that the animal-borne freight-hauling capacity of eighteenth-century north India was not wildly different, on a per-person basis, from Werner Sombart’s estimate for Germany in 1800.12 And both China and India had long purchased warhorses and some other livestock from central Asia, which had enormous amounts of pasture. After 1700, the Qing dynasty controlled much of this territory and bred its own warhorses. Had the Chi- nese needed to import other animals, this would have been ecologically feasi- ble, too.13 Nor do we see other signs of a shortage of transport capital in Asia. Such a shortage would presumably inhibit marketing, particularly of bulky goods such as grain. Yet in one of the most crowded societies of all—China—the share of the harvest that was marketed over long distances seems to have been considerably higher than that in Europe. Wu Chengming has conservatively estimated that 30,000,000 shi of grain entered long-distance trade in the eigh- teenth century,14 or enough to feed about 14,000,000 people.15 This would be more than five times a generous estimate of Europe’s long-distance grain trade at its pre-1800 peak16 and over twenty times the size of the Baltic grain trade in a normal year during its heyday.17 Furthermore, Wu’s figure includes only the largest of many grain-trading routes in China and uses cautious estimates even for those. He omits, for in- stance, Shandong province, which had a population of about 23,000,000 in 180018 —slightly larger than that of France—and was neither particularly com- mercialized nor particularly backward. It imported enough grain in an average eighteenth-century year to feed 700,000–1,000,000 people—more than the Baltic trade fed—and exported roughly the same amount.19 Thus, if we treat the grain entering and exiting this nation-sized piece of China as the equivalent of “international trade” in Europe, we find that this one province engaged in a 10 Smith 1937: 637–38. 11 Habib 1990: 376–77. 12 See appendix A. 13 See, for instance, Gardella 1992b: 101–2. 14 Wu 1983: 277. Ond shi was approximately 103 liters; a shi of rice weighed about 160 pounds. 15 Perkins 1969: 297–307; Marks 1991: 77–78. 16 Braudel 1981: 127. 17 Jones 1981: 81; DeVries 1974: 170. 18 Huang 1985: 322. 19 Xu Tan 1995: 86.
  • 7. E U RO P E B E F O R E A S I A? 35 grain trade comparable to all of Europe’s long-distance grain trading; and there must have been quite a bit of grain trading within the province as well, since even this volume of imports could not have met the demand from its urban areas (not to mention its cotton and tobacco growers). Nor was China unique. Many cities in various parts of Asia (and probably one or two in precolonial America) were larger than any European city before eighteenth-century London, and several were larger than London as well. It has been estimated that 22 percent of Japan’s eighteenth-century population lived in cities, versus 10–15 percent for western Europe;20 and the Malay archi- pelago, though sparsely populated overall, may have been 15 percent urban.21 Many of these cities—as well as some in south Asia and the Middle East— were heavily dependent on long-distance shipments of bulky foods. Overall, then, it seems very hard to find evidence of a European advantage in transportation. A last possibility would be that European animals provided a crucial difference by providing power for industrial activities, such as turning millstones. But the rice-eating parts of Asia needed less milling to begin with, since rice (unlike wheat) was often eaten without being turned into flour. When rice was to be pounded into flour, this was generally done in very small quan- tities at a time, but not for lack of animal power; rather, it was the nature of rice itself, which spoils very rapidly once unhusked, which called for hand- processing small daily amounts.22 Moreover, most mills and other industrial facilities, whether in Europe or Asia, were small; they also took many days off due to limited demand, customary restraints such as holidays, and other short- ages (e.g., of fuel for forges). Thus, large numbers of animals were not gener- ally needed, and there is nothing to suggest that a shortage of animal power was a significant brake on industrial development anywhere. So if Europe’s animals made a difference, it would not have been as a “cap- ital good,” but only as an item of consumption: i.e., as a source of protein for which other areas had no adequate substitute. Europeans certainly ate more meat and far more dairy products than most peoples in Asia. But this advan- tage was declining, not growing, in the early modern period, and doing so rapidly: meat consumption in Germany, for instance, fell by about 80 per- cent between the late Middle Ages and 1800.23 Furthermore, meat was not an irreplaceable source of protein: many Meso-Americans and North Americans seem to have gotten the most important amino acids in meat from corn, beans, and squash, and east Asians from bean curd. More generally, any argument based on one aspect of diet—or one other feature, such as having more brick and stone buildings—is shaky. How are we to decide which differences constitute being “ahead in standard of living”?24 20 Smith 1958: 68. 21 Reid 1989: 57. 22 Bray 1984: 53; Palat 1995: 60. 23 Braudel 1981: 196. 24 Jones 1981: 7.
  • 8. C H A P T E R O N E36 Why emphasize Europe’s probable edge in housing, rather than, say, the re- markable supply of safe drinking water in much of Japan, China, and south- east Asia?25 Or the greater comfort and durability of cotton, which was avail- able to even the poor through most of Asia and preferred by even the rich in Europe once it became available? The only definitive answer would be that Europeans’ particular mix of material goods made them healthier, longer- lived, or more energetic—and our admittedly limited evidence indicates no such thing. Paul Bairoch, projecting backward from twentieth-century data, has generated estimates of per capita income for most of the world circa 1800. In his figures “Asia” as a whole is very slightly behind western Europe but ahead of Europe as a whole, and China remains ahead of even western Eu- rope.26 But Bairoch’s exercise is also fraught with many difficulties. Rather than rely on the single number he generates for each economy, I will build my own case for the economic “ordinariness” of eighteenth-century Europe, pro- ceeding topic by topic. Living Longer? Living Better? Life expectancy at birth in England (perhaps the most prosperous part of Eu- rope) was about thirty-two in 1650 even for the children of peers; it passed forty only after 1750.27 John Knodel finds life expectancy for the people of fourteen west German villages to have fluctuated between thirty-five and forty throughout the eighteenth and nineteenth centuries, a figure which, as we shall see, is higher than nineteenth-century aggregates for larger German popula- tions.28 The massive study by Wrigley and Schofield of English villages gives life expectancies in the mid- to high thirties throughout the eighteenth century, climbing to forty in the nineteenth century and not going much above that level until after 1871.29 Although these figures suggest that England as a whole had, rather surpris- ingly, a life expectancy only slightly worse than that cited by Stone for the sons of peers, we should not leap to that conclusion. Other scholars suggest that Wrigley and Schofield have not fully corrected for the underreporting of births and deaths among the common folk before 1780; this would increase the distance between commoners and the better-documented peers by decreasing the calculated life expectancies of ordinary folk. Peter Razzell estimates that true English infant mortality between 1600 and 1749 may well have been anywhere from 60 percent to 100 percent higher than Wrigley and Schofield’s numbers indicate.30 This alone would depress a life expectancy at birth of 37 25 Hanley 1997: 104, 110–11, 117, 119–20; Reid 1988a: 36–38, 41. 26 Bairoch 1975: 7, 13, 14. 27 Stone 1979: 58. 28 Knodel 1988: 68–69. 29 Wrigley and Schofield 1981: 230, 708–13. 30 Razzell 1993: 757–58.
  • 9. E U RO P E B E F O R E A S I A? 37 to somewhere between 31.6 and 34.0, and Razzell suggests that other age- specific mortalities should also be adjusted upward, especially for the earlier part of this period.31 Life expectancy for France’s much larger population was significantly lower: between 27.5 and 30 at birth for both sexes between 1770 and 1790.32 Figures for slightly later (1816–60) in various parts of Germany are roughly comparable to those for France: 24.7 in east and west Prussia, 29.8 in the Rhine province, and 31.3 in Westphalia.33 Various groups of Asians seem to have lived at least as long as these western Europeans. Hanley and Yamamura estimate mean life expectancies at birth in two Japanese villages of 34.9 and 41.1 for males and 44.9 and 55.0 for females in the late eighteenth and early nineteenth centuries.34 Smith, Eng, and Lundy calculate the total life expectancies of those who made it to age one in a well- documented eighteenth-century village as 47.1 for males and 51.8 for fe- males.35 Thus it appears that rural Japanese—a group that does not include aristocrats, who were legally required to live in castle towns—lived at least as long as Europeans, and probably longer. Chinese longevity is less impressive but still quite comparable to Euro- pean longevity. The case can be made for other Asian populations as well. Tel- ford’s study of genealogies from a relatively prosperous area suggests a mid- eighteenth-century life expectancy of 39.6 at birth, though with a decline to 34.9 (still comparable to estimates for England) by the early nineteenth cen- tury.36 Lee and Campbell, working with unusually good data for a village in rural Manchuria in the years 1792–1867, arrive at an expectancy of 35.7 for one-year-old males and 29 for one-year-old females.37 These figures are a bit lower than Telford’s numbers for the mid-eighteenth century, though for fe- males they may be depressed by what seems to have been an unusually strong preference for sons in this population. At any rate, they are still comparable to those for prosperous parts of rural Europe. Lavely and Wong find many rea- sons to doubt any late eighteenth-century decline in life expectancy; they also assemble measures of Chinese life expectancy from various studies and find 31 Ibid., 759–63; calculations of adjusted life expectancies are my own. 32 Blayo 1975: 138–39. 33 Nipperdey 1996: 89. 34 Hanley and Yamamura 1977: 221–22. 35 Smith, Eng, and Lundy 1977: 51 give figures of 46.1 and 50.8 in the table, which is of future life expectancy. It should also be noted that here, as in recent Chinese studies, the finding of high rates of infanticide (often not due to terrible scarcity) produces an unusually large gap between life expectancy at birth and at age one, and makes the latter a better guide to overall conditions. Anyone unable to believe that infanticide could be anything but a desperate measure should not only consider its prevalence among well-to-do Chinese and Japanese, but the persistence among well-off urban Europeans of sending their infants to rural wet-nurses long after it was clear that this greatly increased infant mortality. 36 Telford 1990: 133. 37 Lee and Campbell 1997: 60, 76–81.
