John H. Coatsworth

John CoatsworthPresident of the Association, 1995

A paper read before the American Historical Association at its meeting in Atlanta, Georgia, January 5, 1996. Published in the American Historical Review 101, no. 1. (Feb. 1996), pp. 1–17.


The history of welfare, of change over time in human physical and spiritual well-being, encompasses most of the research and teaching that historians do. Whatever the explicit object of their research may be, historians can scarcely avoid finding evidence of the changes that have affected, for good or for ill, the health or happiness of their human subjects. Judgments about the welfare implications of historical research can be found embedded in the structure and language of virtually all historical narratives, even when their authors try, as many historians still do, to avoid making explicit the theoretical and ethical commitments that inform their work.

In contemporary political debate in the United States and several other developed countries, a curious linguistic inversion has loaded the term “welfare” with negative connotations. For most of this century, welfare has referred to the efforts of modern governments to improve the living standards of individuals or family groups whose incomes would otherwise fall below a level deemed minimal by policymakers and their constituents. In a somewhat broader definition, the term has also come to include governmental programs in education, health, housing, culture, old age and unemployment insurance, the environment, and the like. Too much welfare, we now hear, threatens our “civilization.”

Historians have much to contribute to the study of the relationship between welfare and civilization. I will address only one aspect of this immensely complex and fascinating field of research, that is, the relationship between improvements in human physical well-being (or welfare) and the advances in economic productivity and political organization that have made them possible. At the same time, I hope to suggest that research on the material conditions of life faced by human populations in the past can contribute to our understanding of many other historical questions.

In the past decade or so, historians have begun to make systematic use of various measures of physical well-being pioneered in the natural and human sciences to reexamine longstanding interpretations of historical change. Skeletal remains from prehistoric to more recent times are yielding masses of evidence on life expectancy, nutrition levels, the incidence of chronic disease, labor-related physical stress, and trauma injuries among diverse populations in many regions of the globe. Analysis of the refuse generated by human settlements has yielded additional information on nutrition and diet as well as the evolution of food products and food-producing technologies. For eras with written records, historians are now using data on height, body mass, and other physical characteristics along with demographic and epidemiological information to revise much of the received wisdom on a stunning array of historical issues.

This revisionism extends back in time to the dawn of what, for lack of a better term, we refer to as complex societies. The transition from the precarious nomadism of hunting and gathering to the more settled and secure life of sedentary agriculture and pastoralism marked a great advance for humankind. In the Old World, this transition first occurred in the millennium between 9000 and 8000 B.C., shortly after the last ice age, following the disappearance of the megafauna that had made hunting so productive. In the New World, the transition occurred 4,000 years later in Mesoamerica and the Andes, perhaps in part because the more favorable ratio of people to resources made hunting and gathering productive for a much longer time. With the diffusion of new productive techniques, there followed a long era in which the earth’s human population increased more rapidly than ever before (though still quite slowly by modern standards), human society became more complex, and the first great city-based territorial empires arose. Historians still refer to these societies as “civilizations” to denote their distinct achievements in art, religion, science, and law making.1

Not until recently, however, did we come to know much about the great price our common ancestors paid to achieve these advances in technology, social organization, and high culture. Bioarchaeologists have linked the agricultural transition to a significant decline in nutrition and to increases in disease, mortality, overwork, and violence in areas where skeletal remains make it possible to compare human welfare before and after the change. These results seem to hold whether the transition occurred gradually over a long time or was forced by conquest (as in parts of the New World). Civilization, we now know, stunted growth, spread disease, shortened life spans, and set people to killing and maiming each other on an unprecedented scale.2

Moreover, we now know that until relatively recently—until the twentieth century in most parts of the world—the cities where science and high culture achieved so much were such unhealthy places with such high mortality rates that none came close to reproducing, let alone expanding, its population by natural increase. The populations of cities as diverse as ancient Teotihuacan and industrial London grew in population only by luring or forcing people to move in from elsewhere. Advances in technology and organization associated with the rise of city-based empires increased the productivity of agriculture and the surpluses available for urban elites and their retainers, but the dense populations that settled in the wake of these advances sooner or later outstripped the growth of agricultural productivity and suffered from increased risk of disease due to overcrowding, poor sanitation, overwork, and malnutrition. These afflictions persisted nearly everywhere until the twentieth century.3

