Louise A. Tilly

Louise A. TillyPresident of the Association, 1993

Presidential address delivered at the American Historical Association meeting in San Francisco on January 7, 1994. American Historical Review, Vol. 99, No. 1. (February 1994): 1-17.


Many of us have been moved by Edward Thompson’s death last August to reflect on his deeply felt social project in history and rededicate ourselves to its principles.1 Thompson insisted on the agency of ordinary people faced by far-reaching economic and social changes. He expressed passionate concern to rescue working men and women “from the enormous condescension of posterity.”2 Although the Thompson of The Making of the English Working Class was dubious about abstract theory, he was much more than an inspired rhetorician. When he spoke of “class as process,” relationships among classes, culture as a “way of living” embedded in the material world, and insisted on struggle as the engine of history, he spoke as a Marxist historian, even as he withdrew his support from the party and regimes whose practice he believed betrayed the tradition.

Like most great historians, Thompson had many sides. There was the Thompson who saw history as the discipline of context and argued that looking at the local level was critical to any synthesis.3 And the Thompson who brilliantly reminded his readers over and over of the connections between apparently disparate forms of collective action—such as eighteenth-century food protest or nineteenth-century Luddite machine-breaking—and the far-reaching structural transition from market to industrial capitalism, shared understandings of political economy and the law, and the networks through which these traditions were communicated.

From the early 1960s, Thompson and others inspired a generation of social historians who challenged conventional historical concepts and methodology and expanded the scope of historical research and writing.4 Since the mid-1980s, however, historians have increasingly questioned this populist, critical, and realist approach in social history. Local case studies have proved to be interesting and important but difficult to generalize. Social historians encounter stubborn empirical details that cannot be compared simply to cases from other settings or time periods; analyses of processes and structures are only with difficulty connected to detailed cases. Many nevertheless have persisted in social history, fine-tuning or combining new types of sources or interpretation with the old. Others have abandoned efforts to explain, retreated to description, and simply emphasized the complexity of the past. And still others have rejected the very possibility of explanation and adopted a radical skepticism toward any reliable knowledge of the past, or they have borrowed linguistic and literary critical approaches that take them far from the social and economic history that their mentors espoused.

In this epistemological crisis, Thompson can still inspire us, even if we do not completely agree with him. With his work in mind, I wish to present a vision of social history that focuses on connections between structure and action, individuals and processes, the past and the present, and settings distant in space. Before offering some general conclusions, I will illustrate this vision through a discussion of cotton textile industrialization up to the mid-nineteenth century and its effect on workers and their families.

The social history I envision studies past economic, political, and social structures, as well as collectivities—groups defined by class, occupation, sex, family position, geographic location, ethnicity, ideological commitment, religion, and so on. It studies the connections between and among structures, processes of change, and human action. It posits an interdependence of structure and action—human agents produce structures, intentionally or not, even as structures facilitate or constrain human action.5 Individual actors (or groups) build structures and in turn are affected by structure through institutions such as families, courts, legislatures, churches, schools, firms, employers’ associations, labor markets, unions, and the press. And all of these institutions operate within cultural contexts of shared understandings. Social history entails historical realism insofar as it presumes that popular experience is accessible to historical analysis, that this experience is coherent and cumulative, and that human agency makes a historical difference.

Social history gives serious attention to time and place. Narrative is therefore central in describing and interpreting temporal sequences of events. But narrative cannot be a complete explanation, just as causal explanation based on structural analysis cannot tell the whole story. Both are needed to describe past structures and events and to explain their connections; together, they communicate the essence of history.

Epistemologically, social historians argue not that complete or precise knowledge about the past is possible but that we can attain a modest knowledge, based on the evidence we can unearth, critically evaluated. We emphasize systematic variation among cases, not typical ones, and share a belief in an accessible reality. We focus on connections in our explanations of change or continuity. But what kinds of connections?

I have already discussed social history’s defining connections, those among structures, processes, and human agency. Two other sets of connections are also central to our project. These are, first, spatial links between groups, regions, political units connected by trade, production, migration, religion, or political relations. And, second, temporal connections between past and present. Since the present is always becoming the past, and the past constrains or facilitates present action, it is the connections between past and present—continuities or breaks—that matter. I will illustrate these three sets of connections more fully below in historical vignettes on the interplay of textile industrialization and the familial social relations of workers in India, England, and France.6

The interaction of pre-factory textile production in India, England, and France and the outcomes for their hand spinners and weavers of the Industrial Revolution illustrate three forms of connections—spatial, temporal, and those joining structure with people’s action—in popular history on a world scale. Although historians sometimes portray such cases as at best parallel experiences distantly linked by an international market, a close examination reveals ongoing interplay among the histories.7

I begin in India, whose political, economic, and social histories are multiply connected to those of England, where the first Industrial Revolution took place.8 Politically, of course, India was not a unified polity in the seventeenth and eighteenth centuries, although the Mughal empire was the largest political entity in south Asia. Following the Portuguese, who had seized Goa a century before, private Dutch, French, and English trading companies established commercial relations with Indian merchants in port cities during the seventeenth century. Revolts against the Mughal emperor served as a wedge for the British to expand their claims, decisively defeat French, Indian, and Dutch armies in the 1750s and 1760s, and establish direct rule or overlordship in most of India in the first decade of the nineteenth century. Conquest of the rest came in the 1840s, dissolution of the East India Company and direct British rule in the 1860s.

India was at least doubly connected to the English Industrial Revolution and its early manifestation in the cotton textile industry. The first connection runs through a highly specialized long-distance trade from India to Britain organized by commercial capitalists in the seventeenth century, when the importation in Britain of Indian pure cotton textiles spurred an expanding market for these products. In the eighteenth century, regulations prohibiting the importation of Indian cotton were passed. By the end of the century, the English cotton textile industry had successfully substituted home production for imports through revolutionary technologies that interacted cumulatively and through new organizations of production. The second connection, which began in the late eighteenth century, runs through British export of cotton yarn to India; in the course of the nineteenth century, cotton cloth was exported as well. British land policy promoted Indian agriculture, not manufacturing, and India’s textile sector was permitted no tariff protection against British imports. Indigenous cotton manufactures declined.

Henry St. George Tucker, who spent years in India for the East India Company, later becoming one of its directors, offered this analysis in 1823: “[C]otton fabrics, which hitherto constituted the staple of India, have not only been displaced in [England], but we actually export our cotton manufactures to supply a part of the consumption of our Asiatic possessions. India is thus reduced from the state of manufacturing to that of an agricultural country.”9 Such an interpretation was adopted by Indian nationalists at the beginning of this century, but it has been widely debated by Indian and Western historians in the last forty years or so; a closer look at the history of the Indian textile industry is in order.

There were four widely separated major textile manufacturing regions in India in the seventeenth and eighteenth centuries; each had its own specialties, based on the tastes of its market region and the availability of different qualities of raw cotton fibers. Weavers in these specialized regions and others dispersed throughout the country who served mostly local markets not only met the home market’s demand but also supplied a vigorous long-distance commerce. Exports included the fine light muslin of Dakha (Bengal), “painted” chintzes, and printed calicoes, as well as coarser cloth destined for more humble users, like that traded in the Indonesian islands or transshipped from Europe and the Middle East to Africa in the slave trade and to the Caribbean and North American plantations for the clothing of slaves. West European countries were the chief trading partners of Bengal in the first half of the eighteenth century.10

Contemporary observers and historians alike agree on the fine quality of certain types of Indian cotton cloth. For example, Robert Orme exclaimed hyperbolically in his Historical Fragments of the Mogul Empire (1783):

The women ... spin the thread designed for the cloths, and then deliver it to the men, who have fingers to model it as exquisitely as these have prepared it. For it is a fact, that the tools which they use are as simple and plain as they can be imagined to be. The rigid, clumsy fingers of an European would scarcely be able to make a piece of canvass, with instruments which are all that an Indian employs in making a piece of cambric.11

K. N. Chaudhuri concludes that the “technological superiority of Asian fabrics rested largely on human skills transmitted on the basis of hereditary knowledge.” Robert Orme’s use of the word cambric to describe the fine Indian cotton suggests another spatial connection; just as the word calico carried the name of Calicut (a city in southwestern India) to England and France, the original meaning of cambric (an English word) was the fine linen cloth of Cambrai, in northern France. Thus a connection to one of the chief textile-producing regions of France.