  • 10. C H A P T E R O N E38 them to be generally greater than those for comparable groups of northwest Europeans until the nineteenth century.38 Recent studies of the Qing imperial lineage—perhaps the best-documented large premodern population anywhere, and not a universally well-to-do one— present a mixed picture, but one that generally supports the idea that “Chi- nese”39 lived as long as western Europeans. Life expectancies at birth seem low, in part because of very high rates of infanticide—perhaps as many as 25 percent of female newborns were killed, with the rate peaking in the eigh- teenth century.40 (Infanticide was widely used as a family planning device, and the unusually good records for this population make it possible to see just how widespread it was.) However, life expectancies for those who made it to age one were at or slightly above forty by the late eighteenth century,41 which makes them quite comparable with the best-off among the western European populations discussed above. That Chinese life expectancies were comparable to European ones can also be inferred from other demographic data. As we shall soon see, China’s birthrates appear to have been lower than European ones, while its population growth rate was first higher (1550–1750) and then comparable (both China and Europe roughly doubled 1750–1850): this is only possible if Chinese death rates were also lower than European ones. (Europe had more emigration, but not enough to make an important difference until the end of this period.) Granted, further research may suggest higher birth- and death rates for China than those found so far (especially if we find good data for poorer parts of the country), but our European data are also drawn dispro- portionately from relatively prosperous areas. The rough comparability of life expectancies in better-off parts of eigh- teenth-century China and Europe (with perhaps a slight advantage for China) are also mirrored by our scattered data on nutrition. We should not assume too close a correspondence between mortality rates and nutrition, a practice that assumes preindustrial populations had few ways of consciously influencing death rates, leaving fluctuations in available resources (and exogenous crises such as plague or war) as the main influence. Lee and Wang, for instance, have made a good case that new public health measures (e.g., the spread of smallpox invariolation), long-standing patterns of personal sanitation (using soap, boil- ing water), and changes in popular attitudes (about everything from seeking medical care to killing or neglecting certain infants) may have had more im- pact on eighteenth-century Chinese life expectancies than we would expect from research on premodern European populations. But even so, the basic Malthusian insight that per capita food supplies affect death rates cannot be 38 Lavely and Wong 1998, especially pp. 721–24. 39 The members of the imperial lineage were Manchus, but were living in China and were in many ways quite assimilated. 40 Li Zhongqing 1994: 7. 41 Ibid., 9.
  • 11. E U RO P E B E F O R E A S I A? 39 ignored; it is thus reassuring to find that Chinese, who lived relatively long lives, seem to have had relatively abundant food. Braudel finds a huge variety in European reports of calorie intake before 1800 and notes that most come from sources on the lives of the privileged; he suggests 3,500 calories per day for people doing hard physical labor (e.g., crews in the Spanish fleet) and around 2,000 calories per head for the “great urban masses.”42 Nineteenth-century English data assembled by Clark, Huber- man, and Lindert run 2,000–2,500 calories per adult male equivalent for vari- ous groups of non-farm laborers’ households, and almost 3,300 for rural farm laborers in the 1860s.43 Ming-te Pan, working backward from the rations re- ported for farm laborers in a seventeenth-century agricultural manual from the Yangzi Delta, notes that these rations would have worked out to 4,600 calories from grain alone.44 Estimates of grain consumption for the entire Chinese pop- ulation in the eighteenth-century vary, but they average about 2.2 shi of rice equivalent per day,45 yielding roughly 1,837 calories per person per day. If the age structure of the population was the same in the eighteenth century as it was in John Buck’s samples from the 1920s and 1930s,46 this would convert to 2,386 calories per adult equivalent, plus whatever nongrain consumption they had. Conversion to adult male equivalents, though desirable for comparability with England, is complicated by the fact that the differences between adult male and female consumption in both seventeenth- and twentieth-century rural Chinese data are considerably larger than in English samples; but if we use the late nineteenth-century English ratio, then our Chinese figure becomes 2,651 calories per adult male. This compares well with all but one of the various British samples, including those from the much more prosperous late nine- teenth-century, and quite far above Braudel’s estimate for the “great urban masses” of Europe as a whole.47 Data for southeast Asia are extremely spotty, but a parish register from early nineteenth-century Luzon suggests a life expectancy at birth of forty-two.48 Other scattered evidence suggests that between 1500 and 1800, elite southeast Asians may have lived a bit longer than their European peers, and European 42 Braudel 1981: 129–30. 43 Clark, Huberman, and Lindert 1995: 223–26. 44 Pan, unpublished: 10. 45 Marks 1991: 77–78. 46 Cited in Perkins 1969: 300. 47 For England, see Clark, Huberman, and Lindert 1995: 226n. 25. Pan 1994: 327 and accom- panying notes makes a reasonable case for estimating adult male consumption as double that for adult females. If this were true, Chinese consumption per adult male equivalent would be an even more impressive 3,181 calories from grain alone, but such a lopsided distribution of calories between men and women would make “adult male equivalents” a somewhat deceptive standard of comparison. Data for 1930s Shanghai, however, suggest that the grain consumption of adult fe- males was 77 percent of average adult male consumption (Shanghai shehuiju 1989: 183); this is quite close to the .733 conversion ratio used by Clark, Huberman, and Lindert with their English data. 48 Ng 1979: 56, cited in Reid 1988a: 48–49.
  • 12. C H A P T E R O N E40 visitors in this period often remarked on how healthy the indigenous popula- tion was.49 For many other areas, we simply lack data. Only in India are the calculated life expectancies that we have markedly inferior to most of those for northwest Europe: probably somewhere between twenty and twenty-five at birth circa 1800, based on shaky data from one area.50 As we shall see repeatedly, a combination of enormous variety and weak data make it particularly hard to generalize about south Asia, or even to make the sorts of statements about subregions that are possible for China, Japan, and Europe. In this case, it is particularly noteworthy that India had a much greater variety of labor regimes than the even larger (but politically more unified) Chinese empire; the range of variation seems at least as broad as it was across Europe and thus much greater than it was in western Europe alone. It would not be surprising if this led to equally large differences in income distri- bution and living conditions, even among areas with similar natural endow- ments. (This was, of course, the case in Europe, too, while in China the rela- tionship between regional ecologies and standards of living seems to have been more direct.) Meanwhile, even a life expectancy of twenty-five is only slightly below Blayo’s figure for France; moreover, a recent study suggests that the food-purchasing power of at least south Indian laborers (both agricul- tural and artisanal) in the mid-eighteenth century generally exceeeded that of the English working class.51 Birthrates If European death rates were not exceptionally low, neither were their birth- rates; and thus European families had no special advantage in preserving their patrimonies. When John Hajnal first outlined the ways in which the Euro- pean fertility regime, with its high rates of celibacy, of adolescents and young adults spending years away from home as servants before they could marry, and relatively late marriages, would produce birthrates lower than those in a “preindustrial demographic regime” (in which nothing was done within marriage to prevent procreation), it was widely assumed that most, if not all, of the rest of the world was characterized by such a “premodern” system.52 There were, indeed, few large societies outside Europe that had compa- rable institutions to delay marriages or depress the rate of people ever mar- ried, and comparativists looking outward from Europe were simply not pre- pared to find effective fertility control within marriage before the time it began to appear in Europe (roughly, the end of the eighteenth century). But it is 49 Reid 1988a: 45–50. 50 Visaria and Visaria 1983: 472–73. 51 Parthasarathi 1998: 79–109. 52 Hajnal 1965, 1982; see especially 1982: 476–81.