The Maya civilization in its “Classic” era from roughly 300 to 900 A.D. was once thought to have mysteriously escaped this ancient logic. The Classic Maya civilization of southern Mexico and northern Central America consisted of a series of splendid ceremonial centers ruled over by peace-loving priests who spent much of their time on astronomy, mathematics, and designing complex hieroglyphs to record their discoveries and confound archaeologists. The rest of the people earned their living by means of a relatively primitive slash-and-burn agriculture that yielded only enough to support themselves and their scientists. They came together for religious festivals and donated their time in the off-season to hauling rock for pyramids, but they built no cities because they could not have supported a large non-agricultural population. Nor did they engage in such other modern pastimes as trade, class struggle, or slaughtering their neighbors.4

This romantic vision of the Maya has crumbled over the past fifteen years. We now know that the Classic Mayan ceremonial centers were actually cities with populations as large as 80,000. To feed them, the Maya exploited the ecological resources of the regions they dominated with techniques more advanced and productive than previously thought. Ruled by warrior kings, they traded with places as far away as the Valley of Mexico and the Pacific coast of South America, and suffered from class divisions so sharp that the differences in nutritional levels between elite and commoner can be measured in the heights of the skeletons they left behind. Warfare was endemic. In short, the Maya were as civilized as any other civilization.5

The progress of our species out of these ancient cycles of rise and decline and into an era of sustained increase in levels of physical well-being is the unique achievement of the twentieth century. This conclusion has emerged in part from studies that use historical data on the height of adult populations to gauge levels of nutrition in human populations over the past three centuries. While the height attained by any given individual is affected by genetic and other idiosyncrasies, the average height of population groups is determined mainly by net nutrition in childhood, especially early childhood, and in adolescence.6 Chronic malnutrition of pre-modern populations kept adult heights well below modern levels through the nineteenth century in all countries and until quite recently in many less developed regions.7

Life expectancy provides another summary measure of the physical welfare of human populations. From prehistoric times until recently, low life expectancy was due mainly to the effects of chronic malnutrition, which made most people susceptible to the diseases that killed them at relatively early ages, especially in early childhood. Life expectancy in the cities of early modern Europe, such as Amsterdam, Geneva, and London in the sixteenth and seventeenth centuries, was probably not much higher than that of the ancient empires (though Rome was an especially deadly place).8 Lasting improvements in life expectancy did not even occur in the developed world until well into the twentieth century.9

During the industrial revolution of the nineteenth century, welfare stagnated or fell for decades. As in the ancient world, increases in productivity and the rise of cities did not lead to improvements in levels of physical well-being. In the United States, where scarce labor and abundant resources had produced an unusually well-fed population, the average height of the native-born U.S. male population stagnated from 1780 to 1830 and then fell by nearly five centimeters to a low point in the 1880s. The U.S. population did not recover the average stature it had achieved by the late eighteenth century until the 1920s. Life expectancy fell for a half-century after 1790 and then stagnated until near the end of the century.

One may wonder, given what we now know, why sensible people would choose to settle down in agricultural villages, move to cities, or pursue industrial development. The evidence suggests that choice had little to do with these processes. These great transitions seem to have occurred at times when older ways of life had become untenable or unattractive: the exhaustion of readily available wild sources of food, the promise of security or salvation in towns, the dearth of land or jobs in agriculture.10 Whether pushed or pulled, by lord or by market, most of those who joined these trends faced few alternatives. Even those whose individual role history has traditionally celebrated, from monarchs to moguls, pushed mainly at the margins of these trends. Their creativity rarely pushed beyond the context of their times.

The striking contrasts I have cited between technological and even cultural achievement, on the one hand, and changes in levels of physical welfare, on the other, contradict once-cherished assumptions. But productivity advance has never translated immediately or automatically into improvements in living standards any more than it has guaranteed spiritual comfort.

It is nonetheless true that significant advances in physical welfare cannot occur in economies that fail to grow. Without an increase in the output of goods and services per capita, that is, without economic growth, the ancient afflictions of our species cannot be wished away. Growth in productivity is not a sufficient condition for improvements in welfare, but it is a necessary one.

