Technology innovation in Chinese agriculture

It is a commonplace in world history to observe that China had achieved a high level of sophistication in science, medicine, and astronomy by the Middle Ages, but that some unknown feature of social organization or culture blocked the further development of this science into the expansion of technology in the early modern period. Chinese culture was "blocked" from making significant technological advances during the late Ming and early Qing periods -- in spite of its scientific advantage over the West in medieval times; or so it is believed in a standard version of Chinese economic history.

A variety of hypotheses have been offered to account for this supposed fact. For example, Mark Elvin argues that China's social and demographic system created conditions for a "high-level equilibrium trap" in the early modern period in The Pattern of the Chinese Past. According to Elvin, Chinese social arrangements favored population growth; innovative and resourceful farmers discovered all feasible refinements of traditional agricultural techniques to refine a highly labor-intensive system of agriculture; and population expanded to the point where the whole population was at roughly the subsistence level while consuming virtually the whole of the agricultural product. There was consequently no social surplus that might have been used to invest in discovery of major innovations in agricultural technology; so the civilization was trapped. (Here is a more developed discussion of Elvin's argument.)

Other historians have speculated about potential features of Confucian culture that might have blocked the transition from scientific knowledge to technology applications. The leading Western expert on Chinese science is Joseph Needham (1900-1995), whose multi-volume studies on Chinese science set the standard in this area (Science and Civilisation in China. Volume 1: Introductory Orientations; Clerks and Craftsmen in China and the West). And Needham attributes China's failure to continue to make scientific progress to features of its traditional culture.

But here is a more fundamental question: is the received wisdom in fact true? Was Chinese technology unusually stagnant during the early-modern period (late Ming, early Qing)? Agriculture is a particularly important aspect of traditional economic life; so we might reformulate our question a bit more specifically: what was the status of agricultural technology in the seventeenth and eighteenth centuries (late Ming, early Qing)? (See an earlier posting on Chinese agricultural history for more on this subject.)


Economic historian Bozhong Li considers this question with respect to the agriculture of the lower Yangzi Delta in Agricultural Development in Jiangnan, 1620-1850. And since this was the most important agricultural region in China for centuries, his findings are important. (It was also the major cultural center of China; see the concentration of literati in the map above.) Li makes an important point about technological innovation by distinguishing between invention and dissemination. An important innovation may be discovered in one time period but only adopted and disseminated over a wide territory much later. And the economic effects of the innovation only take hold when there is broad dissemination. This was true for Chinese agriculture during the Ming period, according to Li:

The revolutionary advance in Jiangnan rice agriculture technology appeared in the late Tang and led to the emergence and development of intensive agriculture composed of double-cropping rice and wheat. But this kind of intensive agriculture in pre-Ming times was largely limited to the high-fields of western Jiangnan. In the Ming this pattern developed into what Kitada has called the 'new double-cropping system' and spread throughout Jiangnan, but only in the late Ming did it become a leading crop regime. Similar were the development and spread of mulberry and cotton farming technologies, though they were limited to particular areas and cotton technology's advances came later because cotton was introduced later. Each had its major advances in the Ming. Therefore, technology advances in Ming Jiangnan agriculture were certainly not inferior to those of Song times which are looked at as a period of 'farming revolution'. (40)

Li also finds that there was a significant increase in the number of crop varieties in the early Qing -- another indication of technological development. He observes, "The later the date, the greater the number of varieties. For example, in the two prefectures of Suzhou and Changzhou, 46 varieties were found in the Song, but the number rose to 118 in the Ming and 259 in the Qing" (40). And this proliferation of varieties permitted farmers to adjust their crop to local soil, water, and climate conditions -- thus increasing the output of the crop per unit of land. Moreover, formal knowledge of the properties of the main varieties increased from Ming to Qing periods; "By the mid-Qing, the concept of 'early' rice had become clear and exact, and knowledge of 'intermediate' and 'late' strains had also deepened" (42). This knowledge is important, because it indicates an ability to codify the match between the variety to the local farming environment.

Another important process of technology change in agriculture had to do with fertilizer use. Here again Li finds that there was significant enhancement, discovery, and dissemination of new uses of fertilizer in the Ming-Qing period.

A great advance in fertilizer use took place in Jiangnan during the early and mid-Qing, an advance so significant that it can be called a 'fertilizer revolution'. The advance included three aspects: (a) an improvement in fertilizer application techniques, centring on the use of top dressing; (b) progress in the processing of traditional fertilizer; and (c) an introduction of a new kind of fertilizer, oilcake. Although all three advances began to appear in the Ming, they were not widespread until the Qing. (46)

And the discovery of oilcake was very important to the increases in land productivity that Qing agriculture witnessed -- thus permitting a constant or slightly rising standard of living during a period of some population increase.

There were also advances in the use of water resources. Raising fish in ponds, for example, became an important farming activity in the late Ming period, and pond fish became a widely commercialized product in the Qing. Li describes large-scale fishing operations in Lake Tai in Jiangnan using large fishing boats with six masts to catch and transport the fish (62).

So Li's estimate of agricultural technology during the Ming period is that it was not stagnant; rather, there was significant diffusion of new crops, rotation systems, and fertilizers that led to significant increases in agricultural product during the period. "In sum, in the Jiangnan plain, land and water resources were used more rationally and fully in the early and mid-Qing than they had been in the late Ming" (64).

Two points emerge from this discussion. First, Li's account does in fact succeed in documenting a variety of knowledge-based changes in agricultural practices and techniques that led to rising productivity during the Ming-Qing period in Jiangnan. So the stereotype of "stagnant Chinese technology" does not serve us well. Second, though, what Li does not find is what we might call "science-based" technology change: for example, the discovery of chemical fertilizer, controlled experiments in rice breeding, or the use of machinery in irrigation. The innovations that he describes appear to be a combination of local adaptation and diffusion of discoveries across a broad territory.

So perhaps the question posed at the start still remains: what stood in the way of development of empirical sciences like chemistry or mechanics that would have supported science-based technological innovations in the early modern period in China?

Agrarian history -- the Weber edition

One of Max Weber's early areas of research was what might be called "macro agrarian history". This was a field of research that Weber himself largely invented. He undertook to document and explain the large patterns of economic development in the ancient world, including especially the social systems surrounding farming and animal husbandry. Weber's cases include Mesopotamia, Egypt, Israel, Greece, and Rome. How did the fundamental material activities of farming, trading, and consuming contribute to the development of major civilizations? This research culminated in several important manuscripts, including in particular Agrarian Sociology of Ancient Civilizations (1909; translated and edited by R. I. Frank in 1976). Weber was particularly interested in the causal and structural features of the specific forms of property relations, labor, trade, taxation, and consumption that characterized the social economies of the ancient world. And he tries to show how certain features of the agrarian system influenced the emergence of certain political and legal forms.

The title "Agrarverhältnisse in Altertum" is translated as "Agrarian Sociology of Ancient Civilizations"; but this is a little misleading. It's not so much a work of sociology as it is a work of economic and political history at a civilizational level. The essay is offered as an account of the history and particularity of the rural economies of the ancient world -- closer to economic history than sociology.

Weber's treatment of ancient Greece is particularly detailed. So let's inventory some of the main components of his account to get a flavor of his approach. This will give us a better idea of how he set about trying to understand and explain the social world that is presented through ancient Greek literature. Here are the first few paragraphs of his description of the agrarian society of "historic Greece" -- the contemporary society of Homer, or roughly 1000 BCE:

The Greeks raised spelt, barley, and wheat, alternating this in each field with grass (hence leases were for an even number of years). This system continued to be followed except in those areas specializing in a particular crop. Occasionally, it seems, the three-field system was used, but there was no change of crops, except that legumes were grown on fallow land. The use of manure is mentioned in Homer (but green manure was not used until later times); in general, however, agricultural techniques were stabilized at a rather primitive level and thereafter did not develop.