  • 13. E U RO P E B E F O R E A S I A? 41 now clear that Asians (or at least east Asians) did have some control over marital fertility. Data from Japan were the first to show surprisingly low birthrates. Much of this seems to have been an indirect—and perhaps inadvertent—result of cus- tomary arrangements in which young women were employed away from their home villages, often for years at a time, thus producing effects on fertility similar (though more pronounced) to those observed by Hajnal for Europe.53 Moreover, we also have unmistakable evidence of more direct human efforts to control the number and sex of children a family had, including abortion and infanticide, and perhaps contraception and abstinence as well. Still more re- vealing, it has become increasingly clear that these direct methods—including infanticide—were not only used as survival strategies in times of economic hardship, but as part of accumulation and mobility strategies in good times as well.54 Indeed, there is evidence that Japanese infanticide was actually more common among the well-to-do than among the poor.55 Evidence from southeast Asia is sparser and less compelling, but also strongly suggests that couples made various sorts of efforts to control fertil- ity—particularly the many families in which women engaged in migratory trade.56 Most recently, it has become clear that Chinese families of various classes, and in both good and bad times, employed a variety of strategies to limit their family size, space their children, and select their genders.57 The most widely used strategies appear to have been delaying pregancy in marriage and then preventing pregnancy after establishing a family; recent research sug- gests that this made the reproductive careers of Chinese women significantly shorter, on average, than their European peers, despite virtually universal early marriage.58 The result was birthrates per marriage and per woman that were well below those of western Europe throughout the 1550–1850 period.59 In sum, it appears that various groups of Asians were at least as able and determined as any Europeans to keep birthrates down for the sake of maintain- ing or improving their standards of living.60 Moreover, the evidence of Chi- nese and Japanese birthrates lower than European ones supports the evidence for lower death rates (and thus a fairly high standard of living), and vice versa. And if east Asians were as well- or better-off than Europeans, there is no prima facie reason to think they engaged in less household-level accumulation of capital; the next section considers arguments that various macro-level factors made Europeans’ efforts more effective, nonetheless. 53 Cornell 1996: 43–44; Hayami, cited in Goldstone 1991: 405. 54 Smith, Eng, and Lundy 1977: 107–32. 55 Skinner, cited in Goldstone 1991: 407. 56 Reid 1988a: 16, 160–62. 57 Li and Guo 1994: 1–38; Li Bozhong 1994a: 41–42, 46–52. 58 Lee and Wang forthcoming: 20–21; Lee and Campbell 1997: 90–95. 59 Li Zhongqing 1994: 3. 60 Li Bozhong 1994a: 57–58.
  • 14. C H A P T E R O N E42 Accumulation? There seems, then, little reason to think that most Europeans—even northwest Europeans—were uniquely well-off, even as late as 1750. It thus seems un- likely that their capital stock was more valuable, since it does not seem to have enabled them to produce a better standard of living for themselves. Yet another possibility suggested by Jones—that Europe’s capital stock suffered less de- preciation—deserves separate attention. There are possible scenarios in which a more durable capital stock was for a long time offset by other differences (e.g., a lower rate of gross investment or lack of skilled labor) but gradually made itself felt later when those other differences became less important. At present, though, there seems little reason to place much weight on any such scenario. Europe’s buildings may well have weathered disasters better than those in China and Japan, both of whom used less brick and stone. However, we lack adequate data to say that Europe led all other societies in this respect, or that no other compensating differences in the vulnerability of capital stock existed. Jones also argues that Europe’s most common disasters—principally epi- demics, wars, and harvest failures—mostly destroyed labor, rather than capi- tal, while earthquakes and floods, which were more common in many parts of Asia than they were in Europe, were more likely to destroy capital. But again, there are reasons to doubt that this gave Europe any significant advantage. True, populations usually recovered from all but the worst disasters within a generation or two, while some destruction of capital stock had longer-lasting effects: the centuries-long decline of parts of Iran and Iraq after thirteenth- century warfare destroyed the irrigation system may be the most famous exam- ple.61 But if the basic fabric of a society was not destroyed, even elaborate kinds of infrastructure could often be rebuilt in little more time than it took for populations to recover from epidemics. For instance, the water-control sys- tems throughout China’s Yangzi Valley were rebuilt fairly quickly once stabil- ity returned after years of warfare, plague, depression, and depopulation in the seventeenth century62 and within just a few years after comparable absolute (though not proportional) levels of destruction in the mid-nineteenth century.63 And floods and earthquakes are presumably no more likely to destroy a soci- ety’s basic fabric than is plague or drought. Thus, unless basic social order suffered more from warfare in Asia than it did in Europe—a hard case to make given the frequency of war in early modern Europe, its much lower incidence in at least China and Japan, and the limited extent of physical destruction in most southeast Asian wars64 —the argument that Europe benefited from lesser depreciation of its capital becomes very shaky. (In a later work, Jones shifts his 61 Abu-Lughod 1989: 193–97. 62 Will 1980; Perdue 1987: 211–19. 63 Bernhardt 1992: 129–34. 64 Reid 1988a: 121–28.
  • 15. E U RO P E B E F O R E A S I A? 43 emphasis from differences in actual physical destruction to a claim that the legacy of the Mongol era saddled Asia with particularly conservative regimes, a claim we will deal with later.65 ) And finally, Jones gives us no reason to think that it was necessarily more burdensome to replace ruined physical capital than to replace the human capital that Europe seems to have lost at least as rapidly as China, Japan, and perhaps Southeast Asia. Nor is there any sign that Europe’s weavers, farmers, or other workers were significantly more productive than their peers in various parts of Eurasia—as they should have been if they had either more or better capital. We have al- ready seen that they do not appear to have lived longer or better—a point not only significant in itself, but because it suggests that in competition between European and Asian goods, European manufacturers were not disadvantaged by paying higher real wages. So had their workers been more productive, they should have been able to sell their products in Asian markets. But as all ac- counts agree, European merchants had far more difficulty selling their goods in Asia than in finding markets at home for Asian goods, both for elite and mass consumption. (It is possible that despite eating just as well, Asians had less of other goods than did Europeans, but we shall see in chapter 3 that the Chinese and Japanese probably did as well.) True, the largest single source of Asian manufactured exports to Europe—the Indian subcontinent—was also one large Asian region for which many scholars believe that workers’ living standards were unusually low (as much because of very unequal income distri- bution as because of actual levels of per capita production, as we shall see in chapter 3). But Chinese textiles and other goods also found a significant Euro- American market (and not only among the rich) throughout the eighteenth and much of the nineteenth centuries.66 What about Technology? By 1850, at least northwest Europe already had a marked technological advan- tage over the rest of the Old World, and this cannot be entirely a nineteenth- century creation. But as the previous sections make clear, it seems unlikely that eighteenth-century Europeans were, on the whole, more productive than, say, Chinese or Japanese; and that means we need to carefully circumscribe claims of overall European “technological superiority” circa 1750 and target our explanations accordingly. The results admit the importance of cultural and institutional factors that helped spread a “scientific culture” but leave open, pending further research, how unique this culture was. They also tend to mini- mize the role of more specifically politico-economic factors (from patent law 65 Jones 1988: 130–46, especially 145–46. 66 Hao 1986: 28; Morse 1966: II: 61, 180, 256, 266, 322, on the size of the American market in particular, and on the relatively modest price of the cloth.
  • 16. C H A P T E R O N E44 to near-constant war-making to the high cost of British labor) highlighted by many other scholars. Meanwhile, such results increase the prominence of knowledge gained overseas for certain crucial technologies and of a set of “permissive factors” related to geography and resource availability. If Europeans were, as I have argued, not ahead in overall productivity in 1750, then it is unlikely that the average level of technology they deployed was superior; but it is more plausible that the best available technologies deployed anywhere in Europe (mostly in Britain, the United Provinces, and parts of France) for various important sectors were already the world’s best. The spread of those technologies over the next century would have then narrowed the gap between Europe’s best and average technologies and created much of the productivity advantage we see by 1850. (Clearly, for instance, Newtonian mechanics allowed Europeans in 1750 to devise some pumps and canal locks better than any in existence elsewhere, but the ubiquity of, say, Chinese canals probably gave them a continued edge in the average degree to which they had exploited the possibilities of inland waterways until somewhat later.) And even if one insists on the alternate position—that all of Europe’s advantage in 1850 sprang from post-1750 inventions—one would want to ask what basis existed for this sudden burst of inventiveness. Much of the credit for both the accelerated diffusion of best practices after 1750 and the burst of new innovations must go to elements of the “scientific culture” that Margaret Jacob and others have seen emerging, especially in En- gland, in the 150 years before 1750: increased literacy and printing, the spread of scientific societies, relatively accessible public lectures, and so on. Behind these phenomena stood a strong sense that the investigation of a mechanical nature was to be encouraged, because it offered both material benefits to the individual and a socially stabilizing alternative to two other epistemologies with political implications: dogmatic “priestcraft” and/or popular assertive- ness based on intuitive, revealed, or magical knowledge of a living nature, God, and social order.67 Some parts of this configuration were indeed unique to northwest Europe, but not all of them were. It is worth noting, for instance, that Chinese interest in the physical sciences and mathematics increased mark- edly in the seventeenth century, especially afer the Manchu conquest in 164468 and that publishers found that medical books were a particularly good way to sell lots of books, fulfill a commitment to improve the world through their work, and steer clear of the post-conquest minefields of political controversy.69 More generally, the European configuration, however fruitful it proved, did not represent the only path to technological progress. Other areas still led or 67 See especially Jacob 1988: 27–30, 58–59, 64, 77, 81–82, 89, 110, 123, 150–51, 158, 209, 223. 68 See, for instance, Henderson 1984; Kawata 1979. 69 Widmer 1996: 95–99, 103–4, 107–8, 113–15.