Scholarly efforts to quantify and measure the differences in productivity across time periods and between nations have a long history, but the modern origin of such efforts may be found in the work of Simon Kuznets and his collaborators in the 1950s.11 Since then, economic historians have devoted much of their time, energy, and skills at furious debate to producing ever more reasonable estimates of aggregate productivity in the ever more distant past. These efforts have concentrated on the economies of the North Atlantic region, where most of the world’s quantitative economic historians reside, but in the past two decades work on other regions has also advanced rapidly. Despite a daunting array of obstacles, including poor and missing data as well as serious conceptual and technical complexities, increasingly robust estimates of economic performance now exist for many countries as far back as the eighteenth century. A few brave souls have even produced calculations, full of plausible extrapolations and conjectures, for as far back as the early Roman Empire.12

Among the most interesting results of such efforts is the discovery that the gap in productivity between what are today the less developed regions and the more advanced economies is of relatively recent origin. The per capita output of the developed economies now averages about four times that of the less developed world.13 This gap probably did not exist before about 1700, by which time most of the West European economies had probably attained a level of per capita income not much above that of the Roman Empire (minus Egypt, which was poorer) in the first century A.D.14

The modern-day division of the globe between rich nations and poor thus originated in the eighteenth century, when a small number of North Atlantic economies began to grow slowly but irreversibly while most of the rest of the world did not. By the mid to late nineteenth century, the gap between “core” and “periphery”had already reached its modern dimensions in most areas. By that time, rising exports of minerals and agricultural products were producing sustained productivity advances in much of the less developed world. Catching up, however, would have required that the poorer countries attain higher growth rates than those achieved by their richer trading partners. Most failed to do so. Over the course of the twentieth century, economic growth rates in Asia, Eastern Europe, and Latin America have enabled these regions more or less to keep pace with the North Atlantic economies, growing faster in some periods and more slowly in others, but no region has managed to close the gap significantly. In the case of the African continent, growth rates in this century have consistently fallen below those of the rest of the world, so the African gap has nearly doubled in the past hundred years.15

These long-term regional trends coincided with growing disparities within most world regions.16 The per capita income spread between the world’s richest nation and its poorest could not have been much greater than 4 to 1 in 1820. By 1989, it was 39 to 1.17 A few of the nineteenth century’s poor nations have grown rapidly and gone far toward catching up, while others have fallen even further behind. In the past half-century, for example, Japan, Korea, and Taiwan have developed rapidly, while others, such as much of Africa and parts of Latin America and Asia, have fallen further behind.18 At the end of the twentieth century, the world’s production and productivity, and thus the essential building blocks for improvements in human welfare, are more unequally distributed across the globe than ever before.

Even in those economies where productivity has improved over the past century or two, the benefits of growth have spread unevenly across the populations that made them possible. Variations in the distribution of goods and services within societies explain how some productivity advances have translated into rapid improvements in physical welfare, while others made little difference or even accompanied notable declines in living standards.

Patterns of income distribution have usually followed (and sometimes exacerbated or otherwise transformed) existing social and cultural cleavages. For example, class divisions within human societies as diverse as ancient Mexico and nineteenth-century Britain were so sharp that they determined access to basic nutrients and thus physical stature as well as life expectancy. Among ancient Mesoamericans, ruling elites of nobles, priests, and warriors controlled access to food, particularly scarce sources of protein. Declining agricultural productivity in eras of urban overpopulation appear to have provoked elites to codify and enforce sumptuary laws that reserved the consumption of certain foodstuffs to themselves. Widening physical differences between commoners and elites coincided with the periods of upheaval, perhaps open class warfare, that seem to have preceded the destruction or abandonment of urban centers and the disintegration of states and empires.19

Similar differences between social strata were characteristic of most industrializing societies until relatively recently. In England in 1800, for example, the adult male members of the titled nobility stood a full five inches taller than the population as a whole.20 Income became even more concentrated during industrial revolutions in Britain, the United States, and elsewhere in part because the salaries of scarce, highly skilled or educated labor rose rapidly while the wages of the more abundant unskilled labor force stagnated until after the turn of the twentieth century.21 In the Third World, the shocks associated with the onset of export-led growth included waves of usurpation of peasant lands, often coinciding with the construction of railroads that raised land values and thus provided incentives for aristocratic land grabbing. Proletarianization of labor forces everywhere and mass immigration to some areas of the New World increased the supply of unskilled workers and depressed wages.22

Gender, ethnic, and age differences also mattered. In pre-modern times, during periods of prolonged pressure on food supplies, the fact that women often suffered more than men can be measured in the growing gap between the heights of the two sexes. In the modern era, the persistence of customs and institutions that inhibit the occupational mobility of major portions of the modern labor force, including female majorities nearly everywhere, has worked to preserve inherited inequalities unrelated to productivity in the distribution of income and wealth throughout the globe.