Thus the main features of Greek agriculture continued to be ploughing with a hooked plough (for long entirely of wood) drawn by oxen, sowing seed in furrows, hacking and weeding the grain fields, and harvesting with sickle and threshing board. Hence labour intensity was considerable, and since virgin land was no longer available, it was difficult to shift from subsistence agriculture to market production, even though grain prices were high in later times.

Cattle raising does not seem to have been much reduced in extent by farming until the age of the tyrants, who favoured the peasantry. The Homeric epics indicated a diet based mainly on cheese, milk, and -- among aristocrats -- meat. Clothing was made of wool and skins. (147)

So -- crops, technology and practice, labor productivity, and consumption patterns. And what about the social relationships within which these farm activities took place?

In historical times the prevalent form of living unit was the patriarchal nuclear family. Women and children were in much the same position as among Semitic peoples: women could be bought or else married with dowry; husbands had the right to send away their wives, and could sell, rent, expel, or kill their children. Later the laws governing legitimacy, and the feeling for kin ties cultivated among the great families combined to reduce these powers of the father over his children; by the time when the Gortyn Code was first framed these changes had all taken place. (148)

But whereas the masses lived in nuclear families, nobles and kings -- at first the same -- lived as elsewhere in large households including agnates of a clan (genos). The purpose of this was to preserve the unity of inheritable landed estates. Thus both separate and group inheritance are mentioned in the Homeric epics; see the homosipuoi of Charondas, equivalent to the homogalaktes of Attic law. (148)

Here we have some social detail: the family, property in land, inheritance, and a "class" distinction between the farmer class and the elite families. And we have regular reminders of the primary historical sources underlying these descriptions: the Homeric epics and other surviving texts from Greek literature and history.

Weber quickly identifies social inequality in ancient Greek society and points to a system of wealth extraction from the masses to the elite families:

Those people who did not belong to a numerous and economically established clan, who were in short without land, found themselves forced to enter the clientele of one or another aristocrat. This was a later development, as the supply of new land declined and differences in wealth developed; originally membership of the community and ownership of land each presupposed the other. (149)

What about politics -- the use of military force to extract resources and compel population behavior?

The Dorian cities were fundamentally military states, and so everywhere they maintained the same three tribes. Elsewhere there was great variety, but everywhere and always the formal division of a community into tribes signified one thing: that a people had constituted itself as a polis ready for war at any time. It should be noted here that the proper word for a tribe in a non-urbanized community was ethnos, as is shown by the documents of the Delphic Amphictyone. (151)

More precise information is not available for the political and social structures of autonomous communities in early times. If, however, we rely on analogies with other peoples, then we can assume that in each community the position of ruler (anax) came to be hereditary in a family made prominent by wealth in cattle and marked out as favoured by the gods by success in war and equity in judgment. (151)

When he moves on to the later period in ancient Greek history he extends the list of concepts used to characterize society; he describes the development of cities, the extension of long-distance trade, the social and cultural influence of Asia Minor, and the growth of political power wielded by aristocratic kings.

This is just a small snapshot of Weber's historical reasoning in Agrarian Sociology of Ancient Civilizations. The book is methodical, and it moves through an orderly series of objective concepts of social organization in order to provide a description of the historically given society.

So the text is full of interesting claims and details about the agrarian regimes of human society 3,000 years ago. But here is the question that needs posing in the context of UnderstandingSociety: What is Weber trying to accomplish in this piece of writing? And how does his analysis shed light on his intuitive views of social and historical knowledge? What can we discover about Weber's sociological imagination from this book? Several summary comments seem fairly clear.

First, much of the intellectual work here is devoted to de-coding the social and economic implications of surviving Greek literature, including particularly the Homeric poems. Weber is trying to piece together a coherent story of the modes of farming, technology, property, family, kinship, and kingship that are implied in the many small clues about social life contained in the Homeric corpus. So there is a large part of the work that is intended to be conceptual and descriptive.

Second, Weber lays bare one of his own interests in his construction of the text: a concern for viewing large social formations -- civilizations or social-economic regimes -- and discovering some of the material and situational factors that influence their development. This introduces an explanatory ambition to the work; Weber wants to be able to explain how some developments are the causal result of other developments.

Third, there is an important element of comparison involved in the book. Weber is plainly interested in noticing the differences as well as the similarities in the agrarian regimes he describes in Egypt, Greece, or Rome.

Fourth, there is not much attention given to the role of ideas or ideologies in this book -- in contrast to the central role that ideas play in his comparative religion research some years later. And there is no trace of "interpretive" inquiry in this book either. Weber is trying to discover an objective vocabulary in terms of which to describe the material and social arrangements that are implied by the Homeric corpus, without attempting to say how the agents experienced or represented these social relations. This is not a "hermeneutic" book; if anything, it more resembles an ethnographically sensitive materialism. We might even say that the book is Weber's, but it is something other than Weberian.

Finally, the topic of power returns repeatedly throughout the book. Weber demonstrates his recurring interest in the role that coercion and military force play in the organization of society and the concentration of wealth.

Several generations later, the great classical historian M. I. Finley undertook a parallel investigation of the agrarian economies of Greece and Rome in The Ancient Economy (1973). Finley's work is more clearly organized by the goal of showing how the material, technical, and social relations he describes constitute an economic system -- a system of production and reproduction. And it reflects a modest kind of Marxism in its effort to describe the forces and relations of production in the ancient world. Certainly the body of historical data that was available to Finley about the social arrangements of the ancient world was much greater than what was available to Weber; in this sense we are likely to judge that Finley's account is more likely to be accurate in detail. But it is very interesting to read the two books side-by-side; they have a lot in common. And each presents an ambitious view of what is involved in knowing how ancient societies worked.

(It is interesting to discover that M. I. Finley and R. I. Frank, the translator and editor of Agrarian Sociology of Ancient Civilizations, had a bit of a dust-up in the New York Review of Books in 1970 over a review that Finley wrote of several books on the Roman Empire.)

Malthus blogging on the Corn Laws

I think that Thomas Malthus would have been very much at home in the blogosphere. He weighed in on the issues of the day, bringing careful logical analysis of economic theory to bear on the policy issues that were up for debate. And he was very interested in making the connection between economic principles and real empirical evidence. This is particularly true in his contributions to the debate on the Corn Laws in 1814 and 1815. Malthus authored pamphlets on these issues in 1814 ("Observations on the effects of the corn laws"; link) and 1815 ("Grounds of an opinion on the policy of restricting the importation of foreign corn"; link), and they repay scrutiny today; they are powerful instances of a very smart economist probing the theory and the facts surrounding a complex policy issue. (Here is a nice survey of Malthus's theories; link.)

The Corn Laws might be thought of as a form of "stimulus package" for the British economy in the early nineteenth century. By setting a high tariff on the import of wheat and other grains, Parliament aimed to protect the agricultural sector and to encourage the expansion of grain production to make Britain more independent from external grain providers. One might also compare the debate to the NAFTA debate or to policy deliberations in the 1960s concerning "import substitution" strategies. Opponents argued that removal of the tariffs would bring down the price of grain, a central component of the wage basket; this would help the poor and would also permit a significant reduction of the wage as well. So the issue divides the interests of land owners, industrialists, and the poor.