  • 17. E U RO P E B E F O R E A S I A? 45 stayed even in various technologies and continued their own patterns of both invention and diffusion. In many areas, various non-European societies remained ahead. Irrigation, which we have already mentioned, was perhaps the most obvious; and in many other agricultural technologies, too, Europe lagged behind China, India, Japan, and parts of Southeast Asia. A Welsh agricultural improvement society founded in 1753 took this as a truism, dedicating itself to bringing closer the day in which Wales might be “as flourishing as China.”70 Indeed, once we have seen that life expectancies were similar—making it unlikely that Europeans were vastly better nourished—the huge differences in population densities be- tween Europe and east Asia stand as impressive testimony to the size of that difference.71 To this we might add the ability of Chinese and Japanese agricul- ture to also keep up (as European agriculture stopped doing after 1800) with soaring demand for textile fibers and evidence (to be discussed in chapter 5) that even relatively backward North China was doing better at conserving the fertility of its soil better than, say, England or France. As we shall see later, Europeans groping for ways to combat deforestation and soil degradation in their tropical colonies near the end of the eighteenth century found much to learn in both India and China, but they did not apply the lessons at home in any systematic way until well into the nineteenth century. Take away the enormous amounts of extra land that Europe gained across the Atlantic (through luck, smallpox, and violence, as well as navigational and commercial skills) and it is easy to imagine Europe’s marked technological backwardness in the largest sector of eighteenth-century economies having a significance as great as what- ever advantages it had in other sectors. There were also other sectors in which late eighteenth-century Europeans still had catching up to do. In many areas of textile weaving and dyeing, west- ern Europeans were still working on imitating Indian and Chinese processes; the same was true of manufacturing porcelain. As late as 1827 and 1842, two separate British observers claimed that Indian bar iron was as good or better than English iron, and the price quoted for 1829 was less than half that of English iron in England.72 Various parts of Africa also produced large amounts 70 Bayly 1989: 80–81. 71 The difference between the population densities supported by Shandong and the Netherlands, discussed in note 7 above, is a particularly interesting example, since irrigation was not a sig- nificant factor in Shandong agriculture. On Chinese agricultural technology generally, see Bray 1984. For a non-Chinese example (which does involve irrigation) consider the fact that in the Ka- veri delta in South India, cultivators gave up about 94 percent of their output, but survived (Van Schendel 1991: 44). This suggests that one farmer could feed sixteen people (though probably not very well)—suggesting that productivity per worker, not just per acre, could be dramatically higher in parts of Asia than anything found in Europe. 72 On iron, see Dharampal 1971: 243–44, 246–47, 260; for English iron prices (and conversion from pig iron to bar iron), see Deane and Cole 1962: 222n. 5, 223 n. 1. On weaving and dyeing, see Mitra 1978: 13.
  • 18. C H A P T E R O N E46 of iron and steel that were of a quality at least as good as anything available in early modern Europe, though shortages of wood (for fuel) limited production to certain areas and could make iron quite expensive in areas distant from the forests.73 Medicine was probably not terribly effective anywhere in the world, but east (and probably southeast) Asian cities were far ahead in crucial matters of public health, such as sanitation and the provision of clean water.74 One of the few important medical advances of the seventeenth and eighteenth cen- turies—smallpox prevention—seems to have been developed independently in Europe, China, and India.75 Recent studies have suggested that at least in the area of maternal and infant health, Qing medicine—popular knowledge of which seems to have been spreading rapidly—remained superior to its Euro- pean counterpart, despite making (as far as we know) no basic conceptual breakthroughs comparable to Harvey’s work on circulation.76 The list could go on much further. Overall, then, arguments that Europe in 1750 already enjoyed a unique level of technological sophistication need significant qualification. Even in the gen- eration and use of energy—probably Europe’s most important advantage in the nineteenth century (as I will argue later)—the situation was much less clear a hundred years earlier. Smil estimates that energy use per capita was probably comparable in China and western Europe circa 1700.77 And though the effi- ciency of individual power-generating machines (from waterwheels to— soon—steam engines) was probably one of Europe’s greatest areas of advan- tage, China had an equally marked advantage in the efficiency of its stoves, both for cooking and heating.78 In retrospect, it is clear that given Europe’s nineteenth-century switch to available and abundant fossil fuels, European advances in finding ways to use heat had a greater revolutionary potential than China’s edge in capturing heat efficiently—but only in retrospect, and only with the advantage of favorably located coal. Had fuel shortages slowed Europe’s industrial growth and a breakthrough occurred elsewhere first, the wastefulness of European hearths might not appear as a minor “exception” to a story of growing technical supe- riority but as a prime example of technological weakness that had held this area back. Or had the New World not provided enormous amounts of textile fibers, European precocity in mechanizing spinning and weaving might seem more like interesting curiosities than the centerpiece of a great transformation, and we might be invoking the low level of per-acre agricultural yields in Eu- rope as a sign of serious technological weaknesses that necessitated keeping 73 Thornton 1992: 45–48. 74 See Hanley 1997: 104–5, 110–11, 119–20; Reid 1988a: 38. 75 Dharampal 1971: 141–64 on India; Du Jiaji 1994: 154–69 on China. 76 Xiong 1995 on infant and maternal care; Unschuld 1986: 183–97; Widmer 1996: 95–115, and Bray 1997: 311 on the popularization of printed medical works. 77 Smil 1994: 234. 78 See, e.g., Anderson 1988: 154.
  • 19. E U RO P E B E F O R E A S I A? 47 most land in food crops, and thus had caused these clever but nonetheless insufficient inventions to languish until they were imitated elsewhere. We will return to the crucial examples of steam and spinning—and their relationship to resource windfalls—near the end of this chapter. The point to emphasize for now is that non-European societies retained significant techno- logical advantages in many areas even in the late eighteenth century, and it was not inevitable that they would turn out to seem relatively unimportant in the long run. Nor, even once European technology began to advance faster and on a broader front, was it inevitable that this would overcome remaining weak- nesses in land management, conservation, and market extension, or do so soon enough so that development would not be directed, with lasting effects, along paths requiring precisely the sorts of labor-intensive solutions found in east Asia and a few atypical parts of western Europe (such as Denmark). Nor should we assume that these areas of non-European advantage were merely the lingering effects of once great, but now stagnant, traditions. While eighteenth-century Asia produced none of what Joel Mokyr calls “macro- inventions”—radical new ideas that suddenly alter production possibilities all by themselves—Europe produced few of these during the period from 1500 to 1750, and even during the years usually defined as the Industrial Revolution (1750–1830).79 Meanwhile, smaller technical improvements of various sorts continued to be made in many different geographic and technological areas. European dyes that briefly enjoyed a strong vogue in China were then imitated by native innovators,80 just as happened with many Asian products in Europe. In the seventeenth century, somebody discovered that a certain kind of cellar would trap enough humidity to allow cotton-spinning during the many dry months in cotton-growing North China; this innovation spread like wildfire over the next hundred-plus years, allowing a region with a population far ex- ceeding that of any European country to produce its own textiles and greatly reduce seasonal unemployment. Just as it is only the rise of fossil fuels (which made getting the most out of every ounce of combustible material much less important than before) that made the efficiency of Chinese stoves a footnote rather than a crucial fact, it is only because we know that within another cen- tury home-based textile production of any sort would come to seem “back- ward” that these cellars do not appear as a simple but vital technical break- through, disseminated at an impressive rate.81 The example of spinning cellars is also revealing because though we know extremely little about how this innovation was disseminated, we know it was. Though the design was simple, the people who needed to learn about it were among the poorest, most dispersed, and least literate members of society. That this sort of diffusion could occur fairly rapidly over a large area with the mech- anism being invisible to us should make us cautious about asserting that in the 79 Mokyr 1990: 13, 57, 83. 80 Greenberg 1951: 87. 81 Bray 1997: 217–20.