In many societies, ethnic minorities (or oppressed majorities) have poorer diets, suffer more from disease and overwork, and live shorter lives than favored groups. In the ancient world, the forced migrations of subject peoples to imperial capitals and densely populated work sites tended to increase mortality from disease compounded by chronic malnutrition. Indigenous peoples in the Americas still suffer from nutrition levels so low that in many countries they are visibly shorter than the “Europeans” for whom they work. In Cholula, Mexico, inhabitants whose lifestyle marks them as indigenous are no taller today than the inhabitants of the region in the pre-Hispanic era.23

Institutional mechanisms that have historically turned the output and labor of distinct ethnic or social groups into the uncompensated property rights of others through serfdom or slavery have generally tended to concentrate income in fewer hands. Periods of rapid growth in slave plantation agriculture in Latin America as well as the United States coincided with an intensification of labor, diminished net nutrition, and harsher slave codes, thus exacerbating existing inequalities.

The most vulnerable populations, historically, have been poor children. High infant and child mortality rates, more than any other factor, account for the failure of the world’s cities to maintain or expand their populations by natural increase until recently. In the southern United States, slave children, often fed separately, appear to have been deliberately undernourished by their owners.24

Just as economic growth has never guaranteed automatic and instantaneous improvements in the physical welfare of populations, effective productive effort has never ensured commensurate returns to those individuals who make them. The significance of the gap between private gain (or loss) and social benefit (or cost) from economic activity has propelled an entire subfield of economic history to explore the institutional determinants of productivity advance.25 Institutional changes that reduce this gap contributed, and continue to contribute, to stimulating economic growth. The modernization of legal, judicial, and regulatory systems, the abolition of caste distinctions and of slavery later on in the nineteenth century, the end of mercantilist trade monopolies and the development of freer international trade regimes, the growth of capital markets that facilitated vast flows of capital and technology across international boundaries, and a host of less dramatic changes have contributed enormously to productivity advance throughout the world. Most of the historical work on these modern achievements has focused appropriately on institutional changes that improved returns to entrepreneurs and innovators; it has produced major breakthroughs in our understanding of the institutional requisites for economic growth.

A similar approach to the history of distributional regimes would no doubt yield comparable insights into the ways institutional changes have affected welfare as well as productivity.26 Human societies distribute property rights and other entitlements in so many different ways and with such exquisite attention to detail that it is easy to lose sight of their long-term effects. Economic forces have played a major role in determining the distribution of wealth and income in most societies, even pre-modern ones, but they work within institutional constraints whose effects economic historians have not always studied systematically.

Simon Kuznets noticed four decades ago that income distribution in the developed countries had become less equal in the early stages of economic growth but more equal thereafter. He hypothesized a similar trajectory for the less developed regions. The evidence of the past half-century confirms his hypothesis for some countries but not for others. The trend toward greater equality in Western Europe has slowed in the past decade or so. In the United States, it has been reversed. A few of the fast-growing East Asian countries have become more equal; most of Latin America has not. And in most of Eastern Europe, relatively egalitarian systems have been replaced by regimes that recall Karl Marx’s concept of “primitive accumulation.” These trends have implications for the welfare and productivity of vast populations.

The twentieth century has witnessed unprecedented improvements in the welfare of human populations, finally translating the increased productivity of the industrial age into measurable achievements in nutrition, health, and thus life expectancy. Without greater productivity, little could have been accomplished to improve living standards. At the same time, as the examples I have cited amply demonstrate, human societies have frequently found it difficult or even impossible to convert productivity advance into improvements in physical well-being. Resolving this dilemma, economic historians are now discovering, awaited the discovery of an appropriate and effective mechanism for addressing it. That mechanism, deployed with increasing impact over the past century, is the “welfare state.”

Precedents for effective government intervention may be found in earlier times, of course. The history of famine relief is a good example. Widespread hunger due to weather-related harvest failures has killed countless millions from prehistoric times until today. Until recently, famines were treated by historians as evidence of humanity’s victimization by a recurrently irritable Mother Nature. We now know that most historic famines were man-made, produced not by Mother Nature but by earthly officials of the other sex who simply failed to move available food supplies to stricken areas. Famines ended in England only after food riots in the late 1750s pushed the government back toward Tudor-Stuart “paternalism” by intervening more vigorously in the grain market during periods of scarcity.27 In India, as Amaryta Sen showed in a classic study, British colonial authorities could have saved millions during the 1943 Bengal famine with relatively little effort. It was not the Green Revolution that put an end to famine in India, it was better understanding of how famines work and an independent government that, gave higher priority to preventing or stopping them.28 Contemporary famines occur mainly in countries without effective governments.