Eighteenth-century Indian weavers were specialized skilled craftsmen belonging to castes found in all parts of India. They wove cloth in styles defined by social convention on simple looms; bleaching, printing, glazing, and other finishing processes were done by specialists subcontracted by urban weavers or in rural settings by traders in local markets. Despite their skill (which would appear to put them in a strong economic position), they experienced considerable geographic mobility; such mobility was sometimes due to shifting market opportunities, other times to warfare, crop failures, or other natural disasters. Merchants and brokers served as intermediaries between the English East India Company and theweavers.12

The system was part of the development of commercial capitalist structures on a world scale. In seventeenth and early eighteenth-century India, textile production was facilitated by an economy-wide division of labor, with a growing agricultural sector keeping pace with the expanding textile manufactures. The Indian system had something in common with European proto-industrialization, if we understand the latter not as the first stage of industrialization but as the commercially controlled expansion of labor-intensive, low-cost production in households or small shops of textiles and other products for distant markets. However, instead of giving weavers yarn with which to produce the desired lengths of fabric and controlling the finishing and marketing of their cloth, Indian brokers made cash advances that covered both purchase of yarn and living expenses for the weaver and his dependents while the cloth was being woven. The system was dynamic and responded to market demand and long-distance trade possibilities, but weavers apparently never had the possibility of accumulating capital and moving into commerce, as they sometimes did in England.13 Instead, over the eighteenth century, the European trading companies came more and more to dominate the local Indian economies and pull them into an international division of labor with its centers in Western Europe.

The Indian economic and social order was shaken at the end of the eighteenth century by wars and famines, as the weavers’ lot became more precarious, especially in the west and south. At times, they turned to agricultural labor; in their own industry, they had by then largely become simple wage laborers. Chaudhuri concludes that the very success of Indian cotton manufactures in international markets and the long-established patterns of specialization and hierarchical control made it difficult for Indian middlemen and producers to change their ways in response to the rising external challenge of English technological innovation.14

Most scholars agree that, like many European proto-industrial regions, India was not on the verge of an industrial revolution but that its indigenous commercial capitalism was undermined by privileged English competition. The handweaving of cotton fabrics did not disappear in India; over the course of the nineteenth century, and into the twentieth, there were crisis periods, some linked with changing demand for the products, others with famines and disease (which killed both weavers and consumers). Coarse fabrics continued to be woven in most areas for local consumption, and the finest specialties also endured; the proletarianization of weavers of coarse cloth continued, but they hung on by reducing their prices and immiserating themselves and their families. According to the 1901 census of India, there were still 5.8 million handloom weavers and only 350,000 workers in mechanized mills. In the same period, Indian mill production accounted for about 11 percent of consumption, handlooms for 23 percent, and imports (largely from Britain) for 66 percent. Handweaving continued to be a source of livelihood for many, but it was no longer the link to the world economy than it had earlier been.15

Up to the end of the eighteenth century, hand spinning for commercial weavers, a female occupation permissible in all castes, produced yarn of higher quality than could be produced in contemporary England.16 But English mule-spun yarn, which became available in the century’s last decades, thoroughly undermined the hand-spun Indian product. Dr. Francis Hamilton Buchanan’s report of his travels in northern India and Bengal, from 1808 to 1815, describesthe parlous condition of women spinners in the early period of English competition.17 Since these women had often been spinning as members of weavers’ households, some doubtless began to spend time at the loom instead of the spinning wheel, or they sought work in agriculture. The cash contribution that spinners could make to their household budgets was severely compromised as hand spinning virtually disappeared in the nineteenth century. (However, women in the households of the remaining hand weavers probably provided them at least some of their yarn.) Widowed spinners who had been rejected by their in-laws were particularly disadvantaged, because they had to support themselves and their children. As these women were losing their means of livelihood, British officials and Indian intellectual modernizers abolished sati (widow immolation, a custom most often practiced among upper-caste Hindus) but did nothing for widows displaced by imported, industrially spun yarn.18 The few women who continued to spin were those who simply had no alternative way to live.19

To what extent and in what ways were parent-child and adult gender relations among the common people of India modified by the changing fortunes of its cotton manufactures in the late eighteenth and early nineteenth centuries? Putting together sparse evidence about family life, we know that members of both rural and urban families in the period were likely to think in terms of group rather than individual well-being. Agriculturalists and craftspeople depended on cooperation in familial interest, a tendency that was supported culturally by women’s marriage at a very young age and the common Hindu joint family form (ideally, three generations but sometimes brothers, their wives, and offspring co-residing). Because daughters would leave their family of origin at marriage, and because of the preference for sons that was the corollary of this custom, girls were often neglected compared to their brothers. Marriages were planned without consultation of the young people involved. And wives were unlikely to develop much sense of independence when they came as child brides (no more than fifteen and often considerably younger) into their husband’s extended household and grew up with its expectations and values.20

George Forster, an intrepid eighteenth-century world traveler, noted: “The entire system of domestic ordinary and economy of the Hindoos is founded on a firm yet simple basis ... the wife depends for ... most of the ordinary accommodations of her life, on the immediate existence of her husband.”21 Although Forster may have based his testimony on observing upper-class families, wives of urban or rural workers living in Hindu joint family households were also dependent on their husband’s survival, because they were not permitted to remarry, yet they had few rights as widows living with their in-laws. A wife worked with other women, ate separately from her husband, shared child care with female kin, and saw her husband little. She was subordinate not only to him but to all others in his family of origin.

Indian women became worse off in terms of the family division of labor and resources when they abandoned hand spinning and lost the possibilities for earning that went with it. The agricultural labor available to them was not as well paid and required less skill; it was an alternative source of income, no more, and, at best, probably left gender relations in families unchanged. Women’s farm work was in locations removed from the home, while they continued to be responsible (in cooperation with other household women) for food preparation and child care. The females whose lives were most at risk were daughters who were neglected by their birth families or married at excessively young ages and widows whose in-laws had rejected them.

Children also lost their helping role in household spinning and weaving, which meant fewer opportunities to work in or close to their home (although girls continued to share in women’s domestic and child-care tasks) and the lower wages brought by agricultural work. Children almost certainly had not received separate wages in household textile production, and even if they were paid individually when they sought agricultural day labor elsewhere, the expectation was that they would contribute their pay to the household. The decline of skilled handwork in Indian manufacturing and the increasing importance of agriculture in the economy did little to modify the expectation that boys and girls, men and women, would all contribute to household economies in whatever way they could; the division of labor was little changed, but life became if anything more precarious. The Hindu cultural preference for male children, the dominance of husbands and fathers in families, and the lack of alternatives for women other than time-consuming household tasks and agricultural labor gave females little opportunity to develop their capabilities or skills; among the poor, simple survival was often all that could be expected.22

In sum, the process of English cotton textile industrialization had negative effects on India as a commercial partner and eventual colony. India lost its major export markets for cotton cloth to English industry and its export-oriented products; Indian commercial capitalism was stunted, while England’s machine-spun yarn and cloth were protected in Indian markets. Indian hand-produced coarse cotton cloth was restricted to local trade, although the finest specialties continued to be prized by elites throughout the country and sometimes exported; both spinners and weavers were pressed into agricultural labor to survive. These outcomes were shaped by India’s previous history, especially the ways in which its specialized and subdivided productive system rigidified as it was forced into dependency, first by the East India Company’s commercial capitalist trading system and, later, by British colonial policy. Indian family relations, which included both pressure for cooperation and potential for conflict, changed little. Patriarchal control of children’s lives continued unabated, and adult gender relations continued to be structured in ways disadvantageous to females, given their weak position in their birth family, youthful marriage, dependence on the good will of their husband and his family, and lack of rights as widows.