  • 20. C H A P T E R O N E48 absence of scientific societies and Newtonian clergymen, China (and other societies) lacked adequate means for spreading new and useful knowledge. At this point, we know relatively little even about scientific discussions among the elite, and, as Benjamin Elman and others have shown, these discussions were far livelier in the eighteenth century than we have generally supposed.82 Granted, the discussion proceeded mostly in classical Chinese and largely by the exchange of letters rather than in more institutionalized settings, but these letters were not really private documents and the discussions in them were wide-ranging, sophisticated, and often quite practical. Without organized sci- entific societies, the popularization of complex findings was likely to be slower than it was in England or Holland and might well have made cross-pollination between elite science and artisanal knowledge more difficult. But much re- mains to be learned about the possible contribution of vernacular publications in science and technology, especially now that we have become aware of a lively trade in vernacular medical texts (admittedly a more prestigious subject than other kinds of science or technology). Moreover, unlike in Europe, where these formal scientific societies were often essential to protecting science from a hostile established church, in China there was no such powerful and hos- tile body, and it is not clear why the particular kinds of institutions that de- veloped in Europe should have been the sine qua non of scientific or techno- logical progress everywhere. So rather than search for reasons why Chinese science and technology “stagnated” in general—which they did not do—we need to look at why the paths on which they continued to progress did not revolutionize the Chinese economy. By the same token, while giving full credit to the institutions that helped European science and technology advance unusually rapidly and on a broad front, we also need to think about which particular paths of development proved economically critical and look for fac- tors that allowed them to be so. To borrow Joel Mokyr’s metaphor (though with a different aim) we must compare not only the motors of technological change, but also the steering wheels—and the terrains over which different societies steered. Not only did western Europe not lead in all areas of technology, but of the areas in which they did lead, only some had long-term importance. For in- stance, western Europeans had the world’s most efficient waterwheels by this time,83 but this alone did not give the European industries that used water power a competitive edge capable of overcoming high transport costs (or high costs in other aspects of production) and conquering markets elsewhere. And at any rate, this was an advantage that could be deployed at only a limited number of sites and could not be expanded indefinitely even at those sites. The same was true of many, many other technologies, whether created in Europe or elsewhere. 82 Elman 1990: 79–85. 83 Smil 1994: 107.
  • 21. E U RO P E B E F O R E A S I A? 49 Later in this chapter, I will argue that the most important innovations for creating sustained growth were land-saving ones in one way or another, partic- ularly those associated with fossil fuels, which reduced reliance on forests for energy. But it has been far more common to argue that the crucial phenomenon was the rise of a labor-saving emphasis in European technological innovation. The common argument is that economic differences (principally the fact that western European laborers were free and allegedly received relatively high wages) caused Europeans (or in some versions of the argument, Britons) to focus their attention on labor-saving innovations, while other societies saw little or no need to economize on labor. (The reliance of this argument on Hajnal’s demographic argument and/or Brenner’s institutional one, both dis- cussed above, should be fairly clear.) The unique western European need to cut down on the use of expensive labor, so the story goes, ultimately led to ma- chinery, modern industry, and vastly improved per capita productivity and living standards, while other societies were more interested in looking for in- novations that economized on land, capital, or some specific scarce material. Thus, Europeans were not necessarily more creative, but high wage costs steered their efforts in the one direction that led to a real transformation. Ver- sions of this argument have been put forward by scholars as diverse as J. B. Habbakuk (Britain versus continental Europe), Mark Elvin (China versus Eu- rope), David Washbrook (India versus Europe), and Andre Gunder Frank (Asia generally versus Europe);84 and it dovetails with the common claim that Europe was already richer than the rest of the world before industrializa- tion. But the argument does not work, except perhaps in one or two specific industries. First there are empirical problems. As we have seen in the first half of this chapter, it seems likely that average incomes in Japan, China, and parts of southeast Asia were comparable to (or higher than) those in western Europe even in the late eighteenth century. If this is true, then the case that European manufacturers faced higher wage costs would have to rest on one of two possi- bilities. It is conceivable that the distribution of income could have been more equal in western Europe (or at least Britain, if one accepts that the Industrial Revolution began there), so that workers were receiving a larger share of a comparable average per capita income than workers elsewhere. Alternatively, a society could have had a system of unfree labor such that even though work- ers received fairly high aggregate payment for working, they received no in- cremental payment for working harder and could not seek other work if their patrons have no productive work for them to do. In such a scenario, despite what appear to be high wages, it would make more sense for elites to try to squeeze more hours of labor out of their subordinates than to invest in labor- saving technology. 84 Elvin 1973; Frank 1998; Habbakuk 1962; Washbrook 1988.
  • 22. C H A P T E R O N E50 This latter scenario may well describe the situation in certain parts of south- east Asia, where highly skilled artisans, though scarce enough that they were often well rewarded for their work, were bound to aristocratic patrons who “protected” them and monopolized their output.85 It may apply to some parts of India as well; but formally free or semi-free (if often poorly paid) artisans seem to have been more common there, at least until British rulers legislated against various techniques weavers had used to maintain autonomy vis à vis those who advanced them their working capital.86 And such a model has little relevance for most Chinese artisans even in the 1400s, and virtually none once the system of government-designated hereditary artisans collapsed in the 1500s. As we shall see in the next chapter, Chinese labor may well have been “freer” than early modern European labor; it was certainly not much less so. The bound-labor scenario might at first seem more relevant to Tokugawa Japan, in which various occupational statuses, restrictions on mobility, and hereditary patron-client relationships were supposedly fixed by edict; but as we shall see in the next chapter, the reality was very different from the statute books. The argument about very cheap wage labor is knottier. In chapter 3 we shall see some evidence that the distribution of income in Qing China and Tokugawa Japan was actually more equal than that in western Europe in gen- eral and late eighteenth-century Britain in particular. (For India, on the other hand, the bulk of the anecdotal evidence presented in chapter 3 suggests that income distribution was more unequal than it was in Europe; quantitative evi- dence is scarce, with some pointing in each direction.) However, even the east Asian evidence is far from conclusive and mostly suggests that the very top of society claimed no more of national income in China and Japan than Europe’s elite did; China and Japan could nonetheless have had a larger layer of desper- ately poor people than western Europe did, who pushed unskilled wages down to a level significantly below those in Europe. Although I see no particular reason to think that this was the case—and the anecdotal testimony of most Europeans visiting east Asia before 1840 suggests the opposite87 —the possi- bility cannot be dismissed. Moreover, there is a distinct but related—and more likely—scenario that would reconcile high living standards in Chinese and Japanese cores with wage bills lower than those confronting at least Dutch and English employers. Despite the rural location of much Dutch and English industry in the mid- seventeenth and eighteenth centuries, there is strong evidence that by this time relatively few workers in those countries moved seasonally between farm and 85 Reid 1989: 61, 69–71; Reid 1988a: 135. 86 Mitra 1978: 37–41; Hossain 1979: 324–38; Arasaratnam 1980: 259–60, 263, 265, 268, 272, 278. 87 See, for instance, Staunton 1799: II: 138.
  • 23. E U RO P E B E F O R E A S I A? 51 non-farm labor.88 Before this period, many industrial laborers had worked in agriculture at peak season, at least in the Netherlands, and earned relatively high wages for doing so. As the agricultural and industrial labor markets be- came more separated, day wages had to rise to enable what were now less fully employed workers to survive; such a wage increase indeed occurred, but at the price of increased unemployment.89 By contrast, many Chinese and Japanese handicraft workers were almost certainly less fully detached from agriculture; thus at least in theory, they could earn less for their weaving, spinning, or tile-making and still enjoy a standard of living as high or higher than their Dutch and English counterparts. Such a scenario is plausible, though far from established, and if correct, it would reconcile our other findings with a particu- larly strong incentive for at least some European employers to find ways to use less labor. (It would also mean that English employers would have had less trouble keeping their factories going all year-round than employers whose workers also farmed. Thus they would have more incentive to invest in central- ized plant and equipment.) European employers also faced the problem of relatively high food prices, which meant that even if they did not have to pay higher real wages, they did pay higher cash wages than many, if not all, of their Asian competitors.90 But even if we grant provisionally the argument that western European wages were higher than any Asian ones, there are problems with inferring that this stimulated the technological changes of the Industrial Revolution. Indeed, under early modern conditions, high wages could as easily discourage techno- logical innovation in general as it could encourage labor-saving inventions. Joel Mokyr suggests this seemingly paradoxical conclusion based on a model that seems fairly close to eighteenth-century realities.91 Assume, he says, that new technology must be embodied in new capital equipment, which must be paid for. Assume further that wages make up the bulk of most manufacturers’ costs and that there are few ex ante differences in technology large enough to give a firm or country with a higher wage bill lower total production costs for a particular product. Thus, those with higher wage bills will generally have lower profits than their competition. If—as was also generally true until well into the nineteenth century—bank financing for the purchase of new capital equipment is either nonexistent or, to the slight extent that it exists, dependent on a firm’s earnings, then any equipment embodying new technology will 88 “Relatively few” is, of course, a relative term. While DeVries and Allen, comparing the Netherlands and England to other parts of western Europe and to earlier periods, are struck by how little workers moved between proto-industry and agriculture according to the season, Sokoloff and Dollar 1997, comparing England to the United States, are struck by how many English people worked part-time in both agriculture and industry, even in the late nineteenth century. We will return to the U.S. example and its implications in chapter 6. 89 DeVries 1994a: 57–62, Allen, cited in Postel-Vinay 1994: 72. 90 Parthasarathi 1998: 101–2. 91 Mokyr 1991: 177–81.