Poor relief also originated long before the twentieth century, as did free primary and even secondary schooling in some countries, efforts to make urban places more habitable, early pension schemes mainly for war veterans and their families throughout the Western Hemisphere and northwestern Europe, the first steps toward developing modern research centers in science, medicine, and technology in the industrializing countries, and so on. Nearly all of these undertakings helped to improve living standards, but even taken together their effects were limited by the relatively small scale of the resources assigned to them and by constraints on the effective regulatory activity of governments.

The remarkable improvements in physical well-being we have experienced over the course of this century were achieved mainly as a result of massive growth in government spending and regulation. Late in the nineteenth century, governments in Europe and North America began to translate medical and technical knowledge into efforts that made cities less deadly places to live. Aggressive regulation of housing conditions and public health programs ranging from quarantines to immunizations reduced disease. Subsidized housing programs for workers and the poor, vigorous efforts to clean up air, water, and food supplies, and investments in other urban amenities became commonplace.

In the post. World War II era, if not before, income-maintenance schemes, including direct subsidies or transfer payments to the unemployed, the indigent, veterans, and the elderly, expanded enormously throughout the North Atlantic region. In the developed countries, national health systems providing “free”medical care for all citizens became commonplace, except in the United States. Mass educational systems, pioneered in the United States, had already proved effective in virtually eliminating illiteracy and extending primary, and in some countries secondary, schooling to most citizens of the developed regions. In the 1950s and 1960s, most developed countries extended these systems to accommodate growing numbers of post-secondary students.

Much of the less developed world moved more slowly. Most of the Latin American countries resembled Mexico, a country that did not reach the 1800 U.S. literacy level of 80 percent until the 1980s. Improvements in urban living conditions, including better sanitation and the development of public health efforts, especially after World War II, lowered death rates sufficiently to push population growth up along with stature and life expectancy. Similar achievements were recorded in most of Asia and Eastern Europe, though much of sub-Saharan Africa again lagged behind.

In assessing these trends in human welfare, economic historians have made many important contributions to our understanding of the relationship between productivity and physical well-being. Three are of particular significance. First, the recent work has pointed to the impact of greater equality on income distribution to improvements in physical welfare. Second, improvements in the physical welfare of populations have been shown to raise the productivity of entire economies. Finally, much work over the past two decades has also demonstrated that education, and investment in “human capital” more generally, raises both productivity and living standards.

Studies of the relationship between stature and a measure of the degree of equality or inequality in the distribution of income (the Gini index) tend to the conclusion that greater equality has historically yielded higher average nutrition levels.The population of colonial British North America, for example, reached an average stature well above the level that the colonies. per capita income would have permitted in a society where income was more concentrated. Growing inequality during the nineteenth century probably contributed, along with urbanization, to making Americans shorter.29 In the twentieth century, increases in equality have helped to propel rising standards of living, particularly when the beneficiaries have included the lowest income earners.

The contribution of better health and nutrition to increasing productivity has only recently been recognized. In Great Britain, as late as 1800, with average life expectancy at birth hovering at about thirty-five years, as much as 20 percent of the British adult population was so undernourished as to be physically unable to work or indeed to do anything more strenuous that a few hours of easy strolling each day. As health improved, a larger portion of the British population acquired the stamina to work regularly with increasing intensity and to do so over longer productive lives. In this century, higher wages made for better diets, while the British government programs that made city living more healthy also lowered the nutritional needs of urban populations because they reduced the energy that had to be expended warding off disease. In his Nobel Prize lecture, Robert Fogel estimated that 30 percent of the entire increase in the productivity of the British economy between 1790 and 1980 was due to improvements in gross nutrition.30

Finally, rising investments in human resources around the world have contributed both to improved welfare and to rising productivity. Since education induces greater mobility, both occupational and geographic, and at higher levels also provides training in specific skills, investment in education has historically raised incomes by increasing the productivity of those who get it. Cost-benefit analyses by governments and international agencies, as well as a growing number of historical studies, have shown consistently high welfare and productivity effects associated with investing in human resources.31