The English Industrial Revolution, closely connected to India through Lancashire manufacturers’ efforts to compete with Indian products, began in one region, Lancashire and the northeastern part of Cheshire. English cotton textile production (the cloth was woven with linen warps until the late eighteenth century) had localized in Lancashire long before industrialization. By the 1680s, the putting-out (or proto-industrial) organization of production was common in cotton manufactures. Yarn spun by women was bought by merchants or their agents, who distributed it to male weavers and then sold their cloth in distant markets. Both spinners and weavers lived in rural villages and earned a modest but respectable living working in their own cottages, unsupervised and paid by the piece, not by time.23

In the early eighteenth century, the supply of Indian cotton cloth was temporarily restricted as import regulations (passed under political pressure from woolen manufacturers) first prohibited imports, then imposed heavy tariffs and even sought to forbid the very use of imported chintz. Indian cottons nevertheless continued to be imported indirectly or illegally.

In a first response, Lancashire cotton producers increased yarn production, spun on the old-fashioned wheels, producing one thread at the time, by recruiting more spinners.24 But, about the same time, enterprising cotton manufacturers introduced and implemented new organizations of production and technologies, first for weaving (the flying shuttle), and later for carding, roving, and spinning yarn strong enough for warps and fine enough for the weft of the popular cotton textiles. Spinners could run the first spinning machines—jennies—in their own cottages; women’s wages increased markedly with the higher productivity that the jennies made possible. In the past, wrote an observer of Lancashire in 1780, “the chief support of a poor family arose from the loom. A wife could get comparatively but little on her single spindle ... [but the jenny made] a prodigious difference” to women’s earnings; spinners could sometimes outearn male weavers—for a brief period.25 The development of new technologies in response to changing consumer demands contrasts markedly to the Indian case, in which labor costs were kept low by the East India Company’s increasing domination of the weavers’ link to commercial capital and Western merchants’ mediation of growing consumer demand.

English innovation continued, moreover, and the search for more and more efficient machines produced the mule, which was heavy enough that animal or water (and, later still, steam) power was needed to run it. The mules, and the sheds or factories in which they were located, were greeted with protest, and sometimes destruction, by displaced workers in household production and their families and neighbors in the last decades of the eighteenth century. Women continued to be employed on the earlier, lighter spinning machines like the water frame, now in factories, but with the spread of mules—which required not only non-human power but also great physical effort from the humans who ran them—most factory spinners were male by the end of eighteenth century. It was now their turn to be held up for public admiration; E. Baines declared in his History of the Cotton Manufacture in Great Britain (1835) that a mule spinner assisted by his children was able to “live more generously, clothe himself and his family better than many of the lower class of tradesmen.”26

Alongside these technological innovations, there were organizational ones in calico printing, in which women painstakingly patterned printing blocks, or in chintz “printing,” where women hand-painted designs on cloth—both processes modeled after those current in India. But the British cotton-printing industry rarely succeeded in producing prints or dyes that satisfied Indian tastes, and most cotton cloth exported to India in the nineteenth century was unprinted.27

The abundance of cheap and strong cotton yarn led to both increases in the number of hand weavers producing cotton cloth and efforts to mechanize the weaving process. Women displaced from spinning began to move into several occupations that could be done in their cottages. One was removing stems and dirt from raw cotton to ready it for the factory processes, but workers’ wages in this occupation dropped precipitously by the mid-1790s.28 More attractive to adult women was cotton handweaving, which now involved less heavy lengths of cloth (because of the lighter cotton thread used for warps). Weavers who had previously specialized in wool or linen weaving and migrants from Ireland also flocked into cotton weaving, as demand for the cloth continued to expand. These newcomers, together with the former artisanal weavers, flourished briefly in the industry’s expansion in the late 1780s. Writing in 1828, William Radcliffe recalled the period from 1788 to 1803 as a “golden age,” in which “the operative weavers on machine yarns ... might be said to be placed in a higher state of ‘wealth, peace and godliness,’ by the great demand for, and high price of, their labour, than they had ever before experienced.”29 Some of his contemporaries disagreed: the diary of William Rowbottom, an Oldham fustian (mixed linen and cotton) handloom weaver, reported the “most torturing misery” and the “relentless cruelty of the fustian masters” in 1793. Other products of Oldham weavers included calicoes and checks—both modeled on Indian cotton cloth. A French traveler observed, “The fact is, there are too many labourers, and the only remedy is for a less number of young men to take to the loom.”30

Edward Thompson demonstrated the mythic quality of Radcliffe’s rosy picture, pointing out that the older artisan weavers were increasingly equated with the flood of newly recruited unskilled Weavers. As had happened in India, both artisanal and less-skilled English weavers became dependent on the merchants and agents who distributed yarn and subject to rate cuts; a steady erosion of wages began after 1805, as the weavers’ vulnerability became more and more palpable.31

Handloom weaving could only be a temporary haven; in the mid-1820s, powerloom weaving spread, providing new factory jobs primarily for women and children. By the 1860s, most handlooms had been replaced by powerlooms. During the long period of decline, women and children in handloom weavers’ households—unlike the case of India—could find work in either spinning mills or powerloom weaving factories; adult males were less likely to change occupation, partly at least because “male” jobs were scarce in some areas. One member of the 1840 Parliamentary Commission on the State of the Handloom Weavers wrote that “the wife and children of a weaver in most cases contribute very materially to their own support.” Households with a mixed family economy (including both home and factory workers) were better able to earn a livelihood and maintain the dignity of adult male hand weavers.32 Thus, although the local effects of innovation in England on hand spinners and weavers resembled those in India, manufacturers’ readiness to innovate eventually produced new industrial jobs for English former hand workers.

By the mid-nineteenth century, then, cotton textile production—from bales of ginned cotton to sized cloth—was a factory process. Both women and men were factory weavers, and men were mule spinners; each group was assisted by children. (Women also did auxiliary tasks or spun on lighter machines, and some men lingered at handweaving.)

What were the consequences of these changes for family and gender relations? The young people from handloom weavers’ households who worked in factories were making vital financial contributions to their families, more important sometimes than those of their fathers. To what extent did this change their relationship to their parents? A demographic study of Preston, Lancashire, in 1850 asks questions about how households allocated resources and decided who would work, who would care for the children, and what obligations children had toward their parents. It shows that connections between parents and children, and within a wider web of other kin, were extensive; family and household interdependence did not erode but continued strong. Parents expected that children would contribute economically to the household and be ready to help out in life-cycle crises, such as parents’ old age; married couples lived separately from, but often close to, their parents, who could help them with child-care needs if a young mothers’ wages were needed for family subsistence.33

Changes in adult gender relations in the Lancashire cotton industry were structured and limited by the system of job segregation, which undergirded wage differentials based on sex. Job segregation refers not to spatial segregation as in housing but to the fact that men and women were assigned different jobs, with the wage structure of women’s jobs systematically lower than those of similar men’s jobs and few opportunities for women’s advancement. In the production of cotton, the one task that stands out because of the change in the sex of workers is spinning. There had been a sex division of labor in household production and small-scale manufacturing; industrial capitalists sometimes copied, other times revised, the sex identification of occupations, but they almost always made it more rigid; they were willing either to hire women as cheaper labor or buy men workers’ loyalty with privilege and higher wages. Either way, women workers were disadvantaged.34

In sum, in Lancashire, household production became untenable, and men eventually followed women and children first into the spinning mills and, later, into weaving factories. The process was neither linear nor easy for working-class families, given the cycles of boom and bust that accompanied the uneven process of mechanization and the entrepreneurial risks of a period of rapid innovation. But, compared to contemporary changes in India, Lancashire’s position in the forefront of industrialization offered its displaced workers, both male and female, better opportunities for wage earning.