Malthus's position in the two essays is somewhat different. In the first article he promises to lay out the issue dispassionately, dispelling false opinions about what the effects of the proposed policy might be and diving into the advantages and disadvantages of the policy. He writes that "some important considerations have been neglected on both sides of the question, and the effects of the corn laws, and of a rise or fall in the price of corn, on the agriculture and general wealth of the state, have not yet been fully laid before the public." A bit further on, he writes:

My main object is to assist in affording the materials for a just and enlightened decision; and whatever that decision may be, to prevent disappointment, in the event of the effects of the measure not being such as were previously contemplated. Nothing would tend so powerfully to bring the general principles of political economy into disrepute, and to prevent their spreading, as their being supported upon any occasion by reasoning, which constant and unequivocal experience should afterwards prove to be fallacious.

So--"let's do rigorous and systematic analysis based on the principles of political economy and our best understanding of the facts." Good advice for a policy debate.

Quite a bit of the analysis is devoted to refuting an idea that Malthus attributes to Adam Smith -- that "corn" is a unique commodity because increases and decreases in its price have no effect on agricultural production. The idea seems to be that the price of corn determines the wage and thereby determines all other prices in the economy; a small increase in the price of corn immediately causes an equal increase in the price of labor and all other prices. So corn is a "numéraire" to the whole economy; therefore the economy cannot be influenced by tariffs that alter the money price of corn. This is a notion that Malthus quickly and efficiently dispatches. Referring to the poverty studies of Frederick Morton Eden (1797) (whom Marx also relies on in the Grundrisse), Malthus examines the budget of the laboring poor and finds that only 40% of this budget is directly influenced by the price of grain. So the effect of the price of grain on the wage is only partial. (If we assume that the wage is at equilibrium when it equals the cost of the minimum wage basket, then an increase of 10% in the cost of the bread in the basket (40%) will have only 4% effect on the total cost of the wage basket.) Moreover, he points out that labor markets are different from many other commodity markets in the sense that it is not possible to quickly decrease the supply of labor; so the effect of changes in the price of corn will work only slowly into changes in the price of labor. "It is manifest therefore that the whole of the wages of labour can never rise and fall in proportion to the variations in the price of grain."

Second, on the production side, Malthus goes into quite a bit of detail in the micro-economics of farming: land quality, choice of crops, improvements of land, amount of hired labor and draft animals. And he demonstrates that an increase in the price of the crop will give a clear economic incentive to investors to expand and intensify agricultural production. so quantity responds to rising prices. Higher prices => higher profits => rational incentive to invest more capital in expanded production. In short, the Smithian idea that "corn is unique" is untenable.

The central "general principles" of political economy that Malthus attributes to Smith -- and which he endorses himself -- are essentially these: that the price of a good varies over time according to fluctuations in supply and demand; that the quantity of a good increases when producers have a price-based incentive to invest more capital in its production (and vice versa); and that capital will flow across productive uses in such a way as to bring about an equal rate of profit across sectors and industries. Malthus argues that the idea that corn is a unique commodity directly contradicts these principles. "Corn is subjected to the same laws as other commodities, and the difference between them is by no means so great as stated by Dr Smith."

This conclusion has a direct relevance to the topic of the corn laws. Malthus hereby concludes that a tariff or other restriction on imported grain will in fact have the effect of stimulating additional domestic production. So Britain's grain production would increase under this set of laws. However, here Malthus makes a point about Britain's agricultural potential and the density of its population:

On the whole then considering the present accumulation of manufacturing population in this country, compared with any other in Europe, the expenses attending enclosures, the price of labour and the weight of taxes, few things seem less probable, than that Great Britain should naturally grow an independent supply of corn.

So food self-sufficiency is impossible for Britain (in 1814!). Moreover, there is a Malthusian demographic consequence of free trade in grain that Malthus directly recognizes:

As one of the evils therefore attending the throwing open our ports, it may be stated, that if the stimulus to population, from the cheapness of grain, should in the course of twenty or twenty five years reduce the earnings of the labourer to the same quantity of corn as at present, at the same price as in the rest of Europe, the condition of the lower classes of people in this country would be deteriorated.

Malthus explicitly avoids coming to a conclusion of the overall advisability of the corn laws in this essay; but it is hard not to feel that the balance of arguments here appear to favor "open ports" -- no tariffs or restrictions on the import of grain.

So it is surprising to turn to the second essay, less than a year later ("Grounds of an opinion"), because here he lays out a specific argument in favor of the tariffs. He offers as a central reason for this conclusion the fact that protection will stimulate more agricultural production which will make Britain more nearly grain-sufficient. The argument turns less on economic principles and more on the predicted behavior of grain-producing nations such as France in times of dearth. Recent historical experience demonstrated to Malthus that countries will limit their exports of grain at times when the supply is short and prices are high; but this is precisely when Britain would need to continue to have unfettered access to foreign grain markets. In effect, we might read the first essay as creating the argument in principle for free trade in grain, and the second essay as an argument based on the specific historical facts of international trade that make the free trade policy unwise. Here is how Malthus tries to reconcile theory and empirical experience:

I am very far indeed from meaning to insinuate, that if we cannot have the most perfect freedom of trade, we should have none; or that a great nation must immediately alter its commercial policy, whenever any of the countries with which it deals passes laws inconsistent with the principles of freedom. But I protest most entirely against the doctrine, that we are to pursue our general principles without ever looking to see if they are applicable to the case before us; and that in politics and political economy, we are to go straight forward, as we certainly ought to do in morals, without any reference to the conduct and proceedings of others.

It is interesting to observe that Malthus refers on more than one occasion to the effect that a proposed policy will have on the condition of the poor; he affirms that it would be decisive in favor of free trade in grain if it could be shown that this would improve the standard of living of the poor.

If I were convinced, that to open our ports, would be permanently to improve the conditions of the labouring classes of society, I should consider the question as at once determined in favour of such a measure. But I own it appears to me, after the most deliberate attention to the subject, that it will be attended with effects very different from those of improvement.

However, he believes he can show that this is not the case. (The passage quoted above, for example, follows a line of argument something like this: cheaper food => larger families => more competition for work => lower wages.) Moreover, free trade in grain would have the effect of rapidly reducing British agricultural production, and expelling more thousands of agricultural workers from the farm economy. But industrial labor is unlikely to increase sufficiently to absorb this influx in the medium term:

Our commerce and manufactures, therefore, must increase very considerably before they can restore the demand for labour already lost; for a moderate increase beyond this will scarcely make up the disadvantage of a low money price of wages.

Malthus offers something of a paean to the positive effects of industrialization that is worth quoting from "Observations":

Yet, though the condition of the individual employed in common manufacturing labour is not by any means desirable, most of the effects of manufactures and commerce on the general state of society are in the highest degree beneficial. They infuse fresh life and activity into all classes of the state, afford opportunities for the inferior orders to rise by personal merit and exertion, and stimulate the higher orders to depend for distinction upon other grounds than mere rank and riches. They excite invention, encourage science and the useful arts, spread intelligence and spirit, inspire a taste for conveniences and comforts among the labouring classes; and , above all, give a new and happier structure to society, by increasing the proportion of the middle classes, that body on which the liberty, public spirit, and good government of every country must mainly depend. If we compare such a state of society with a state merely agricultural, the general superiority of the former is incontestable.

Who says economics is the dismal science?

China's agricultural history

Consider the discipline of the agricultural history of China. The following represent a sampling of research problems concerning the social and economic history of rural China in recent research:

• What were the patterns of population growth, growth of cultivated land, and growth of net output, in traditional China? (Perkins 1973)
• What was the distribution of land tenure arrangements in north China? (Arrigo 1986)
• What was the structure of rural marketing hierarchies? (Skinner 1964, 1965)
• What was the urbanization rate in 1893? (Skinner 1977)

These are all factual questions about features of economy and social institutions. The findings on these topics are, of course, fallible and revisable.