  • 24. C H A P T E R O N E52 have to be financed out of retained earnings—and those with higher wages will be less able to do that. Thus, rather than stimulating labor-saving technical innovations, a high wage bill may just as plausibly discourage any sort of new technology. And though this model may seem counterintuitive today, it ap- pears to work for earlier eras: it has been used, for instance, to help explain why the very sophisticated and very high-wage Dutch economy was remark- ably late to adopt mechanized industry. Furthermore, though the industrialization of the last two-hundred years has generally been labor saving and capital demanding, it is anachronistic to as- sume that this was always the reason for the major innovations. The applica- tion of coal and steam power to all sorts of processes eventually led to enor- mous labor savings, but the eighteenth-century innovations that made coal usable in making iron, glass, beer, and so on were aimed at saving money on fuel (coal was cheaper than wood), not at saving labor; and the steam engines that pumped water out of coal mines did not substitute for men doing the same work so much as they simply made it possible to exploit certain mines that no number of men could otherwise have used. Other developments in glass- blowing, iron-making, and so on were not particularly concerned with saving on any factor of production—they were concerned with making a higher- quality product. If the makers of the Industrial Revolution were primarily economizing on expensive labor, they were unaware of it. In a study of eigh- teenth-century English patentees, Christine MacLeod finds that most declared the goals of their innovation to be either improving the quality of the product or saving on capital (a goal that makes more sense when we remember that unlike post-1870 technological change, the first one hundred years of the In- dustrial Revolution mostly came embodied in relatively cheap capital goods); only 3.7 percent cited saving on labor as a goal.92 And if inventors were not particularly intent on saving labor, those who judged their inventions were even less so; as late as the 1720s, it apparently counted against a patent appli- cant if he said that his machine saved labor.93 The long-run results of change were no doubt labor saving; but for an argument that high wages focused efforts in a particular direction, conscious motivations would seem to be the heart of the matter. And finally, since most of the capital goods involved were relatively low- cost ones themselves, even a producer who enjoyed a fairly low wage scale would have had an incentive to try them; indeed, it has been hard to show that low wage costs inhibit the adoption of labor-saving technology, even in our own age of much more expensive capital goods.94 (Such arguments sometimes have sometimes held up where the differences in labor costs are vast—e.g., contemporary Pakistan and Germany—but not where the wage differentials were real but not huge—e.g., Victorian Britain versus the United States. And 92 MacLeod 1988: 158–81. 93 Jacob 1988: 92–93. 94 Mokyr 1990: 166.
  • 25. E U RO P E B E F O R E A S I A? 53 immense wage differentials are hard to find before the mid-nineteenth century, since differences in national wealth were not nearly what they are today.95 ) If pre-nineteenth-century entrepreneurs were profit maximizing then the only innovations they should have passed up because of cheap labor were ones that provided only marginal labor savings anyway; to pass up something like cotton-spinning on these grounds alone, a manufacturer would have had to enjoy virtually costless labor. In chapter 2 we will see various examples of Chinese farmers spending money in order to save themselves labor, even though Mark Elvin and other proponents of the wage incentive argument would claim that Chinese manufacturers ignored labor-saving devices because Chinese labor (unlike European labor) was so cheap. But the high wages hypothesis might still be relevant for one crucial sector: cotton textiles, for which both Braudel and Frank assert its importance.96 Here there was very little ambiguity about what innovations in spinning did: they cut, perhaps by over 90 percent, the amount of labor needed to spin a given amount of yarn.97 And while such enormous savings should have been attrac- tive to employers paying virtually any wage rates, they may well have been particularly attractive to English makers of cotton textiles, who faced much higher nominal wage bills than the Indian producers with whom they com- peted for various price-sensitive markets (in west Africa, the Middle East, and especially the New World, where slaves wore the cheapest cottons). The tex- tiles that China exported in this period (and increasingly, even the ones that Jiangnan, China’s leading textile region, sold in other parts of China) were fairly high quality and did not compete primarily on price;98 but British cotton manufacturers could not possibly compete against Indian cottons in the Middle East, Africa, and the New World, unless they cut their wage bills. Of course, British textile producers could easily have failed do so and lost this battle with Indian producers; necessity does not always yield invention. And for Britain as a whole the issue of whether its textiles makers would conquer these markets need not have seemed crucial ex ante, since the East India Company marketed their rivals’ goods: even though these textile markets were quite strategic, any “necessity” operating here was a necessity for the textile producers themselves, not for “England.” (The most strategic of these markets was west Africa, since a ready supply of desired textiles was essential for buying slaves there. But at least some of the cloth needed there was expen- sive, high-quality material, and British slave traders were less concerned with the price of this cloth than with getting enough of it—first from India and only later from the mother country.99 ) 95 Lazonick 1981: 491–516; Bairoch 1975: 3–17 on the scale of differences in national income circa 1800 and the much larger gaps that exist today. 96 Braudel 1982: 522, 575; Frank 1998: 289–91. 97 Chapman, cited in Mokyr 1990: 98–99. 99 H. Klein 1990: 291–93.98 Li Bozhong 1998: 108.
  • 26. C H A P T E R O N E54 So even here, the “high wage/necessity” argument faces problems. None- theless, in this restricted but important case, it may well have some merit; it at least suggests how the patterns of world textile trade and the ways in which English manufacturers competed against Bengal in particular—which was both a low-wage economy (or at least a low cash-wage economy) to start with and one in which the East India Company used increasing amounts of violence to enforce below-market prices for textiles after 1757100 —may have intensified the search for mechanized spinning and weaving. Furthermore, it does illus- trate, among other things, how important it is to look for explanations of par- ticular innovations, rather than of “industrialization” in general, to root those explanations in the specifics of the relevant industries and in what people at the time thought certain innovations could accomplish—while also trying to choose examples that were critical to the broader phenomenon of emerging European supremacy. Armed with knowledge of how the Industrial Revolution did happen, one is tempted to look for European advantages connected to its two most important and dynamic sectors: textiles and the coal/steam/iron complex, especially the latter. And one does find some relevant European advantages, but often in surprising places. In textiles, the Chinese had long had machines that differed in just one crucial detail from both Hargreaves’s spinning jenny and Kay’s flying shut- tle.101 Thus, one could hardly say that western Europe had any significant lead in technology for this sector until those inventions were actually made. Nor can one conclude that just because the last piece needed in both cases seems simple in retrospect, its absence shows that technological innovation in China stopped altogether. Much of eighteenth-century European technology was al- most developed 150 years earlier, but the intervening wait does not indicate technological “stagnation”;102 we must remember that what now seems obvi- ous was often anything but obvious beforehand. Moreover, English textile innovations could easily also have become foot- notes to history rather than major milestones. At the time that the British pioneered major improvements in cotton-spinning, cotton was a minor fabric in Europe; the mechanization of flax-spinning and wool-spinning took sig- nificantly longer. And, as we shall see in chapter 5, there were serious ecolog- ical and social obstacles to the further expansion of either wool or flax produc- tion in Europe. Cotton came from abroad and was available only in fairly limited quantities throughout most of the eighteenth century; indeed, the increased demand for raw cotton that the new spinning technology created 100 See Mitra 1978: 46–47, 51, 63–66, 75–92, 113–15, 126–27, 14–15; for wage comparisons, see Chaudhuri 1978: 157, 273. 101 See, e.g., Mokyr, 1990: 221. 102 E.g., Hobsbawm 1975: 38.
  • 27. E U RO P E B E F O R E A S I A? 55 produced very sharp price rises, which would have greatly limited the useful- ness of this technology without the rise of cotton-growing in the American South.103 This problem can be phrased in a more general way. Histories of technology often imagine one breakthrough creating a “bottleneck” that concentrates ef- forts on a specific problem and so leads to another breakthrough, as when advances in weaving created incentives to speed up spinning. But such bottle- necks are just as often addressed by allocating more resources, without any change in techniques, and the longer that process of reallocation of resources continues, the less incentive remains to find a technological solution. (A good example is the massive increase in the number of coal miners in the late nine- teenth century, as the uses of fossil fuels for all sorts of processes soared with- out much change in the productivity of mining itself.)104 In the case of mecha- nized textile production, a bottleneck was created in the growing of cotton (and other fibers), which required the application of more land and more labor. As we shall see in chapters 5 and 6, it is unlikely that the necessary land to relieve this bottleneck could have been found in Europe. (Though sheep- raising did expand in Poland and Russia,105 it was nowhere near enough, and cotton production remained minimal.) Meanwhile, the labor that was applied to this bottleneck was largely that of African slaves: to the extent that Euro- pean labor was applied to this bottleneck, it was labor used in sailing, trading, coercing, and manufacturing (of goods swapped for slaves in Africa and for the cotton itself). As chapter 6 will show, that particular way of reallocating labor to solve this bottleneck was far more advantageous to Europe in the long run than it would have been to increase the agricultural labor force in order to grow more fiber at home, even if the land to do that had been available. (China and Japan both went this route, squeezing more food and fuel out of some land in labor-intensive ways while converting some lands from both forest and food crops to fiber-growing, but they did so at considerable long-run cost.) And while the case of cotton is unusually clear-cut, various other growing indus- tries, and the rising population’s demand for food, also created bottlenecks that were ultimately solved without using more European land or putting more labor onto that land. While Parthasarathi sees industrialization as in part Brit- ain’s way of escaping a vicious cycle of low per-acre yields⇒ costly food⇒ high cash wages⇒ competitive difficulties,106 it is well to remember that in- dustrialization alone could not solve the problem that allegedly induced the technological gains in industry unless it could also meet the agricultural needs of industries and workers. And since, as we will see, British yields per acre did 103 See, for instance, Bruchey 1967: table 2-A (unpaginated). 104 W. Parker 1984: 38; Mokyr 1985a: 107–8. 105 Gunst 1989: 73–74. 106 Parthasarathi 1998: 107.