By about mid-century, despite the huge structural changes in Lancashire industry, its workers’ family relations had changed relatively little. As in India, families were one of the few resources available to the poor, even though relationships within families could be exploitative as well as cooperative, and any collective gains were likely to be unevenly distributed among family members. Men received better-paying jobs than women. As adult male wage earners, they had first claim on family resources; children were pressed to submit to parental expectations; wives and mothers were expected to accept full responsibility for the domestic economy yet also to earn wages when needed. Those individuals who rejected familial cooperation in Lancashire were, by all accounts, much better off than children or women on their own in India. Their lives would be tough, but the rising economy and the less rigid prescriptions of English family relations eventually produced tolerable, though gender-unequal, conditions.

Like England, France was an importer of Indian cottons in the seventeenth century, as suggested by the words calicot and indienne (a lightweight printed cotton cloth), which linger in the French vocabulary to this day. At the end of the seventeenth century, a Norman merchant from Rouen developed a light cotton cloth woven with linen warps (called siamoise, perhaps an effort to suggest another exotic origin of the cloth that it imitated, which was surely Indian); the siamoises were explicitly developed to compete with the Indian imports that were favored in France as they were in England. Rouen’s textile manufactures boomed in the eighteenth century; so, too, did parallel non-guild-regulated production in the pays de Caux, an agricultural region to the east of the city, where cheap labor was to be found. The siamoises were woven in red or blue-striped or checked white cloth, or solid red, blue, or white, sold to peasants, urban workers, or exported to the colonies to clothe slaves. (Both the patterns and markets were the same as those of Indian cloth.) In the rural sector, adult female and child carders, and spinners worked more intensively in textile production than did male weavers, who alternated with agricultural labor.35

The first Norman cotton spinning mills (using animal-powered or water-powered machines) were built in the 1780s borrowing English technology, but the region’s economy was shaken by the Eden Treaty of 1786, which opened French markets to English machine-spun yarns.36 Machine spinning, first on jennies run by women in their cottages, then in mills (located in areas distant from the Cauchois, where hand spinning had been concentrated) expanded production duringthe first decade of the century, suffered from British competition after 1810, and resumed growth in the 1820s and 1830s.

Normandy, like Lancashire, combined mechanical spinning and handweaving put out to rural weavers, but the system lingered much longer in France, into the 1870s. As in other areas where the mechanization of weaving came slowly, wages were low and workers vulnerable to the frequent commercial crises. In the pays de Caux, women who could no longer sell their hand-spun yarn moved into handloom weaving, where workers were in great demand to transform the new machine-spun yarns into cloth. Isidore Mars, a Cauchois weaver, described the household division of labor in that period:

Adolescents, the father and the mother wove: and one heard only the regular sound of the shuttle which passed and repassed through the threads, and the beat of the loom ... and it was often the children of the family who were obliged to take care of the demands of commercial affairs and the agricultural work. They ... took the cloth [to the merchants], and worked the land, planting and harvesting the crops.37

Although both adults in this family wove, they were not weaving the same type of cloth.

A division of labor by sex assigned heavy fabrics to men and lighter weight cottons (which were later printed) to women. A weaver explained this difference in terms of women’s lack of “physical strength and their inferior intelligence” and men’s “natural” inclination to textiles that were “more laborious and difficult because they procure higher benefits.” It turned out, however, that the products of the women’s branch were the more popular, and they came to dominate the market. Wages for both women and men weavers fell, the women’s because they had no alternative occupation and the men’s because they were competing with lower-paid women. (Even though men’s wages were low, they continued to be higher than women’s.) And, as in England, handweaving with machine-spun yarn was simply a transitional phase that lasted only as long as poorly paid women weavers kept the cloth competitive with imports. Powerloom weaving factories built close to Rouen starting as early as the 1820s forced the country weavers of the Caux out of business by the 1860s; the region became more fully agricultural, and out-migration to the cities expanded.38 Thus many hand spinners and weavers in the pays de Caux lost their means of subsistence, which had developed in the course of competition with Indian cotton imports, and were constrained, like Indian workers, to turn to agriculture or to migrate cityward with the cotton industry.

The Rouennais was one of three French regions that developed cotton yarn and textiles as major products; the others were Alsace (the region around Mulhouse) and the Lille region. There, Roubaix was the headquarters of cotton, linen, and woolen putters-out competing with its urbane neighbor Lille’s manufactures. Pierre Deyon argues that the “bitter competition between the two cities was for the Roubaisiens an ongoing school of initiative and creative imagination.”39 The definitive end to guild controls in 1791 gave Roubaix entrepreneurs free rein; although the revolution and Napoleonic wars (in which the English were the implacable enemy) slowed down developments, the first spinning mills (using English machines) were installed in the brief interlude of peace during the Napoleonic period, the first steam-driven mill in the 1820s. Weaving changed more slowly; the flying shuttle was not fully adopted until 1820, and put-out handweaving in rural cottages using the machine-spun yarn from Roubaix’s factories lingered until the 1870s. Meanwhile, the city’s archivist gushed in 1864, Roubaix was the “French Manchester.”40

Roubaisien patterns of merchant capitalist-sponsored proto-industrial household production resembled those in India, Lancashire, and Normandy, in their employment of all family members, including children, in some aspect of cloth production. In the 1820 census of Roubaix, for example, all boys and girls twelve or over in weavers’ households were designated as weavers or helpers. The situation in Roubaix and its hinterland resembled England more than did that in Normandy or India, however, for, with time, northern households were more and more completely cut off from agriculture. In the Norman pays de Caux, in contrast, men especially continued to be involved in agricultural labor while women were the specialists in spinning and, later, weaving.

What were the implications of these developments for parent-child and gender relations? In the proto-industrial period in the pays de Caux, women married late and had relatively few children. They were important contributors to the household budget both as daughters and as wives. The late marriage pattern of the Cauchois suggests that children were discouraged by parents from youthful marriage because of their importance as family wage earners. There is little evidence of equality between adult men and women despite women’s importance as wage earners. Their spinning wages were lower than those of weavers in the early period, and when they took up handloom weaving, they worked on smaller looms with lighter yarn for lower piece rates as well. Gay Gullickson points out that women spinners’ sociability took place during the evenings when they gathered to spin in each other’s cottages, while men gathered in cafés to talk, drink, and play dominoes. And even this female work-oriented sociability disappeared when they took to the bulky, immovable loom or spinning mill instead of the portable spinning wheel or distaff and spindle.41

In industrial Roubaix in the 1860s and 1870s, there were heavy family pressures on children to earn wages. In the 1872 census, very high proportions of children age ten to fourteen (38.9 percent of girls and 36.5 percent of boys) were listed as having occupations (mostly spinners’ helpers).42 The vast majority of adult children of both sexes lived with their parents until they married, even when they had jobs. In Roubaix, as in Preston, there were successful family efforts to keep children (who were obviously important as wage earners) in the household, working for the family wage fund. In the nineteenth century, married women were less likely to work than single women, men, and children—not surprisingly, since they had many children. Wives who worked in textiles were burdened with children, while those with older children in the household earning wages dropped out of the labor force, for wives had full responsibility for household and reproductive labor—cooking, laundry, and raising children.