A core study in Chinese economic history is Dwight Perkins' Agricultural development in China, 1368-1968 (1973). This study plays the role in China studies that Deane and Cole (British Economic Growth 1688-1959) plays in English studies. Perkins attempts to provide estimates of population growth, cultivated land, and grain output for a period from the late fourteenth century through the mid-twentieth century. Perkins' central thesis is that China's population increased five- or six-fold between the late fourteenth and early nineteenth centuries, and that the agricultural system was able to keep pace with this increase in equal measure by expanding cultivated acreage and by raising the yield per acre (1973:13).

So what kinds of historical and empirical data provide a basis for these sorts of estimates?

For the pre-twentieth century period, Perkins' findings are largely based on primary research: Ming-Ch'ing tax records on population and cultivated acreage, local gazetteers, agricultural handbooks published over past centuries, memorials by local officials to the Emperor, and so forth. He also refers to a large volume of Chinese and Japanese research on the agrarian history of China. The local gazettes provide a great deal of information about the timing and location of markets; commodity prices; land tenure arrangements; and the activities of local elites.

For the twentieth century there are a different set of sources available: rural studies by American or European investigators (Buck, Tawney, and Gamble); data collected by the China Maritime Customs bureau; provincial gazetteers compiled by the Ministry of Industries; and a series of village studies of North China undertaken by the Japanese South Manchurian Railway Company. The earliest twentieth-century studies of the Chinese rural economy in English include Tawney (1932), Buck (1930, 1937), and Gamble (1963). Tawney and Buck provide statistical data describing the state of the Chinese rural economy in the early twentieth century. An important source in current economic research on the early twentieth century is the Mantetsu surveys. These were Japanese field studies conducted by the Japanese South Manchurian Railway Company during 1935-42, and provide extensive detail concerning the structure and organization of the rural economy in selected parts of North China.

In addition to these source materials, there are a number of core studies that have appeared since 1950 that function in much the way that was seen in English economic history. They represent a synthesis of primary data available at the time of publication which later researchers are authorized to draw upon in support of other claims. In addition to Perkins, these include Gamble (1968), Myers (1970), Rawski (1972), Skinner (1964, 1965), and Huang (1984). There is also an extensive literature in Japanese on Chinese economic history.

Current economic history of China depends on previously underutilized sources--local archives, government records, Japanese studies, and the like. Thus Huang (1984) makes extensive use of the Mantetsu surveys, Board of Punishment reports (1984:47), and county archives (the Baxian archives; 1984:51). William Rowe's study of the economic and social history of the city of Hankow is even more closely dependent on primary sources: county gazettes, records of English companies (e.g., Jardine's), and local and provincial government records. This feature may reflect the differences between stages of development of the two disciplines; in the China case there are still extensive primary sources that have not been investigated, and there are correspondingly large and important questions about Ming and Ch'ing economic history which have not been addressed, let alone resolved, in the existing literature.

Perkins pays careful attention to the problem of validating the key estimates of economic activity upon which his analysis depends, and he refers to some of the ways in which he attempts to check the validity of these sources:

I have, in fact, frequently judged the validity of data for the decades and centuries prior to the 1950's on whether these earlier figures were consistent with those for 1957 and with historical developments in the intervening periods. (Perkins 1969:10)

Perkins gives a consistency test for his estimates of population and acreage:

If the pre-modern estimates of provincial population and acreage had been arrived at by arbitrary methods, one would expect yield data derived from such figures to rise in certain periods and fall in others with no apparent pattern. . . . But most of the estimates in Table II.3 for 1850 bear a close relation to the 1957 figures. (1969:20)

Perkins' research predates the current debate about "involution" in the field of Chinese economic history; but his estimates (and those of Bozhong Li) set the parameters for much of that debate. (See "The Involution Debate" for more on this recent controversy.)

References

Arrigo, Linda Gail. 1986. Landownership Concentration in China: The Buck Survey Revisited. Modern China 12 (3):259-360.
Buck, John Lossing. 1930. Chinese farm economy. Chicago, Ill.,: The University of Chicago press.
———. 1937. Land Utilization in China. Chicago: University of Chicago Press.
Fogel, Joshua. 1987. Liberals and Collaborators: The Research Department of the South Manchurian Railway Company. Association for Asian Studies.
Gamble, Sidney D. 1968. Ting Hsien; a north China rural community. Stanford, Calif.: Stanford University Press.
Huang, Philip C. C. 1985. The Peasant Economy and Social Change in North China. Stanford: Stanford University Press.
Huang, Philip C. 1990. The peasant family and rural development in the Yangzi Delta, 1350-1988. Stanford, Calif.: Stanford University Press.
Li, Bozhong. 1998. Agricultural Development in Jiangnan, 1620-1850. London: Palgrave Macmillan.
Myers, Ramon H. 1970. The Chinese Peasant Economy. Cambridge: Harvard University Press.
Rawski, Evelyn Sakakida. 1972. Agricultural Change and the Peasant Economy of South China. Cambridge: Harvard University Press.
Rowe, William T. 1984. Hankow: Commerce and Society in a Chinese City 1796-1889 Stanford, Calif.: Stanford University Press.
Skinner, G. William. 1964-65. Marketing and Social Structure in Rural China. Journal of Asian Studies 24 (1-3).
Skinner, G. William. 1977. Regional Urbanization in Nineteenth-Century China. In The City in Late Imperial China, edited by G. W. Skinner.
Tawney, R. H. 1966 [1932]. Land and Labor in China. Boston: Beacon.

Farms in historical materialism

Materialist explanations in history generally attempt to discover fundamental features of technology and labor that impose a very deep imprint on the rest of society. Farming is almost always fundamental in this respect; the forces and relations of production are particularly visible, and the products of the farm system are fundamental to the survival of society. The standard of living of a traditional society is largely determined by agricultural productivity and the nature of the farming system -- nutrition, for example, is essentially determined by the ratio of total grain output to population. Finally, virtually all traditional economies are primarily rural; so farm life defines the conditions of ordinary social existence for the majority of the population. So let us consider a brief analysis of the farm. (A. V. Chayanov's classic treatment of the peasant economy, The Theory of Peasant Economy, written in the 1920s, remains highly valuable. Robert Allen's lifetime of research on the English farm economy is highly insightful (Enclosure and the Yeoman: The Agricultural Development of the South Midlands, 1450-1850), as is Bozhong Li's work on the farm economy of the Lower Yangzi (Agricultural Development in Jiangnan, 1620-1850).)

A farm is the basic unit of agricultural production. It represents the coordinated application of diverse factors of production in order to produce crops. The factors of production include labor; land; tools, implements, and machinery; fertilizers; and water resources and irrigation techniques. Crops include both foods (e.g., rice, wheat, millet) and raw materials (e.g., cotton, soya beans). And farms may be organized for a variety of purposes: to satisfy a family's subsistence needs, to create a profit within a market system, or to provide employment for the greatest number of rural people.