  • 28. C H A P T E R O N E56 not rise much between 1750 and 1850, that solution had to involve trading partners who could bring large amounts of additional land into play. But still more basically, it is quite possible to imagine a huge productivity increase in cotton-spinning and weaving that did not lead to a fundamental break with the ecological constraints of the eighteenth century. The fiber needed for textiles still needed land, and competition for land among Mal- thus’s four necessities—food, fuel, fiber, and building materials—was grow- ing ever more intense in much of eighteenth-century Europe. As long as food and fuel prices rose faster than wages,107 as they did in most of eighteenth- century Europe, it is hard to see how demand for textiles could grow in- definitely—even with weaving and spinning costs falling—and the new textile technology had no clear application to other sectors. These developments in cotton textile production could easily have led to just an intensification of processes (to be discussed further in chapter 2) that the long-standing growth of rural “handicraft industries” already represented—processes that included accelerating population growth, increased pressure on the land, greater labor intensity, stagnant real wages, and probably an eventual ecological dead end rather than a breakthrough. Eighteenth-century western Europe faced serious ecological pressures (which will be discussed much more thoroughly in chapter 5). Briefly, the demographic and economic expansions of the “long sixteenth” and eighteenth centuries (especially the second half of the latter) led to massive deforestation in western Europe, with levels of forest cover and per capita wood supplies falling below even those in densely populated China, not to mention India. And deforestation brought other problems in its wake. Archaeological evi- dence from France and Germany suggests that the eighteenth century was one of the two worst in history for soil erosion; documentary evidence confirms this and adds several other deforested areas, which experienced massive dust storms, declining yields, and other signs of serious ecological stress.108 Studies of erosion in modern times suggest that it tends to be the most visible sign of a much broader set of soil problems.109 The late eighteenth century also wit- nessed an unusual weather pattern known as the “European monsoon”—a pat- tern in which unusually long droughts alternated with brief, unusually violent rains. When such rainfall came it was both unusually erosive and of little use to crops, especially since Europeans (unlike, say, Indians) did not have mas- sive irrigation systems to store and channel it. It is not clear what caused this climatic episode, but it appears more often in badly deforested areas,110 since trees moderate the seasonality of local rainfall patterns. One of the few temper- ate zone areas that has such a “monsoon” climate today is badly deforested 107 Goldstone 1991: 186; Labrousse (1984): 343, 346–47. 108 Blaikie and Brookfield 1987: 129–40, especially 138; Kjaergaard 1994: 18–22. For more details, see chapter 5. 109 Blaikie and Brookfield 1987: 139. 110 Ibid., 133.
  • 29. E U RO P E B E F O R E A S I A? 57 North China.111 (North China is also much further south, and thus closer to tropical pressure systems, than is northern Europe.) These ecological pressures did not add up to a Malthusian crisis, in which European living standards were about to collapse. On the contrary, they were brought about in some areas by rising levels of per capita consumption as well as population growth. But they did, as we shall see, pose substantial impedi- ments to further growth. Yet in the nineteenth century, while European popula- tion and per capita consumption accelerated, ecological variables stabilized. Western Europe’s forest cover stabilized some time between 1800 and 1850, after four hundred years of decline, and even increased throughout the nine- teenth century in Britain, France, Germany, and Belgium;112 erosion decreased and soil fertility stabilized or even improved; and the European “monsoon” disappeared and a more typical rainfall pattern returned.113 Clearly, then, a big part of the European achievement in the Industrial Revolution was to escape a long-standing pattern in which all growth placed significant incremental demands on the land. And with a few exceptions (such as Denmark), this achievement did not rely on using large amounts of addi- tional labor to make an acre yield more while protecting its fertility (in the manner famously described by Esther Boserup); in the late nineteenth century, labor inputs per acre even fell substantially. Yet the breakthroughs in chemis- try that today allow capital to substitute for land (and labor) to an astonishing degree (above all through using synthetic fertilizer and through making syn- thetic materials that are not grown at all) belong to the very late nineteenth and twentieth centuries. How, then, did sustained European growth become eco- logically sustainable? To understand how self-sustaining growth became possible, one must look, as E. A. Wrigley has argued, for developments that eased the pressures on the land. Wrigley emphasizes increased use of coal, which yielded far more power per unit of surface than wood ever could.114 To this I would add the adoption of New World food crops, particularly the potato, which yielded what for Europe were unprecedented amounts of calories per acre; improvements in ecological understanding and land (especially forest) management which, as Richard Grove has shown, owed much to colonial experiences; and the enor- mous resources gained by applying existing techniques to vast new territories overseas. The last of these developments was not principally technological and will be the focus of chapter 6; for now suffice it to say that the New World yielded both land-intensive products (cotton, sugar, and later grain, timber, meat, and wool) and land-restoring products such as guano. The potato, ecological 111 Chao 1973: 22–25, 30–31. 112 M. Williams 1990: 181. For some specific countries, see Darby 1956: 203–4 and compare with Cooper 1985: 139n. 2 (France) and M. Williams 1990: 181 (Germany). 114 Wrigley 1988: 80–81.113 Blaikie and Brookfield 1987: 132–33.
  • 30. C H A P T E R O N E58 learning, and coal are part of this chapter’s technology story, as is the general setting that made them so important. The potato produced far more calories per acre than existing European crops. The potato was also adopted in eighteenth-century China and Japan, but almost exclusively as a crop for the highlands, since rice already produced enormous amounts of food per lowland acre. In Europe, where grain yields were much lower (both per acre and relative to seed), the potato also conquered the lowlands in such densely populated areas as Ireland and Belgium (replac- ing 40 percent of cereal calories in Flanders by 1791)115 and, somewhat later, in much of central and eastern Europe. A less widely known factor was, like the potato, a technological advance: in the nineteenth century, Europeans began to apply principles of scientific con- servation to their forests and to understand the importance for the ecosystem as a whole of protecting trees. The path to this particular breakthrough has been carefully traced by Richard Grove. Interestingly, although this advance owed much to the application of European science—Newtonian mechanics played an important role in understanding how trees recycled water and af- fected local climates—some ideas popular in Europe were hindrances: even in the early nineteenth century, many European doctors and botanists blamed forests for disease-bearing “miasmas” and recommended clear-cutting woods as a public health measure.116 The solidification of European ecological understanding—just in time, it would appear, to help stabilize northwestern Europe117 before it suffered the fate of parts of the Mediterranean, or even northern China—was related to empire in two crucial ways. First of all, it was on tropical islands that Euro- peans were able to observe the relationships among changing land use, climate (especially desiccation), and changes in soil quality unfolding at a speed that resolved debates that they could not resolve theoretically; and it was in newly colonized parts of India (where European demand and changes in property rights produced rapid shifts in land use) that they began to see that the same dynamics could affect a continental land mass, too. Moreover, the colonial botanists, surgeons, and officials (often the same people) who worked out these relationships learned an enormous amount about how to manage ecosys- tems from south Chinese and especially south Indian practices, which were in many ways more advanced than their own. (Japanese practices may have been still better, but they were much less accessible to curious foreigners.)118 Fi- 115 Braudel 1981: 170. 116 Grove 1995: 408. 117 As we shall see in chapter 5, continental western Europe was for the most part still better forested than Britain but suffered from more serious fuel shortages and more rapidly rising wood prices in the eighteenth century because most areas lacked any equivalent to Britain’s growing use of coal. 118 On European borrowing from Indian ideas and practices (which Grove argues were “more important . . . than any set of ideas imported from outside India” (382), at least before 1857, see
  • 31. E U RO P E B E F O R E A S I A? 59 nally, the much weaker property rights in the colonies and the relative indepen- dence of colonial regimes from local property owners allowed British, French, and Dutch colonial officials to actually experiment with environmental regula- tion schemes, some of them quite radical, in a way they could not have done back home. This knowledge from overseas, once brought to Europe (and the United States) in the nineteenth century, became the basis for forestry services, for how-to books on using trees to help maintain or improve arable land, and so on.119 Thus, empire helped Europe erase its technological disadvantage in agro-forestry (through the potato, through ecology, and through numerous im- portant influences on botany120 ), providing crucial imports of knowledge along with the imports of resources that we will discuss later. There was, however, no extra-European dimension in the last of our great land-saving technological shifts: the increasing use of coal (especially in Brit- ain) both to replace fuel wood and as the basis for whole new processes. Coal was central to earlier views of the Industrial Revolution. Only cotton, iron, steel, and railways got comparable attention, and except for cotton, these other main sectors depended on coal. But more recently, coal has often been deemphasized. People have noted, for instance, that more early factories were powered by water than by coal and that most of England’s coal was used for the unglamourous and not particularly innovative tasks of home heating and cooking. E. A. Wrigley has reasserted the centrality of coal by calculating that it would have taken 15,000,000 acres of woodland (21,000,000 had he used a less conservative conversion) to match England’s annual energy yield from coal by 1815,121 but it is not obvious what this figure tells us. In the absence of the coal boom, England would not have consumed that much additional wood (nor does Wrigley say it would have) since it did not have it; nor can we say for sure that some specific number of forges would have closed, glass gone unmade, or homes unheated. The adjustments would have involved some com- plex combination of people being colder, buying more clothes, producing less iron, and so on, and we cannot be sure that particular industrial advances— much less industrialization more generally—would have ground to a halt with- out coal. Nonetheless, at least a partial return to the earlier emphasis on coal seems warranted, both for Wrigley’s reasons and for others. Water may for a time have powered more mills than coal, but it was geographically restricted, non- portable, and often seasonally unreliable. Moreover, it was no substitute for Grove 1995: 387–88, 406, 440, 471–72; on Chinese influence, see 187; on earlier periods, see 77–80. For some of the insights and limits of official understanding of ecology in China, see Dunstan 1997. On Japanese silviculture, see Totman 1989. 119 Grove 1995: 435, 463–64, 471–72, 480. 120 Morton 1981: 118–21. 121 Wrigley 1988: 54–55; for more on the conversion issue, see chapter 6, p. 276, n. 50.