Of course, a major portion of the resources deployed by twentieth-century governments has aimed at destroying rather than nourishing or otherwise protecting human life. Some efforts to reduce inequality and improve well-being have not proved effective in achieving their objectives, while others have pursued laudable goals without due regard for efficiency or possible macroeconomic imbalances. Waste, corruption, and profligacy have appeared in nearly every epoch of human history (though inflation is a largely twentieth-century scourge). In short, some portion of what modern governments do has actually undermined the economic growth on which their accomplishments critically depend. Despite such problems, we now know that what distinguishes modern civilization from earlier epochs is a result in great part of the effectiveness of public programs (throughout the developed world and a growing portion of the less developed regions of the globe) in improving living standards and prolonging life itself.

The progress our age has witnessed in raising standards of physical well-being stands in sharp contrast to the twentieth century’s record of unprecedented civil strife and international violence. More people have lost their lives in episodes of collective violence in this century than in any other epoch in human history. It is tempting to treat war and welfare as distinct and unrelated phenomena, to condemn one and praise the other as though they pertained to different planets. For purposes of analytical focus or coherent narrative, such an approach might well make sense. For understanding the history of human welfare, it does not.

History has favored a tiny minority of the earth’s people with institutions capable of accommodating the aspirations of our species to a better life, to children that survive, grow taller, and live longer. Culture and tradition move us to hope that democratic institutions, in addition to economic growth, will prove equally effective elsewhere in the globe, but the crucial tests of this hypothesis await the next century in most countries. Indeed, as recent trends suggest, we in the developed world may be facing tests of our own.

The history of welfare can, I think, prove helpful in assessing the social, political, and cultural alternatives facing diverse societies, particularly in periods of change and transition. Knowledge about the economic past can be useful for understanding the changing contexts, and thus the meaning and significance, of human thought and expression in history. Nothing in the historical record that I know tells us unequivocally that people who live longer, eat better, get sick less often, and spend more years in school experience greater happiness in life than their short-lived, chronically undernourished, disease-ridden, and generally illiterate ancestors.32 On the other hand, since the material conditions of life do seem to exert such a measurably powerful influence on human behavior, it would be foolhardy to ignore them.

John H. Coatsworth is Monroe Gutman Professor of Latin American Affairs in the Department of History at Harvard University. Coatsworth received his BA degree in history from Wesleyan University in 1963 and his MA (1967) and PhD (1972) in economic history at the University of Wisconsin, Madison. At Harvard, he also serves as director of the recently founded David Rockefeller Center for Latin American Studies. Coatsworth’s research has focused primarily on the social and economic history of Mexico and the international history of Mexico, Central America, and the Caribbean. His most recent book, The United States and Central America: The Clients and the Colossus (1994), surveys U.S.-Central American relations, focusing on the era since World War II.


1. The classic work on the Neolithic Revolution in this tradition was by V. Gordon Child, Man Makes Himself (London, 1941).

2. See, for example, Mark N. Cohen, Health and the Rise of Civilization (New Haven, Conn., 1989); for a New World example, see Clark S. Larsen, ed., “The Archaeology of Mission Santa Catalina de Guale: Biocultural Interpretations of a Population in Transition,” Anthropological Papers, 68 (1990).

3. For an excellent summary of the literature, see Rebecca Storey, Life and Death in the Ancient City of Teotihuacan: A Modern Paleodemographic Synthesis (Tuscaloosa, Ala., 1992), 35–42.

4. This view of Maya civilization was most forcefully asserted and long defended by J. E. S. Thompson, who “[u]ntil his death in 1975 ... dominated modern Maya studies by sheer force of intellect and personality,” as Michael D. Coe put it in Breaking the Maya Code (London, 1992), 123.

5. For a highly readable account of a major site, see William L. Fash, Scribes, Warriors, and Kings: The City of Copdn and the Ancient Maya (London, 1991).

6. Net nutrition refers to the balance between nutrition “inputs” and the physical demands of work, illness, and the like. Higher levels of nutrition are required when work is more arduous and sickness strikes.

7. See the Nobel Prize lecture of Robert William Fogel, “Economic Growth, Population Theory, and Physiology: The Bearing of Long-Term Processes on the Making of Economic Policy,” American Economic Review, 84 (1994): 372. For an excellent survey of this field, see Richard Steckel, “Stature and the Standard of Living,” Journal of Economic Literature (forthcoming).