Farms in different agrarian regimes may be characterized in terms of a set of technical and social features. On the technical side we need to know what the scale of cultivation is (farm size); what techniques of cultivation are employed; what varieties of crops and seeds are available; what types of farm tools and machines are used; what types of irrigation, if any, are in use; what varieties and skills of labor are employed; what types of fertilizers are used; and what forms of agronomic knowledge are available to the farmer. (We might reduce this variety of technical features to a spectrum ranging from low-technology to high-technology farming systems.) On the social side we need to know the purpose of cultivation (family subsistence or commercial profitability); the form of land tenure in place (fixed rent, sharecropping, smallholding, etc.); the forms of labor employed (slave, serf, family labor, hired labor); the forms of supervision employed; and the processes of income distribution embodied in the agricultural system.

These features are the primary subject of research for agricultural historians such as Allen and Li. These features essentially determine the most important economic characteristics of the agricultural system. First, they determine the productivity of the farming system, whether measured in terms of land efficiency (output per hectare) or labor efficiency (output per man-day). For once we know the techniques of cultivation in a given ecological setting, it is possible to form relatively accurate estimates of output for a given input of land, labor, and capital. This set of considerations also determines what we might describe as the net rural product--the total agricultural product over and above the replacement cost of the factors of production. On this basis, Chinese historians such as Dwight Perkins attempt to estimate the overall wealth and income of late Imperial China, including estimates of quality of life for the majority of rural people (Agricultural development in China, 1368-1968).

Second, the full description of the farming system as indicated above will allow us to infer the system of surplus extraction in place; it will be possible to indicate with adequate precision how much surplus is generated and where it goes. Victor Lippit attempts to arrive at such an estimate for traditional China (Land Reform and Economic Development in China: A Study of Institutional Change and Development Finance). This in turn permits us to describe the system of rural class relations that correspond to a given farming regime.

Within this framework we can now indicate a variety of types of farms; and as a working premise, we may postulate that agrarian regimes in which different farm types are dominant will have distinctive patterns of organization and development. The following are advanced as ideal types; variants and mixed examples are possible as well. However, these types are selected as being particularly central in the development of both Asian and European traditional economies. And, significantly for the historian of social change, each farming system creates a distinctive pattern of incentives, barriers, social relations, and modes of behavior that have important consequences for historical change.

• The peasant farm. The peasant farm is small (1-10 hectares), and is organized to satisfy consumption needs of the family. It is managed and run by a peasant family using family labor. Cultivation is divided between subsistence crops and commercial crops with some risk-aversive preference for subsistence crops. There is a very low level of capital available to the peasant farmer, and cultivation is oriented towards labor-intensive techniques. Low-cost traditional techniques and tools are employed in cultivation. The peasant cultivator typically pays rent on the land he cultivates, though smallholding with clear title to the land is also possible.
• The manor. This farm is of medium size (100 hectares). It is managed by a resident lord whose aims are (1) to satisfy the consumption needs of his household, and (2) to produce a marketable surplus to generate cash income. The manor employs a sizable number of bonded laborers (serfs or slaves); it uses traditional techniques of cultivation but benefits from some economies of scale; and it employs foremen as supervisors. Part of the estate is farmed by individual families in circumstances of peasant farming.
• The capitalist farm. This farm is medium to large (50-150 hectares) and is organized to produce a profit. It is therefore located within a commercialized rural economy within which crops may be readily marketed. The farm is organized and directed by the capitalist farmer, who may either own or rent the land. The capitalist farmer has the fiscal resources needed to make capital investments in the process of cultivation; and he is oriented towards cost-cutting in considering various alternative techniques. The capitalist farm employs wage labor, where the wage is determined by local economic circumstances and the minimal cost of subsistence. The capitalist farm represents a rationalization of available techniques aimed at maximizing the profitability of the unit.
• The cooperative farm. The cooperative farm is a large unit (100-400 hectares) owned by the cultivators (75 families). This farm is oriented towards profitability; it uses the labor of the cultivators; and it is organized by a council of the cultivators. The collective has access to investment funds, and is therefore able to invest in new techniques; moreover, the collective farm benefits from economies of scale.

Significantly, this typology of farming units corresponds broadly to the classical Marxist taxonomy of modes of production: feudalism, capitalism, and socialism. The peasant farm, however, represents a form of organization of the productive forces of rural society that is overlooked in Marxist theory: call it the "peasant mode of production." The peasant mode of production may be defined as a system in which agriculture is performed on peasant farms; the bulk of the population consists of free peasant cultivators; and agricultural surpluses are extracted through rent, interest, and taxation.

In short, careful analysis of the circumstances of farming is crucial for large-scale history, including especially the history of economic development, the history of urbanization, and the history of human well-being. And, of course, the twentieth century demonstrated the centrality of peasants in the great political movements of revolution and anti-colonialism.

Primitive accumulation

Marx's treatment of the "so-called 'primitive accumulation'" is one of the most historically detailed sections in Capital: A Critique of Political Economy (Volume 1). And it is one of the most interesting parts of Capital to read as a separate piece. (Here is an electronic text of the section.) It is Marx's account of the historical processes of change in rural life of the fifteenth through eighteenth century in Britain and Ireland, through which peasants were forced off their land and the commons were enclosed. Marx believes that this separation of the peasantry from the land was a necessary condition for the development of capitalism, in that it created the conditions in which there was a pliable and abundant proletariat. This "free" proletariat was needed for the creation of the factory system and the development of manufacturing cities. So the process of primitive accumulation created the changes in social relations, property relations, and the accumulation of wealth that permitted the creation of the capital-labor relation and factory-based capitalism.

Marx sometimes puts this point in a fairly teleological way, looking at primitive accumulation as a necessary step on the road to British capitalism. (For example, he refers to this process as "the revolution that laid the foundation of the capitalist mode of production.") But it is possible, and preferable, to read Marx's analysis here less teleologically, as simply a detailed account of some of the crucial but contingent changes that took place in rural social relations during these centuries, without importing the idea that these changes were functionally related to the later development of capitalism. And read non-teleologically, the section holds up fairly well in relation to modern historical scholarship.

Marx describes the basic social relations of British rural life in the fifteenth century in these terms, as a freeholding peasantry with access to substantial common lands, pasture, and forest:

The immense majority of the population consisted then, and to a still larger extent, in the 15th century, of free peasant proprietors, whatever was the feudal title under which their right of property was hidden. In the larger seignorial domains, the old bailiff, himself a serf, was displaced by the free farmer. The wage-labourers of agriculture consisted partly of peasants, who utilised their leisure time by working on the large estates, partly of an independent special class of wage-labourers, relatively and absolutely few in numbers. The latter also were practically at the same time peasant farmers, since, besides their wages, they had allotted to them arable land to the extent of 4 or more acres, together with their cottages. Besides they, with the rest of the peasants, enjoyed the usufruct of the common land, which gave pasture to their cattle, furnished them with timber, fire-wood, turf, &c.

And he identifies the crucial turn towards expropriation of the free peasant proprietor:

In insolent conflict with king and parliament, the great feudal lords created an incomparably larger proletariat by the forcible driving of the peasantry from the land, to which the latter had the same feudal right as the lord himself, and by the usurpation of the common lands. The rapid rise of the Flemish wool manufactures, and the corresponding rise in the price of wool in England, gave the direct impulse to these evictions. The old nobility had been devoured by the great feudal wars. The new nobility was the child of its time, for which money was the power of all powers. Transformation of arable land into sheep-walks was, therefore, its cry.

(It is interesting to recall that Oliver Goldsmith describes the eighteenth-century version of this process in Ireland in his poem, The Deserted Village (1770):

Sweet smiling village, loveliest of the lawn,
Thy sports are fled, and all thy charms withdrawn;
Amidst thy bowers the tyrant's hand is seen,
And Desolation saddens all thy green:
One only master grasps the whole domain,
And half a tillage stints thy smiling plain.
No more thy glassy brook reflects the day,
But, choked with sedges, works its weedy way;
Along thy glades, a solitary guest,
The hollow-sounding bittern guards its nest;
Amidst thy desert walks the lapwing flies,
And tires their echoes with unvaried cries:
Sunk are thy bowers in shapeless ruin all,
And the long grass o'ertops the mouldering wall
And, trembling, shrinking from the spoiler's hand,
Far, far away thy children leave the land.