The occupational segregation by sex seen in industrial Lancashire had its equivalent in Roubaix. Women were most often auxiliary workers, helpers, or spinners on the lighter outdated machines or, less often, weavers; they were excluded from mulespinning. Wages for women’s occupations ran systematically lower than men’s; their workplace disadvantage rigidified with industrialization.43

In sum, French industrialists emulated those of England; industrialization came later, delayed by less favorable conditions for technological innovation and capital accumulation but also disrupted by the revolutionary and Napoleonic wars. The Norman and Roubaisien cotton areas had distinctive patterns, based on their earlier histories. Roubaix and the pays de Caux resembled each other as sites of rural putting-out in competition with guild-regulated cities but proceeded in quite different directions, depending on the connection between agriculture and manufactures. The Roubaisien experiences resembled those of Lancashire, in that opportunities for factory work were located in the same areas in which earlier industry or manufactures were produced, so the transitions from hand spinning and handloom weaving to factories did not require relocation of populations. This was less true in Normandy, because proto-industrial spinners and weavers either lived in households involved in agriculture or were themselves part-time agriculturalists; the consequence for the pays de Caux was, as in India, a break from the eighteenth-century pattern of textile manufactures linked to the world commercial capitalist economy.

In both French regions, patterns of household interdependence between parents and children and a sex division of labor for adults in the workplace carried over from manufactures to factory industry, leaving children little autonomy and women disadvantaged at work and in the family because of their dual responsibilities: earning wages and caring for husband and children. As in England, displacement occurred, but the process of transition did not have the severe consequences for workers that the loss of textile export markets in Europe and British colonial policy produced in India.

A web of connections links the structures, geographic locations, processes of change, and actors of this history. The structures of the world economy and interregional trade of the seventeenth and eighteenth centuries were modified by the development of industrial capitalism in England with effects on economies as distant as India and as close as France; capitalist industrialization critically affected the ways in which spinners and weavers earned their livelihood in all three settings. In England and France, one the origin of the capitalist industrializing thrust, the other a wealthy and independent state in which indigenous capitalists borrowed English technology early on and built a competitive textile industry, the transformation of the cotton textile industry brought hard times over the medium run for many workers. The break-up of household production brought more and more men, women, and children into workplaces governed by worker-employer relations, not those of kinship. Male workers sometimes found themselves in competition with women or children; as the arbiters of workplace relations, employers played a major role in establishing the division of labor by sex, setting up a hierarchy of inequality to discipline their work force.

English and French industrialization, proceeding in the same regions as was the elimination of hand spinning and weaving, made alternative forms of work available to displaced workers. The loss of ways of work and living and the fact that new jobs were often exploitative and alienating led to collective protest and resistance to change. Here, the chain of collective actions so lovingly documented by Edward Thompson joins the 1867 turnout of male spinners in Roubaix against their employers’ demand that they each mind two looms. In India during the eighteenth and nineteenth centuries, weavers’ proletarianization also induced less well-documented efforts to resist the process.44 These reactions were connected, not through mutual knowledge but through common processes entailing group loss of control over production in the development of commercial, then industrial, capitalist structures with worldwide ramifications.

Early struggles having been lost, workers in all three countries set to building new connections among themselves, at the local level. This was facilitated by the mid-nineteenth-century prosperity of England and France. In India, the more convention-bound weavers and indigenous commercial capitalism were first drawn by the East India Company into an increasingly dependent trading system, then subjected to British colonial policies that effectively prevented any self-determined economic or social outcome in the first half of the nineteenth century. Indian men and women craftworkers experienced enormous moral and material losses.

Just as similar worker responses to proletarianization accompanied the development of commercial capitalism and industrialization, so common responses occurred in families that became critical resources to be drawn on by workers facing displacement and far-reaching change. Parents and children continued to be interdependent. Wives, husbands, and kin needed, and drew on, each other for support, in ways that were sometimes cooperative, sometimes conflictual, and other times exploitative. Any account of how men and women navigated the large-scale structural changes through which they were living must consider the family household power balance and bargaining between husbands and wives, parents and children, as well as labor market conditions. The historically sculpted connections between structure and action operated through the power balance and bargaining situation in labor markets (between capitalists and workers) and those in families (between husbands and wives, parents and children). Together, these shaped social relations for cotton workers in India, England, and France during the Industrial Revolution.

This analysis has followed only one strand of connections in the process that launched the Industrial Revolution. Unexplored connections abound, such as those to northern Ireland, where cotton spinning and handloom weaving were destroyed; to plantation slavery in the Caribbean and the United States; to industrialization in North America when the Napoleonic Wars cut off trade with England; to other textile industrializations, such as that of silk, in which one set of connections would go from China, India, and Italy to the French Lyonnais to Japan and Paterson, New Jersey; to railroad building, which was closely connected to late nineteenth-century factory industrialization in India; to competition for empire in the search for markets; to the development of new printing, metallurgical, and machine technologies, related to the emergence of newly militant workers ready to organize in their own interests (embedded in, but not fully determined by, their structural position); to the growing demand for reform and social and economic rights in east and west, and so on. This web of connections is also multiply related to possibilities for change in family and gender relations; in the period discussed here, such change was limited, but it later became more salient as fertility declined, popular education was instituted, and women involved themselves in movements for reform, for labor’s rights in the workplace, and for their own political and social rights.

Social historians who follow Edward Thompson in their focus on intensive case studies on a local scale can only benefit by expanding the context of their work, taking into account the connections of these histories to large structures and world processes of change, to far-off peoples in the global economy, and to the past, which is constantly shaping the present.

Louise Audino Tilly has been a pioneer in the study of women, work, and family life in 19th-century Europe. She earned her PhD in 1973 from the University of Toronto. Tilly's dedication to social and women's history has opened the study of history to the lives of the common folk, and has paved a pathway for women in the profession. She has taught at many history departments across the U.S. and in Europe, and has been the recipient of several important fellowships and grants. Tilly is author of Politics and Class in Milan, 1881-1901 (1992).


1. I first encountered E. P. Thompson when he presented “Time, Work-Discipline, and Industrial Capitalism,” in his own inimitable fashion at Harvard in the spring of 1966; it was published the following year in Past and Present, 38 (December 1967): 56–97. The Making of the English Working Class, which I read in graduate school a year later, had been published in 1963 (London).

2. Thompson, Making of the English Working Class, 12.

3. Thompson, “Anthropology and the Discipline of Historical Context,” Midland History, 1 (1972): 41–55.

4. My personal vision of social history owes much as well to several other historians who have been my teachers and colleagues. They are Natalie Zemon Davis, Eric Hobsbawm, David S. Landes, Emmanuel Le Roy Ladurie, Joan Wallach Scott, and Charles Tilly. My thanks also to those who commented on earlier versions of this address, including members of the New York Area French History Seminar, Janet Abu-Lughod, Miriam Cohen, Michael Hanagan, Eric Hobsbawm, Ira Katznelson, Ellen Lagemann, Elizabeth Pleck, Abby Scher, Joan Scott, Charles Tilly, and Chris Tilly.

5. As Christopher Lloyd, The Structures of History (Cambridge, Mass., 1993), 193–94, explains, once in place, “structure is relatively autonomous of individual actions and understandings.”

6. Let me note briefly two other connections that are important in the work of social historians but will not be discussed further here. The first connection is between theory and problems or questions. As the study of past and present—conventionally differentiated as the subjects of historians and social scientists—have converged, historians have become more concerned with posing their questions in systematic ways, informed by theory. They need to be self-conscious about their own theoretical underpinnings and state their problems and questions explicitly. Social historians have been eclectic about their choice of social or political theory, ranging from the old stand-bys like Marx and Weber to contemporary American social-scientific notions of social mobility, political development, economic modernization, demographic determinism, and, more recently, Jürgen Habermas’s theories of communicative action and civil society. The second connection is that between past and present historical accounts and the possibilities of a better world. The accumulation of knowledge in local social histories provides an empirical mapping of causal mechanisms in processes of change, explanations, and interpretations of their meaning for individuals and groups. Therein ties the value of social history for understanding the changing world and finding ways to a more egalitarian and just future. I share William H. McNeill’s optimistic faith that “our historical myth making and myth breaking is bound to cumulate across time, propagating mythistories that fit experience better ... If so, ever-evolving mythistories will indeed become truer and more adequate to public life ... so that men and women will know how to act more wisely than is possible for us today.” McNeill, “Mythistory, or Truth, Myth, History, and Historians,” AHR, 91 (February 1986): 9.