  • 32. C H A P T E R O N E60 combustion in all sorts of chemical and physical processes (from brewing to metallurgy to dye-making), nor in the transport revolution that gave such a boost to the division of labor. In the critical iron sector (and thus also steel, railways, and so on) it is hard to see what alternative to fossil fuels could have been found. True, Hammersley has shown that—contrary to some earlier claims—England’s iron industry in the 1660–1760 period did not contract, and probably was not critically short of affordable fuel: he estimates that forest covering 2 percent of the land of England and Wales would have sufficed to supply England’s iron industry in this period.122 But by the end of the eigh- teenth century, only 5–10 percent of Britain was forest.123 Thus even under ideal conditions, the maximum possible output of charcoal pig iron in Britain would have been roughly 87,500–175,000 tons; but by 1820, actual British iron output reached 400,000 tons.124 And aside from needing some wood for other purposes, it was not feasible to mobilize all wood for charcoal iron- making. Forges also needed to be close to both iron and water power (to drive the bellows), and charcoal for iron production could not be transported more than ten to twelve miles (preferably under five): the furnaces needed large chunks of charcoal, but it tended to break into small bits (or even dust) when moved very far.125 So while Hammersley does show that iron production at 1760 levels did not face an “energy crisis”—and a fortiori that deforestation did not cause the breakthrough to coal-based iron—the same figures show that the iron industry’s further growth did require coal. In most other British industries, development of coal-based processes came earlier than it did in iron-making126 and thus substantially predates the enor- mous steam-engine-powered expansion of coal output. Thus the coal/steam engine boom could not have caused those innovations, but that does not make it irrelevant to the growth of those industries. Even if coal was mostly used for home heating, fuel for industry would have been far more expensive had less coal been available. Granted, real English charcoal prices seem to have stabi- lized in the 1700–1750 period after rising sharply for 1550–1700 (though all wood and charcoal prices must be treated with considerable caution).127 And even before steam engines allowed deeper mining, cheap coal was gradually 122 Hammersley 1973: 602–7; see also Flinn 1978: 139–64. 123 M. Williams 1990: 181. 124 Harris 1988: 25, 56. Flinn (1978: 145) also points out that without coal, charcoal shortages could have hobbled the growth of English iron production after 1750; his emphasis is on showing that the earlier rate of output was sustainable and that there was no worsening charcoal crisis that caused the development of coal-based iron-making. 125 Harris 1988: 26; Flinn 1958: 150. 126 Harris 1988: 26. 127 Hammersley 1973: 608–10 points out that high transport costs made wood prices vary enormously by locality, and often one seller or buyer dominated a particular market, making prices a poor guide to scarcity. Moreover, charcoal prices included a significant labor cost, and so were only loosely related to wood prices.
  • 33. E U RO P E B E F O R E A S I A? 61 becoming more widely available thanks to road- and canal-building; but as we shall see shortly, those gradual improvements were quite small compared to those made possible by steam (especially after 1750) and would soon have reached their limits. Moreover, real charcoal prices rose again after 1750, prob- ably due to increased iron output, even with more coal coming on line.128 Vastly more expensive fuel would certainly have put a crimp in the quantita- tive expansion of many industries, and it is not hard to see it limiting innova- tion as well. As we shall see, even the steam engine itself was at first suffi- ciently bulky, fuel-hungry, and dangerous that experimenting with it might not have seemed worth it if its fuel had cost much more and if the coal mines themselves had not been an ideal place to use it. We will have more to say about deforestation (and continental Europe) in chapter 5; for now it suffices to see how essential coal was to Britain’s breakthroughs, especially in iron, steel, steam, power, and transport. Moreover, though it would be too teleological to see in the early nineteenth- century coal boom all the ways in which cheap fossil fuels have eventually relaxed pressures from a finite land supply (even in farming itself, thanks to energy-intensive fertilizers), it was clearly a crucial step; water power, no mat- ter how much the wheels were improved, simply did not have the same poten- tial to provide energy inputs that would significantly outpace a rapidly grow- ing population for decades to come or to permit chemistry to substitute for land. Thus it seems sensible, after all, to look at the mining and uses of coal as the most likely European technological advantage that was purely home- grown, crucial to its nineteenth-century breakthrough, and (unlike textiles) not dependent for its full flowering on European access to overseas resources. Steam engines were crucial here, both as machines that used coal to power other processes and as the power source for more effective water pumps which permitted a huge expansion of coal-mining itself. M. W. Flinn has noted that despite the many ways in which wind, water, gravity, and horses were used to drain mines, none of these would have been much use at the depths where most of the country’s reserves were. Thus, without steam, “mining in Britain could scarcely have expanded [beyond 1700 levels of annual output] and must prob- ably have begun to show diminishing returns.”129 Instead, output grew by roughly 70 percent over the next 50 years and by almost 500 percent more between 1750 and 1830 (making the total increase roughly 900 percent), as steam engines for mining became both more numerous and more effective.130 Steam engines of a sort had been developed in various societies before the eighteenth century, though without ever becoming much more than a curi- osity.131 The Chinese had long understood the basic scientific principle 128 Flinn 1978: 143–45, 147–48; Hammersley 1973: 608–10. 129 Flinn 1984: 114. 130 Flinn 1984: 26, 121–28. 131 For China see, e.g., Needham 1965: 255.
  • 34. C H A P T E R O N E62 involved—the existence of atmospheric pressure—and had long since mas- tered (as part of their “box bellows”) a double-acting piston/cylinder system much like Watt’s, as well as a system for transforming rotary motion to linear motion that was as good as any known anywhere before the twentieth century. All that remained was to use the piston to turn the wheel rather than vice versa. (In a bellows, the jet of hot air moved by the piston was the goal, not a step toward powering the wheel.) A Jesuit missionary who showed off working miniature models of both a steam turbine-driven carriage and a steamboat at court in 1671 appears to have been working as much from Chinese as from Western models.132 In a strictly technological sense, then, this central technol- ogy of the Industrial Revolution could have been developed outside of Europe, too; thus we can never say definitively why it was in fact developed first in Europe. We can, however, identify some reasons why Europe—more spe- cifically Britain—was a particularly likely site for the series of linked develop- ments in coal and steam central to the Industrial Revolution. And when we compare England to the Yangzi Delta—where similar incentives existed to relieve pressure on the local wood supply, and where advanced technology and a highly commercialized economy were also present—Europe’s advantage rested as much on geographic accident as on overall levels of technical skill and much more than on any (probably nonexistent) advantage in the market efficiency of the economy as a whole. The relevant skills in which western Europe led the eighteenth-century world were ones in which Britain led. One of these was mining itself, but the others are not ones whose relevance is immediately obvious: clock-making, gunmaking, and navigational instruments. The story of Chinese mining in general, and coal-mining in particular, is somewhat puzzling. North and northwest China have huge coal deposits, and in the long era when the north included China’s political, economic, and dem- ographic center of gravity, China developed a huge coal and iron complex. Indeed, Hartwell estimates that Chinese iron production around the year 1080 probably exceeded that of non-Russian Europe in 1700. Moreover, this iron and coal complex was not merely large but sophisticated: Chinese ironmakers, for instance, seem to have known things about the creation and use of coke (purified coal) that would not be discovered elsewhere for centuries.133 But in the years from 1100 to 1400, North and Northwest China were hit by a stag- gering series of catastrophes: invasions and occupation (by the Mongols and others), civil wars, enormous floods (including a major shift in the Yellow River), and plague. The Jurchen invaders of the twelveth century often de- manded that some of the most skilled artisans in the capital region be turned over to them as a price for (temporarily) halting their siege; it is unclear 132 Needham 1965: 135–36, 225–26, 369–70, 387. 133 Hartwell 1967: 102–59.

Related Documents