Ill fares the land, to hastening ills a prey,
Where wealth accumulates, and men decay.
Princes and lords may flourish, or may fade;
A breath can make them, as a breath has made:
But a bold peasantry, their country's pride,
When once destroy'd, can never be supplied. )

And Marx emphasizes the coercive nature of this expropriation of the yeoman peasant:

Even in the last decade of the 17th century, the yeomanry, the class of independent peasants, were more numerous than the class of farmers. They had formed the backbone of Cromwell’s strength, and, even according to the confession of Macaulay, stood in favourable contrast to the drunken squires and to their servants, the country clergy, who had to marry their masters’ cast-off mistresses. About 1750, the yeomanry had disappeared, and so had, in the last decade of the 18th century, the last trace of the common land of the agricultural labourer. We leave on one side here the purely economic causes of the agricultural revolution. We deal only with the forcible means employed.

A crucial component of the "primitive accumulation", in Marx's interpretation, was the abolition of common property, culminating in the enclosure acts in the seventeenth century:

Communal property — always distinct from the State property just dealt with — was an old Teutonic institution which lived on under cover of feudalism. We have seen how the forcible usurpation of this, generally accompanied by the turning of arable into pasture land, begins at the end of the 15th and extends into the 16th century. But, at that time, the process was carried on by means of individual acts of violence against which legislation, for a hundred and fifty years, fought in vain. The advance made by the 18th century shows itself in this, that the law itself becomes now the instrument of the theft of the people’s land, although the large farmers make use of their little independent methods as well. The parliamentary form of the robbery is that of Acts for enclosures of Commons, in other words, decrees by which the landlords grant themselves the people’s land as private property, decrees of expropriation of the people. Sir F. M. Eden refutes his own crafty special pleading, in which he tries to represent communal property as the private property of the great landlords who have taken the place of the feudal lords, when he, himself, demands a “general Act of Parliament for the enclosure of Commons” (admitting thereby that a parliamentary coup d’état is necessary for its transformation into private property), and moreover calls on the legislature for the indemnification for the expropriated poor.

We even are afforded a glimpse of the economist's view of the rationality of enclosure. According to John Arbuthnot, enclosure and the creation of private farms is a more productive use of land and labor:

Let us hear for a moment a defender of enclosures and an opponent of Dr. Price. “Not is it a consequence that there must be depopulation, because men are not seen wasting their labour in the open field.... If, by converting the little farmers into a body of men who must work for others, more labour is produced, it is an advantage which the nation” (to which, of course, the “converted” ones do not belong) “should wish for ... the produce being greater when their joint labours are employed on one farm, there will be a surplus for manufactures, and by this means manufactures, one of the mines of the nation, will increase, in proportion to the quantity of corn produced.”

And here Marx describes the final stages of the "rationalization" of agriculture, in the expropriation of the Scots:

The last process of wholesale expropriation of the agricultural population from the soil is, finally, the so-called clearing of estates, i.e., the sweeping men off them. All the English methods hitherto considered culminated in “clearing.” As we saw in the picture of modern conditions given in a former chapter, where there are no more independent peasants to get rid of, the “clearing” of cottages begins; so that the agricultural labourers do not find on the soil cultivated by them even the spot necessary for their own housing. But what “clearing of estates” really and properly signifies, we learn only in the promised land of modern romance, the Highlands of Scotland. There the process is distinguished by its systematic character, by the magnitude of the scale on which it is carried out at one blow (in Ireland landlords have gone to the length of sweeping away several villages at once; in Scotland areas as large as German principalities are dealt with), finally by the peculiar form of property, under which the embezzled lands were held.

Eventually we come to the point where industrial capitalism is feasible and there is a "free" proletariat available for labor:

It is not enough that the conditions of labour are concentrated in a mass, in the shape of capital, at the one pole of society, while at the other are grouped masses of men, who have nothing to sell but their labour-power. Neither is it enough that they are compelled to sell it voluntarily. The advance of capitalist production develops a working-class, which by education, tradition, habit, looks upon the conditions of that mode of production as self-evident laws of Nature. The organisation of the capitalist process of production, once fully developed, breaks down all resistance.

And, of course, Engels picks up the story from the perspective of nineteenth-century Manchester and Birmingham, and the conditions of squalor and oppressive factory labor that resulted. (See an earlier posting on Engels's sociology of the proletarian city.)

The English Marxist historians have devoted a lot of their attention to this period of British social history. (Harvey Kaye gives a very good account of the work of this generation of Marxist historians in The British Marxist Historians.) Maurice Dobb is one of the English socialist historians who has retraced Marx's steps on this subject in some detail. One of Dobb's important books, Studies In The Development Of Capitalism (1963), treats this process of the decline of the English peasantry and the rise of the proletariat, making use of the historical scholarship of the first part of the twentieth century. Some of Rodney Hilton's research on this period can be found in Class Conflict and the Crisis of Feudalism: Essays in Medieval Social History (Rev). Several generations later, Robert Brenner gave a somewhat different interpretation of the history of agrarian change in England from the sixteenth to the eighteenth centuries; his views were developed in the splendid journal Past & Present and were separately published along with a round of critical reactions in The Brenner Debate: Agrarian Class Structure and Economic Development in Pre-industrial Europe.

Economic history analyzed

The history of a region or people encompasses a multitude of aspects of social life: culture, religion, political institutions, social movements, environmental change, technology, population—and the circumstances and processes of economic change that the region undergoes. One does not need to be a reductionist in order to observe that the economic circumstances a society experiences, and the processes of change that these circumstances undergo, have a profound influence on other aspects of social and cultural change. Improved agricultural productivity can support population growth; it can enhance the coercive power of state institutions; and it can make possible the flourishing of intricate institutions of religion and education. Likewise, the constraints created by slow or negative economic productivity growth in a region can stifle the development of other important social processes. So economic history, as a discipline within history more broadly, is a crucially important field of historical inquiry.

Yet the foundations of the discipline of economic history are controversial. Economic historians do not yet agree on the role of mathematical economic theory within their discipline, or the relationships that should obtain between quantitative and qualitative data, or the role of social theories of causal factors in explaining economic change, or the connections that should be established between economic historical research and other fields of social or cultural history.

What is the intellectual task of an “economic history” of a region or country? To start, we might say that it is to provide an evidence-based description of the main economic characteristics of the country or region over a defined period of time: the kinds and levels of agricultural and manufacturing products that are produced, the technologies and institutions through which production and distribution occurs, the size of the population, and the level of material well-being that is experienced by the population. And, second, the task of economic history is to arrive at causal hypotheses that may serve as explanations of some of the patterns of economic change that are discovered.