7. The following discussion of the early effects on spinners and weavers of the new technologies and increased scale and concentration in larger units that lay at the heart of the Industrial Revolution are part of an ongoing larger project that looks at capitalist and state-promoted industrialization, growth of the state, and family and gender relations in a world historical perspective, tracing the effects of the English Industrial Revolution and later regional and national industrialization within the world economy as a continuing process linking distant markets and changing opportunities for capitalists and workers, the outcome of which in any given location is dependent on local institutional and historical context. This is a totally reconceptualized revision of my chapter, “Industrialization and Gender Inequality,” in Michael Adas, ed., Islamic and European Expansion: The Forging of a Global Order (Philadelphia, 1993), 243–310, which compared industrialization in five national states but only superficially examined the connections between and among them.

8. My continuing investigations into the origins and consequences of the Industrial Revolution have been strongly influenced by the following scholars and works: E. J. Hobsbawm, Industry and Empire (London, 1968); David S. Landes, The Unbound Prometheus: Technological Change and Industrial Development in Western Europe from 1750 to the Present (Cambridge, 1969); and Eric R. Wolf, Europe and the People without History (Berkeley, Calif., 1982).

9. Quoted in Romesh Chunder Dutt, The Economic History of India under Early British Rule, Vol. 1: From the Rise of the British Power in 1757 to the Accession of Queen Victoria in 1837 (1904; New York, 1970), 262. Dutt was an Indian civil servant whose study both reflected and shaped the thinking of the Indian nationalist movement in the period. On nationalist claims, see Bipin Chandra, “Indian Nationalists and the Drain, 1880–1905,” Indian Economic and Social History Review, 2 (1965): 103–44; Susan S. Bean, “Gandhi and Khadi, the Fabric of Indian Independence,” in Annette B. Weiner and Jane Schneider, eds., Cloth and Human Experience (Washington, D.C., 1989), 355–76; and, in the same volume, Bernard S. Cohn, “Cloth, Clothes, and Colonialism: India in the Nineteenth Century,” 303–53, esp. 338–45.

10. John Irwin and P. R. Schwartz, Studies in Indo-European Textile History (Ahmedabad, 1966); K. N. Chaudhuri, “The Structure of Indian Textile Industry in the Seventeenth and Eighteenth Centuries,” Indian Economic and Social History Review, 11 (June–September 1974): 127–82; Chaudhuri, Trade and Civilisation in the Indian Ocean: An Economic History from the Rise of Islam to 1750 (Cambridge, 1985); and Chaudhuri, Asia before Europe: Economy and Civilisation of the Indian Ocean from the Rise of Islam to 1750 (Cambridge, 1990). (The latter study looks at the vast land area around the Indian Ocean in a Braudelian analysis of slow-changing structure and material life. It expands Chaudhuri’s analysis of textile manufactures in India first reported nearly twenty years earlier but does not differ in explanation.) Also very useful for the seventeenth and eighteenth centuries are Ahmeeda Hossain, “The Alienation of Weavers: Impact of the Conflict between the Revenue and Commercial Interests of the East India Company, 1750–1800,” Indian Economic and Social History Review, 16 (July 1979): 323–45; S. Arasaratnam, “Weavers, Merchants and Company: The Handloom Industry in Southeastern India 1750–1790,” Indian Economic and Social History Review, 17 (1980): 257–81; and especially for comparisons with European proto-industrialization in the same period, Frank Perlin, “Proto-industrialization and Pre-Colonial South Asia,” Past and Present, 98 (February 1983): 30–95.

11. Orme quoted in Chaudhuri, Asia before Europe, 298; Chaudhuri quote on 314. Obviously, there are other interpretations of Orme’s text. Here, I accept the literal reading of Chaudhuri, who cites Orme for his admiration for Indian skill and the fine cloth it produced with very simple wheels and looms.

12. Chaudhuri, “Structure of the Indian Textile Industry”; Trade and Civilisation; and Asia before Europe. For a cultural interpretation of the ambiguous social status of weavers in Indian society, see C. A. Bayley, “The Origins of Swadeshi (Home Industry): Cloth and Indian Society, 1700–1930,” in Arjun Appadurai, ed., The Social Life of Things: Commodities in Cultural Perspective (Cambridge, 1986), 285–321, esp. 293–97. Bayley looks back from the nationalist position of the early twentieth century (which called for support of swadeshi and the boycott of imports) to examine the historic role of cloth in Indian society.

13. On world-scale commercial capitalism, see Perlin, “Proto-Industrialization and Pre-Colonial South Asia”; and Chaudhuri, Asia before Europe. John K. Walton, “Proto-industrialization and the First Industrial Revolution: The Case of Lancashire,” in Pat Hudson, ed., Regions and Industries: A Perspective on the Industrial Revolution in Britain (Cambridge, 1989), 41–68, makes the argument that among Lancashire’s other advantages as the site for the first Industrial Revolution was its specialization in cotton manufacture. This gave some artisans the chance to move into commerce and, later, introduce technological innovations in the textile production processes.

14. Chaudhuri, “Structure of Indian Textile Industry”; Trade and Civilisation in the Indian Ocean; and Asia before Europe. Hossain, “Alienation of Weavers,” argues similarly that the weavers came to bear more and more of the costs of the English East India Company’s efforts to establish stricter quality control at low cost (which lengthened the hierarchical structure between weaver and merchant); and Perlin, “Proto-Industrialization and Pre-Colonial South Asia,” emphasizes the failure to invest in fixed capital or technological change of the increasingly rigid system erected on low labor costs. See also Tim Dyson, “Indian Historical Demography: Developments and Prospects,” in Dyson, ed., India’s Historical Demography: Studies in Famine, Disease and Society (Riverdale, Md., 1989), 1–15, 8 and following. Drawing on articles on specific regions in the same volume by Roland Lardinois and Simon Commander, Dyson ventures a generalization that India had a “very low, possibly negative, rate of population growth from 1770 to 1820.”