8. Storey, Life and Death, chap. 1; see also Jan de Vries, European Industrialization, 1500–1800 (London, 1984).

9. Robert William Fogel, “Egalitarianism: The Economic Revolution of the Twentieth Century” (Simon Kuznets Memorial Lectures, Yale University, unpublished, 1992), 25.

10. For the Neolithic Revolution, this view is associated with Ester Boserup, The Conditions of Agricultural Growth (Chicago, 1965).

11. Simon Kuznets, “Economic Growth and Income Inequality,” American Economic Review, 45, no. 1 (1955):1–28.

12. Raymond Goldsmith, “An Estimate of the Size and Structure of the National Product of the Early Roman Empire,” Review of Income and Wealth, 30 (1984): 263–88.

13. This ratio is based on Angus Maddison’s sample of forty-three countries in “Explaining the Economic Performance of Nations, 1820–1989,” in Convergence of Productivity: Cross-National Studies of Historical Evidence, William J. Baumol, Richard R. Nelson, and Edward N. Wolff, eds. (New York, 1994), 20–61. Maddison’s estimates have the advantage that they are adjusted to purchasing power parity; unadjusted estimates that tend to understate the income of less developed countries, typically by using exchange rates to convert local-currency estimates into dollars, give ratios of 1 : 8 or more.

14. For an interesting survey of the literature and some rough guesses, see Paul Bairoch, Economics and World History (New York, 1993), chap. 9. For the comparison with Rome, see Goldsmith, “Estimate of the Size and Structure”; life expectancy in Rome was lower than in early modern Europe, but Goldsmith’s other comparisons suggest rough comparability.

15. This and the following paragraph is largely based on Maddison, “Explaining the Economic Performance of Nations,” 22–27.

16. Gabriel Tortella, however, shows convergence in twentieth-century Europe between southern Europe (Italy, Portugal, and Spain) and the north (England and France), in “Patterns of Economic Retardation and Recovery in South-Western Europe in the Nineteenth, and Twentieth Centuries,” Economic History Review, 47, no. 1 (1994): 1–21.

17. Maddison, “Explaining the Economic Performance of Nations,” 23–30.

18. Smaller, though still significant, disparities in economic performance occurred among the developed economies, some of which, such as Germany or Italy, have grown faster than others, such as Great Britain or the United States. See Maddison, “Explaining the Economic Performance of Nations,” 22–23.

19. See the essays in The Collapse of Ancient States and Civilizations, Norman Yoffee and George L. Cowgill, eds. (Tucson, Ariz., 1988).

20. This difference declined, mainly in the twentieth century, to barely one inch today. See Fogel, “Egalitarianism,” 29.

21. See Jeffrey G. Williamson, “Globalization, Convergence, and History” (National Bureau of Economic Research, Working Paper 5259, 1995), 24–26.

22. As Williamson put it, “In the New World, unskilled labor lost and landowners gained from [the forces that raised growth rates relative to the developed world],”“Globalization, Convergence, and History,” 24.

23. Carlos Arturo Giordano Sanchez Verin, “La alimentación como reflejo del desarrollo físico en dos comunidades rurales de México: Cholula e Ixtenco. (unpublished paper, 1994).

24. Richard Steckel, “A Dreadful Childhood: The Excess Mortality of American Slaves,” Social Science History, 46, no. 3 (1986): 427–65.

25. Among the pioneers of this approach, Douglas North stands out. See, for example, “Institutions,” Journal of Economic Perspectives, 5, no. 1 (1991): 97. 112.

26. This point is made cogently by Alexander James Field, “Do Legal Systems Matter?” Explorations in Economic History, 28 (1991): 1–35.

27. See Fogel, “Egalitarianism,” 9.

28. Amaryta Sen, “Starvation and Exchange Entitlements: A General Approach and Its Application to the Great Bengal Famine,” Cambridge Journal of Economics, 1 (1977): 33–39.

29. For a summary of this discussion, see Steckel, “Stature and the Standard of Living. (forthcoming)”

30. Fogel, “Economic Growth,” 383; see also Steckel, “Stature and the Standard of Living,”for a discussion of the U.S. case.

31. For a recent historical study, noteworthy for its sophistication and balance, see Clara Eugenia Nuñez, La fuente de la riqueza: Educación y desarrollo económico en la España contemporánea (Madrid, 1992).

32. The history of migratory flows across international boundaries does suggest, nonetheless, a widespread preference for higher living standards.