Consider the variety of questions that need to be addressed by an economic history of a region or country:

• Demography. What was the absolute population size and distribution at various time points during the period? What were the trends of population growth during the peri¬od? How much urbanization occurred during the period?
• Inputs and technology. How much land was under cultivation? What crops and products were in production? What fertilizer technologies were in use? How much irrigation was available, and what was the trend of extension of land and irrigation?
• Property relations and control of labor. What forms of tenancy and land ownership were in place? How were these arrangements changing during the time period? What forms of labor control were in use? Was there a tendency of change in the conditions and extent of wage labor?
• Productivity. What was the absolute size of the production of central commodities—rice, wheat, cotton? What were the factor productivities for land, labor, capital, or animal power? What trends existed in these quantities?
• Prices and market conditions. How much agricultural activity took place within functioning markets for crops, grain, textiles, and handicraft goods? What were the prices of these goods over time? How sensitive were farmers to changing market conditions?
• Human welfare. What were the income levels and food security of various groups: landless workers, smallholding peasants, tenants and other groups? How extensive were income inequalities within the economy? Where were economic surpluses going? What was the trend of real welfare and inequalities?
• Causal factors. What are the causal relationships that obtain between various large factors: technology, social relations, property systems, state, demographic regimes, and international relations?

Explanation requires a theory of underlying causal mechanisms. What theoretical resources are available to the economic historian to explain patterns and singularities of economic change? It is evident that economic outcomes are the result of human behavior within the context of environmental circumstances and institutional settings. Human behavior, however, is not rigidly segregated into “economic,” “cultural,” and “social” behavior; rather, behavioral outcomes are influenced by all these kinds of factors. So economic history cannot be restricted to the theories associated with neoclassical economics. Rather, the economic historian needs to examine the economic phenomena under study within the broader social and environmental context in which this behavior takes place. And that means that the economic historian must be as much a social historian, a sociologist, or an ethnographer as he is an economist; he needs to pay as much attention to the social and political context of economic trends as he does to the mathematics of equilibrium or the idealized workings of a market.

Marc Bloch’s history of medieval French agriculture (1931) offers a good illustration of the value of a broadly contextualized approach to a region’s economic history (French Rural History: An Essay on Its Basic Characteristics). Bloch surveys the main social institutions and technologies that were in use, and attempts to explain some of the large patterns that became evident (for example, field shape and the geographical diffusion of the wheeled plough). Bloch’s explanation invokes such varied factors as the nature of the soils across France, the availability and timing of technical innovations in the design of the plough, and the nature of village communities in different parts of France. And he makes ingenious use of a wide range of historical sources to permit him to come to assessments of various economic and institutional facts.

Well-developed social theories give us a basis for demonstrating how various factors could be causally relevant. It is then the task of empirical, historical, and theoretical research to arrive at justified conclusions about causation. But theories don't provide scripts for "necessary" processes of economic development. In fact, it is entirely possible that different combinations of causal factors are of primary importance in different historical settings. Historical change is conjunctural and contingent. The general point is that institutions and circumstances matter, and that institutional arrangements in different times and places may impose limits or opportunities that discourage or favor some pathways of development over others. Instead of expecting one grand course of development, we ought to expect a shifting fabric of contingent, fluctuating path-dependent processes.

Two current economic historians whose work I particularly admire are Bozhong Li, whose careful and deep empirical studies of the economic history of the lower Yangzi delta provide a foundation for a much more finegrained economic history of China (Agricultural Development in Jiangnan, 1620-1850, link); and Robert Allen, whose careful studies of the English farm economy and the standard of living in the early modern period are a paradigm of excellent empirical and analytical work in economic history (link, link).

(There is more on this topic on my research webpage, including Epistemological Issues in Economic History and Eurasian Historical Comparisons.)

Historical populations

Historical demography concerns itself with two families of questions -- factual description of population behavior and explanation of the patterns that are observed. On the descriptive side, historical demographers are concerned with estimating the absolute population size of a group or region -- Tuscany in 1400, New England in 1700, the Anasazi in 1000. And they are interested in measuring the underlying demographic rates -- for example, natality, mortality, age of marriage, nutritional and health status, or sex ratios.

And, to make useful comparisons, these measurements need to be linked to the same variables over time and comparable populations over the same time period. This set of observations permits the discovery of trends and patterns -- steady population growth in group A simultaneous with flat population size in group B; high rate of growth for C during 1700-1750 followed by slow growth in the next 50 years; high population density in agricultural region D compared with low density in agricultural region E.

The descriptive task is a difficult one. But it immediately raises a host of explanatory questions. What are the social mechanisms that conjoin to bring about the complex observed demographic outcomes? Population outcomes are the result of myriad choices by anonymous individuals; and yet the outcomes are highly patterned. The task of explanation arises when we try to make sense of the demographic behavior and outcomes that are discovered. What are the material and social circumstances that explain the trends we observe and the differences that emerge across populations?

Thomas Malthus offered an explanation of large population trends based on both material and social factors: the food-production capacity of the land inhabited and the norms and institutions regulating family formation and child-bearing. Malthus's theories suggest relating population size and growth rates to measures of economic prosperity -- for example, measures of the real wage for farm laborers over time, measures of the fluctuations in availability of jobs, or measures of available land per unit of population. The classic Malthusian argument predicts that population growth will lag economic fluctuation: rising wages encourage higher fertility and falling wages discourage fertility. And if population continues to rise in declining economic circumstances, then eventually mortality will rise -- again suppressing population growth.

Source: Massimo Livi Bacci, A Concise History of World Population, P. 70.

The simple Malthusian theory, then, postulates that population expands to the limit created by available sources of food. So increases in farm productivity and increases in arable land should give rise to an increase in population, whereas a rising population density on a given territory will eventually lead to the "negative" checks of famine and nutrition-based mortality if not pre-empted by the preventive checks of reduced natality.

The question of scale, both temporal and regional, makes a large difference in the explanatory quest. We can frame the question of group demographic behavior from the micro- to the macro-levels -- from a handful of Alpine villages in the 16th century (population 5000) to the Yangzi delta in the 18th century (population 300 million). Micro-trends may have a different explanations than macro-trends.

It might be maintained that the Malthusian theory is only well-suited to the largest scale: large territory and long timeframe. In the longterm it is almost tautological to observe that a population will expand to the limits created by environmental resources. But long before those limits are approached, there is ample space for the workings of social arrangements to modulate population behavior. Alternative economic and demographic institutions create different demographic profiles. Birth spacing (reflecting family choices and cultural values) can dramatically influence total fertility. Technologies of birth control -- both traditional and modern -- can move a population from high-fertility to medium- or low-fertility. So historical demography requires social explanation based on discovery of concrete and specific institutions in local circumstances, not simply broad-brush hypotheses about resource limits.

Recent studies of population behavior across the expanse of Eurasia confirm this point. The Eurasian Demography Project finds that there is a wide range of demograhic patterns across the locales of Eurasia, from Finland to Italy to China to Japan. And the Malthusian picture works only at the coarsest level; in particular, it turns out that the Chinese population did not teeter on the edge of Malthusian crisis.

(See Eurasian Historical Comparisons for more on the Eurasian Demography Project.)

"Moral economy" as a historical social concept

The concept of a "moral economy" has proved useful in attempting to describe and explain the contentious behavior of peasants in response to onerous social relations. Essentially, it is the idea that peasant communities share a set of normative attitudes concerning the social relations and social behaviors that surround the local economy: the availability of food, the prices of subsistence commodities, the proper administration of taxation, and the operation of charity, for example. This is sometimes referred to a "subsistence ethic": the idea that local social arrangements should be structured in such a way as to respect the subsistence needs of the rural poor. The associated theory of political behavior holds something like this: peasant communities are aroused to protest and rebellion when the terms of the local subsistence ethic are breached by local elites, state authorities, or market forces.

Here I want to highlight this concept by asking a few foundational questions. Fundamentally, what kind of concept is it? How does it function in social interpretation, description, or explanation? And how does it function as a component of empirical investigation?