15. The 1901 census cited in Encyclopedia Britannica, 11th edn. (1910), 14: 392; material on Indian cloth consumption in 1900–1901 from Peter Harnetty, “‘Deindustrialization’ Revisited: The Handloom Weavers of the Central Provinces of India, c. 1800–1947,” Modern Asian Studies, 25 (1991): 508. Harnetty’s study looks in detail at the Central Provinces but examines as well evidence for the overall decline of handloom weavers in India. Note that about two-thirds of the cloth produced in India in 1900–1901 was handwoven. In the twentieth century, British officials and nationalists made efforts (for different reasons) to revive handweaving and the consumption of indigenously produced cloth. Only in this period was the flying shuttle finally accepted by hand weavers. See also Konrad Specker, “Madras Handlooms in the Nineteenth Century,” Indian Economic and Social History Review, 19 (1989): 132–66; and Sumit Guha, “The Handloom Industry of Central India: 1825–1950,” Indian Economic and Social History Review, 26 (1989): 297–318. A good overview of nineteenth-century English writings on the Indian economy to 1858 is available in K. N. Chaudhuri, ed., The Economic Development of India under the East India Company (Cambridge, 1971). Basic documents in the contemporary debate over nineteenth-century Indian economic developments include, in the special issue of the Indian Economic and Social History Review, 5 (March 1968): Morris D. Morris, “Towards a Reinterpretation of Nineteenth-Century Indian Economic History,” 1–15 (first published in the Journal of Economic History, 23 [1963]: 606–18); Toru Matsui, “On the Nineteenth-Century Indian Economic History—A Review of a ‘Reinterpretation,’” 17–33; Bipan Chandra, “Reinterpretation of Nineteenth Century Indian Economic History,” 35–75; Tapan Raychaudhuri, “A Re-Interpretation of Nineteenth Century Indian Economic History?” 77–100. Also see Amiya Kumar Bagchi, “De-Industrialization in India in the Nineteenth Century: Some Theoretical Implications,” Journal of Development Studies, 12 (January 1976): 135–64; and “Deindustrialization in Gangetic Bihar, 1809‑1901,” in Barun De, ed., Essays in Honour of Professor Susobhan Chandra Sarkar (New Delhi, 1976), 499–522; Marika Vicziany, “The Deindustrialization of India in the Nineteenth Century: A Methodological Critique of Amiya Kumar Bagchi,” Indian Economic and Social History Review, 16 (April–June 1979): 106–46; Amiya Kumar Bagchi, “A Reply,” ibid., 147–61; and Colin Simmons, “‘Deindustrialization,’ Industrialization and the Indian Economy, c. 1850–1947,” Modern Asian Studies, 19 (1985): 593–622.
Starting in 1854, entrepreneurs from Bombay imported British machinery and successfully established cotton mills, which at first specialized in spinning, but some by 1862 were weaving as well. An early warning of the future was issued by R. M. Martin in 1862 (when there were four mills in full operation), who wrote that “even the present generation may witness the Lancashire manufacturer beaten by his Hindu competitor.” Martin, The Progress and Present State of British India (London, 1862), 280–82, quoted in Morris D. Morris, The Emergence of an Industrial Labor Force in India: A Study of the Bombay Cotton Mills, 1854–1947 (Berkeley, Calif., 1965), 25. It took much longer than Martin envisioned, but the handwriting was on the wall for the English cotton textile industry, as other countries developed their own mechanized sectors.

16. Chaudhuri, Asia before Europe, 316.

17. Dutt, Economic History of India, 1: 223, 235, 236, 238, 241, 245, 248, 252.

18. On sati, see Joanna Liddle and Rama Joshi, Daughters of Independence: Gender, Caste and Class in India (London, 1986); and Lata Mani, “Contentious Traditions: The Debate on Sati in Colonial India,” in Kumkum Sangari and Sudesh Vaid, eds., Recasting Women: Essays in Indian Colonial History (New Brunswick, N.J., 1990), 88–126. British and Indian reformers later turned their attention to laws permitting widows to remarry and raising the age of consent for marital intercourse, as discussed by Vina Mazumdar, “The Social Reform Movement in India—From Ranade to Nehru” (1976), in B. R. Nanda, Indian Women: From Purdah to Modernity (New Delhi, 1990), 42–66; the economic problems of poor widows were usually seen as remediable by remarriage, not remunerative work.

19. Nirmala Banerjee, “Working Women in Colonial Bengal: Modernization and Marginalization,” in Sangari and Vaid, Recasting Women, 269–301.

20. Bernard S. Cohn, India: The Social Anthropology of a Civilization (Englewood Cliffs, N.J., 1971), 119–23. Cohn makes clear that the joint family is not universal in India, and I have greatly oversimplified the issue of family structure. See also Amartya Sen, Commodities and Capabilities (Amsterdam, 1985), 80 and following; and Lucy Carroll, “Daughter’s Right of Inheritance in India: A Perspective on the Problem of Dowry,” Modern Asian Studies, 25 (1991): 791–809. V. V. Prakasa Rao and V. Nandini Rao, Marriage, the Family, and Women in India (Columbia, Mo., 1982), discusses primarily mid-twentieth-century family patterns. Rao and Rao’s brief survey of women’s status divides the past into two long periods: the early Vedic (2500–1500 B.C.), for which literary sources prescribe rough equality of men and women, and 1500 B.C. to 1800 A.D., characterized they argue by deterioration of women’s status, their subordination in the family, a rise of female seclusion (practiced not only by Muslim conquerors but adopted as well by upper-class Hindus), and diminished access of girls and women to education.

21. George Forster, A Journey from Bengal to England (1789), 1: 57, quoted in Kalikinkar Datta, Survey of India’s Social Life and Economic Condition in the Eighteenth Century, 1707–1813, 2d rev. edn. (New Delhi, 1978).

22. Here, I adopt the concept of Amartya Sen, who writes that “there is a good case for judging individual well-being, neither in terms of commodities consumed nor in terms of the mental metric of utilities, but in terms of the ‘capabilities’ of persons. This is the perspective of ‘freedom’ in the positive sense: who can do what.” The capabilities approach, Sen argues, can take account of personal characteristics such as sex and age and not simply the resources to which individuals may have access. Sen, “Economics and the Family,” Asian Development Review, 1 (1983): 19. See also Jocelyn Kynch and Amartya Sen, “Indian Women: Well-Being and Survival,” Cambridge Journal of Economics, 7 (1983): 363–80. Sen develops his argument there (365) against John Rawls’s position that individuals’ advantage is based on their possession of widely desired goods like rights, liberties, opportunities, and wealth, and Ronald Dworkin’s case for opportunities as the key to advantage. In “What Did you Learn in the World Today?” American Behavioral Scientist, 34 (May–June 1991): 530–48, Sen argues that equity (fairness in the distribution of “good things”) should be given more weight in evaluating policy than efficiency (referring in part to having more “good things”), and he calls for disaggregated internal comparisons as well as cross-national aggregated ones in analyses of inequality.

23. Walton, “Proto-industrialization,” esp. 42–45, 62–63. The following description of eighteenth-century changes in the Lancashire cotton industry is based on Sidney J. Chapman, The Lancashire Cotton Industry: A Study in Economic Development (1904; Clifton, N.J., 1973); Alfred P. Wadsworth and Julia De Lacy Mann, The Cotton Trade and Industrial Lancashire, 1600–1780 (1931; Manchester, 1965); Ivy Pinchbeck, Women Workers and the Industrial Revolution, 1750–1850 (1930; London, 1981), and the very full documentation they provide.

24. Pinchbeck, Women Workers, 131, notes the increase in cotton spinners around Manchester and elsewhere in Lancashire, in the mid-eighteenth century.

25. Thoughts on the Use of Machines, 14, quoted in A. P. Wadsworth, “The Lancashire Wage-Earners before the Factory System,” Wadsworth and Mann, Cotton Trade, bk. 4, 404.

26. Quoted in Pinchbeck, Women Workers, 153.

27. Maxine Berg, “Women’s Work, Mechanization, and the Early Phases of Industrialization in England,” in Patrick Joyce, ed., The Historical Meanings of Work (Cambridge, 1987), 64–98. Compare the late seventeenth-century comment by a Dutch East India Company agent (D. Havart) on the process of cloth painting in Coromandel quoted in Irwin and Schwartz, Studies in Indo-European Textile History, 34–35, with the description in S. D. Chapman and Serge Chassagne, European Textile Printers in the Eighteenth Century: A Study of Peel and Oberkampf (London, 1981), 95–96 (quoted in Berg, 95–96) of late eighteenth-century English cloth painting. The Indian process was adopted in order to reproduce the desired complex design on English cotton cloth. Harnetty, “‘Deindustrialization’ Revisited,” 463, n. 22, cites D. A. Farnie, The English Cotton Industry and the World Market, 1815–1896 (Oxford, 1979), 101, as evidence of the Indian lack of acceptance of English printed cottons. This footnote seems at least partially inconsistent with Harnetty’s text on the same page, which indicates that printed English cloth imitating some Central Provinces’ textile specialties was well received when it was introduced in the area in the late 1860s.