The concept of moral economy was extensively developed by E. P. Thompson in The Making of the English Working Class (1961) and an important essay, "The Moral Economy of the English Crowd in the Eighteenth Century," originally published in Past and Present in 1971 and included in Customs in Common: Studies in Traditional Popular Culture. The concept derives from Thompson's treatment of bread riots in eighteenth century Britain. In MEWC Thompson writes:

In 18th-century Britain riotous actions assumed two different forms: that of more or less spontaneous popular direct action; and that of the deliberate use of the crowd as an instrument of pressure, by persons "above" or apart from he crowd. The first form has not received the attention which it merits. It rested upon more articulate popular sanctions and was validated by more sophisticated traditions than the word "riot" suggests. The most common example is the bread or food riot, repeated cases of which can be found in almost every town and county until the 1840s. This was rarely a mere uproar which culminated in the breaking open of barns or the looting of shops. It was legitimised by the assumptions of an older moral economy, which taught the immorality of any unfair method of forcing up the price of provisions by profiteering upon the necessities of the people. (MTWEC, 62-63)

After describing a number of bread riots in some detail, Thompson writes, "Actions on such a scale ... indicate an extraordinarily deep-rooted pattern of behaviour and belief .... These popular actions were legitimised by the old paternalist moral economy" (66). And he closes this interesting discussion with these words: "In considering only this one form of 'mob' action we have come upon unsuspected complexities, for behind every such form of popular direct action some legitimising notion of right is to be found" (68). And Thompson often describes these values as "traditional" or "paternalist" -- working in opposition to the values and ideas of an unfettered market; he contrasts "moral economy" with the modern "political economy" associated with liberalism and the ideology of the free market.

In "The Moral Economy of the Crowd" Thompson puts his theory this way:

It is possible to detect in almost ever eighteenth-century crowd action some legitimising notion. By the notion of legitimation I mean that the men and women in the crowd were informed by the belief that they were defending traditional rights or customs; and, in general, that they were supported by the wider consensus of the community. On occasion this popular consensus was endorsed by some measure of licence afforded by the authorities. More commonly, the consensus was so strong that it overrode motives of fear or deference. ("Moral Economy," CIC 188)

It is plain from these passages that Thompson believes that the "moral economy" is a real historical factor, consisting of the complex set of attitudes and norms of justice that are in play within this historically presented social group. As he puts the point late in the essay, "We have been examining a pattern of social protest which derives from a consensus as to the moral economy of the commonweal in times of dearth" (247).

So the logic of Thompson's ideas here seems fairly clear: there were instances of public disorder ("riots") surrounding the availability and price of food, and there is a hypothesized "notion of right" or justice that influenced and motivated participants. This conception of justice is a socially embodied historical factor, and it partially explains the behavior of the rural people who mobilized themselves to participate in the disturbances. He recapitulates his goal in the essay, "Moral Economy Reviewed" (also included in Customs in Common) in these terms: "My object of analysis was the mentalité, or, as I would prefer, the political culture, the expectations, traditions, and indeed, superstitions of the working population most frequently involved in actions in the market" (260). These shared values and norms play a key role in Thompson's reading of the political behavior of the individuals in these groups. So these hypotheses about the moral economy of the crowd serve both to help interpret the actions of a set of actors involved in food riots, and to explain the timing and nature of food riots. We might say, then, that the concept of "moral economy" contributes both to a hermeneutics of peasant behavior and a causal theory of peasant contention.

Now move forward two centuries. Another key use of the concept of moral economy occurs in treatments of modern peasant rebellions in Asia. Most influential is James Scott's important book, The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia. Scholars of the Chinese Revolution borrowed from Scott in offering a range of interpretations of peasant behavior in the context of CCP mobilization; for example, James Polachek ("The Moral Economy of the Kiangsi Soviet" (1928-34). Journal of Asian Studies 1983 XLII (4):805-830). And most recently, Kevin O'Brien has made use of the idea of a moral economy in his treatment of "righteous protest" in contemporary China (Rightful Resistance in Rural China). So scholars interested in the politics of Asian rural societies have found the moral economy concept to be a useful one. Scott puts his central perspective in these terms:

We can learn a great deal from rebels who were defeated nearly a half-century ago. If we understand the indignation and rage which prompted them to risk everything, we can grasp what I have chosen to call their moral economy: their notion of economic justice and their working definition of exploitation--their view of which claims on their product were tolerable and which intolerable. Insofar as their moral economy is representative of peasants elsewhere, and I believe I can show that it is, we may move toward a fuller appreciation of the normative roots of peasant politics. If we understand, further, how the central economic and political transformations of the colonial era served to systematically violate the peasantry's vision of social equity, we may realize how a class "of low classness" came to provide, far more often than the proletariat, the shock troops of rebellion and revolution. (MEP, 3-4)

Scott's book represents his effort to understand the dynamic material circumstances of peasant life in colonial Southeast Asia (Vietnam and Burma); to postulate some central normative assumptions of the "subsistence ethic" that he believes characterizes these peasant societies; and then to explain the variations in political behavior of peasants in these societies based on the moments of inconsistency between material conditions and aspects of the subsistence ethic. And he postulates that the political choices for action these peasant rebels make are powerfully influenced by the content of the subsistence ethic. Essentially, we are invited to conceive of the "agency" of the peasant as being a complicated affair, including prudential reasoning, moral assessment based on shared standards of justice, and perhaps other factors as well. So, most fundamentally, Scott's theory offers an account of the social psychology and agency of peasants.

There are several distinctive features of Scott's programme. One is his critique of narrow agent-centered theories of political motivation, including particularly rational choice theory. (Samuel Popkin's The Rational Peasant: The Political Economy of Rural Society in Vietnam is the prime example.) Against the idea that peasants are economically rational agents who decide about political participation based on a narrowly defined cost-benefit analysis, Scott argues for a more complex political psychology incorporating socially shared norms and values. But a second important feature is Scott's goal of providing a somewhat general basis for explanation of peasant behavior. He wants to argue that the subsistence ethic is a widely shared set of moral values in traditional rural societies -- with the consequence that it provides a basis for explanation that goes beyond the particulars of Vietnam or Burma. And he has a putative explanation of this commonality as well -- the common existential circumstances of traditional family-based agriculture.

One could pull several of these features apart in Scott's treatment. For example, we could accept the political psychology -- "People are motivated by a locally embodied sense of justice" -- but could reject the generalizability of the subsistence ethic -- "Burmese peasants had the XYZ set of local values, while Vietnamese peasants possessed the UVW set of local values."

This programme suggests several problems for theory and for empirical research. Are there social-science research methods that would permit us to "observe" or empirically discern the particular contents of a normative worldview in a range of different societies, in order to assess whether the subsistence ethic that Scott describes is widespread? Are peasants in Burma and Vietnam as similar as Scott's theory postulates? How would we validate the implicit theory of political motivation that Scott advances (calculation within the context of normative judgment)? Are there other important motivational factors that are perhaps as salient to political behavior as the factors invoked by the subsistence ethic? Where does Scott's "thicker" description of peasant consciousness sit with respect to fully ethnographic investigation?

So to answer my original question -- what kind of concept is the "moral economy"? -- we can say several things. It is a proto-theory of the theory of justice that certain groups possess (18th-century English farmers and townspeople, 20th-century Vietnamese peasants). It implicitly postulates a theory of political motivation and political agency. It asserts a degree of generality across peasant societies. It is offered as a basis for both interpreting and explaining events -- answering the question "What is going on here?" and "Why did this event take place?" In these respects the concept is both an empirical construct and a framework for thinking about agency; so it can be considered both in terms of its specific empirical adequacy and, more broadly, the degree of insight it offers for thinking about collective action.