28. J. Aiken, A Description of the Country from Thirty to Forty Miles Round Manchester (1795), cited in Pinchbeck, Women Workers, 152–53.

29. William Radcliffe quoted in Chapman, Lancashire Cotton Industry, 38.

30. William Rowbottom, “The Chronology or Annals of Oldham” (unpublished manuscript diary), quoted by John Foster, Class Struggle and the Industrial Revolution: Early Industrial Capitalism in Three English Towns (London, 1974), 35; T. S. Ashton, “The Standard of Life of the Workers in England, 1790–1830,” in F. A. Hayek, Capitalism and the Historians (London, 1954), 127–59, draws on Rowbottom’s diary for a series of retail prices on basic food items in Oldham from 1791 to 1809, showing large swings in the cost of workers’ typical diets. L. Simond, Journal of a Tour and Residence in Great Britain (Edinburgh, 1815), quoted in Duncan Bythell, The Handloom Weavers: A Study in the English Cotton Industry during the Industrial Revolution (Cambridge, 1969), 107–08. (The observation was made in 1811.)

31. Chapter 9, “The Weavers,” in Thompson, Making of the English Working Class; Pinchbeck, Women Workers, 173; Foster, Class Struggle, 37. Foster emphasizes both historical contingencies, such as the French revolutionary and Napoleonic wars, and structural ones, such as the slowness of mechaniza­tion of weaving in his case study of Oldham, where fustian weaving continued but was increasingly miserably paid.

32. Reports of the Parliamentary Commission on the State of the Handloom Weavers, XXIV, 1840, 556, quoted in Pinchbeck, Women Workers, 166. Bythell, Handloom Weavers; John S. Lyons, “The Lancashire Cotton Industry and the Introduction of the Power Loom, 1815–50” (Ph.D. dissertation, University of California, Berkeley, 1977); and Walton, “Proto-industrialization,” 67, also discuss the mixed family economy in hand weaver families. Compare Tessie Liu, The Weaver’s Knot: Contradictions of Class Struggle and Family Solidarity in Western France (Ithaca, forthcoming), who argues that patriarchal authority and prerogative were little challenged in handloom weavers’ families in France.

33. Michael Anderson, Family Structure in Nineteenth Century Lancashire (Cambridge, 1971); “Household Structure and the Industrial Revolution: Preston in Comparative Perspective,” in Peter Laslett and Richard Wall, eds., Household and Family in Past Time (Cambridge, 1972), 215–35; and “Social History and the Working-Class Family: Smelser Revisited,” Social History, 3 (October 1976): 317–34. See also M. M. Edwards and R. Lloyd-Jones, “N. J. Smelser and the Cotton Factory Family: A Reassessment,” in N. B. Harte and K. G. Ponting, eds., Textile History and Economic History (Manchester, 1973), 304–19. In “Sociological History: The Industrial Revolution and the British Working-Class Family,” Journal of Social History, 1 (Fall 1967): 17–35, and Social Change in the Industrial Revolution (Chicago, 1959), Neil J. Smelser argues in contrast that English workers’ discontent in the 1830s was linked to their concerns to maintain control over wage-earning children, who once had worked in the household and later in spinning mills under their parents’ direct supervision. With heavier and larger mules, fathers found it more difficult to keep their children under their supervision, and the effect of the Factory Act of 1833 (which limited children’s but not adults’ working hours, thus making impossible common work schedules among parents and children) was further to undermine paternal authority. Smelser concludes that from then until the 1850s, when new Factory Acts delineated uniform working schedules, the politics of the common people in Lancaster were dedicated to reconstituting the family. Smelser’s structural-functional approach was one of Thompson’s targets in the preface to Making of the English Working Class.

34. A debate about the gender meaning of the continuing dominance of males as mule spinners may be traced through Heidi Hartmann, “Capitalism, Patriarchy, and job Segregation by Sex,” Signs, 1 (1976), Part 2: 137–69; William Lazonick, “Industrial Relations and Technical Change: The Case of the Self-Acting Mule,” Cambridge Journal of Economics, 3 (1979): 231–62; Mary Freifeld, “Technological Change and the ‘Self-Acting’ Mule: A Study of Skill and the Sexual Division of Labour,” Social History, 11 (October 1986): 319–43; William Lazonick, Competitive Advantage on the Shop Floor (Cambridge, Mass., 1990), chap. 3, “Minders, Piecers, and Self-Acting Mules,” esp. 88–93; Mariana Valverde, “‘Giving the Female a Domestic Turn’: The Legal, Social and Moral Regulation of Women’s Work in British Cotton Mills, 1820–1850,” Journal of Social History, 21 (1988): 619–34; Ellen Jordan, “The Exclusion of Women from Industry in Nineteenth-Century Britain,” Comparative Studies in Society and History, 31 (April 1989): 273–96; Robert Gray, “Factory Legislation and the Gendering of Jobs in the North of England, 1830–1860,” Gender and History, 5 (Spring 1993): 56–80; Colin Creighton, “Richard Oastler, Factory Legislation and the Working-Class Family,” Journal of Historical Sociology, 5 (September 1992): 292–320; and Carol E. Morgan, “Women, Work and Consciousness in the Mid-Nineteenth-Century English Cotton Industry,” Social History, 17 (January 1992): 23–41.

35. This description of cottage industry in the pays de Caux is derived from Gay Gullickson, Spinners and Weavers of Auffay: Rural Industry and the Sexual Division of Labor in a French Village, 1750–1850 (Cambridge, 1986).

36. Landes, Unbound Prometheus, supplies an excellent overview of cotton textile industrialization in the three French regions where it was dominant.

37. Isidore Mars, Derniers souvenirs du bon vieux temps d’Auffay depuis 1793 jusqu’à 1840 environ (Dieppe, 1876), 3, as translated and quoted in Gullickson, Spinners and Weavers, 109. (1 have made some minor modifications in Gullickson’s translation.)

38. Charles Noiret, Mémoires d’un ouvrier rouennais (Rouen, 1836), 33, translated and quoted in Gullickson, Spinners and Weavers, 110.

39. Pierre Deyon, “Un modele à l’épreuve, le developpernent industriel de Roubaix de 1762 à la fin du XIXème siècle,” Revue du Nord, 63 (special issue, Aux origines de la Révolution industrielle) (January–March 1981): 59.

40. Théodore Leuridan, Histoire de la fabrique de Roubaix (Roubaix, 1864), 156.

41. Gullickson, Spinners and Weavers, 84–85, 149–52. See also Gullickson, “Love and Power in the Proto-industrial Family,” in Maxine Berg, ed., Markets and Manufacture in Early Industrial Europe (London, 1990), 205–26; and Liu, Weaver’s Knot, makes an interesting argument about interfamilial inequality between sons and daughters as well as between husbands and wives.

42. The following material is reported in Louise A. Tilly, “Occupational Structure, Women’s Work, and Demographic Change in Two French Industrial Cities, Anzin and Roubaix, 1872–1906,” in Jan Sundin and Erik Söderlund, eds., Time, Space and Man (Atlantic Highlands, N.J., 1979), 107–32; “Individual Lives and Family Strategies in the French Proletariat,” Journal of Family History, 4 (Summer 1979): 137–52; and “The Family Wage Economy of a French Textile City: Roubaix, 1875–1906,”Journal of Family History, 4 (Winter 1979): 381–94.

43. Louise A. Tilly, “Gender and Jobs in Early Twentieth-Century French Industry,” International Labor and Working-Class History, 43 (1993): 31–47.

44. For the eighteenth century, Chaudhuri, “Structure”; Asia before Europe; and, for the later period, Harnetty, “‘Deindustrialization Revisited.’”