The image of rural women hunched over spinning wheels and looms does not figure prominently in the historiography of socialist China. Chao Kang's study of the Chinese cotton industry assumes that manual textile production was phased out in the early 1950s, and other works in English and Chinese concur.Footnote 1 Yet interviews and archival sources show that tubu (homespun cloth) remained the standard garb for many rural people until the very end of the collective period, and that for an entire generation, millions of rural women continued to spend large portions of their waking hours at the spinning wheel or the loom, usually after a full day of collective labor in the fields.
This is perplexing for two reasons: firstly, official discourses in the PRC did not afford legitimacy to the labor of making cloth and clothing; to the extent that such work was perceived at all, it was treated as an insignificant household chore, tainted by its association with a backward peasant economy and a feudal gender system that locked up women in the household. Secondly, CCP leaders sought to curtail household textile work because it tied up two resources that were needed for the construction of the socialist economy: cotton – China's most important cash crop and a major source of government revenue – and women's labor, which was seen as a key ingredient in the economic modernization of the country.
It is easy for us, who no longer observe manual textile work on a daily basis, to forget how enormously labor-intensive spinning and weaving is. In Mao-era north China, it took about sixty labor days a year to provide a family of five with the absolute minimum of clothing. If the family was to be clothed decently according to the modest standards of the times – a good suit for the household head to wear on market days, a new suit of clothes for children every year – a woman had to spend one-third to one-half of her waking hours at the spinning wheel and the loom. Continued manual textile work thus put rural women in conflict with a state that needed women's labor and was, after 1956, in a position to enforce their full participation in collective agricultural work.
In theory at least, rural people had neither the means nor the need to produce their own clothing. The introduction of unified purchase and marketing [tonggou tongxiao] in 1954 put an end to private trade in cotton, cotton yarn, and cotton cloth, and the collectivization of agriculture in 1955–1956 ensured that almost the entire cotton harvest ended up in the hands of the state.Footnote 2 At the same time, a centralized rationing system gave rural and urban people access to modest quantities of state-produced cloth at relatively low prices. Why, then, did rural women continue to engage in time-consuming and back-breaking textile work, in the teeth of state attempts to suppress it? And why did a government that had in earlier years not hesitated to attack traditional gender norms and alter entrenched divisions of labor fail to liberate women from the spinning wheel and the loom?
A first step towards answering these questions lies in unraveling the complex system of interlocking, bureaucratically administered shortages – of grain, cash, cotton, cloth, and labor time – that governed households’ strategies in the collective period. As I will show below, the issue was not simply that state-supplied factory cloth fell short of needs; it was that shortages of cash and grain forced rural households to sell their ration coupons and weave cloth from whatever scraps of cotton they could obtain. This made sense to them because at a time of pervasive shortages, women's labor was the one resource that remained elastic: men expected, and women accepted, that women could always “work an extra shift” [jia ban], carrying out textile work late at night, after men and children had gone to bed. Looked at in this way, handloom weaving survived despite the enormous social cost of having millions of women spend countless hours on a task that could have been mechanized, simply because these costs could be offloaded onto the most powerless group in Chinese society: rural women.
A second factor in the survival of manual textile work, and the one I am going to focus on in this article, was that such work was embedded in complex patterns of social reproduction. Without wanting to idealize the harsh labor regime of pre-1949 rural China, I would argue that textile work in rural China was as much concerned with the making of properly gendered persons as with the making of textiles.Footnote 3 The rural Chinese woman at her loom reproduced herself and others: her labor shaped her body, taught her embodied skills, and inculcated gendered norms and habits. The products of her labor created men and women as specific sorts of persons, as textiles generally do.Footnote 4
Cloth and clothes created social relationships: local customs in many parts of China dictated that no daughter could marry out in style without a trousseau of clothes; no family could find a bride for their son without giving cloth or cotton to her parents; no newborn child could be welcomed into the community without an exchange of textiles; and no person could leave this world in proper fashion without at least one set of grave clothes. The reproduction of a social and moral order was not external to textile work; it was part of what such work was meant to do. Modern Chinese states did not recognize ritual and social needs; they did not even have a language to discuss them, apart from the highly charged, condemnatory terms of “waste” and “feudal superstition”. Village leaders could and did distribute small amounts of cloth or cotton to expectant mothers and betrothed couples, so that birth and weddings could go ahead as planned, but by and large the state failed to come to terms with the fact that cloth (as, in a different way, food) was a means of social, rather than just physiological, reproduction.
Yet for the state, too, larger issues were at stake. Chinese reformers since the late nineteenth century had argued that the nation's weakness stemmed from its economic fragmentation. The main thrust of economic nationalism in the first half of the twentieth century was anti-imperialist, yet economic nationalists were also concerned about the existence of an entrenched “natural” economy, in which self-sufficient peasant households met their needs locally, without participating in a national market. For China to survive, the Chinese had to learn to consume and produce socially, in ways that contributed to the ever-increasing integration of the nation.Footnote 5 The creation of such an integrated national economy was part of the program of the Chinese Communist Party. From the state's point of view, farmers had a patriotic duty to sell cotton to and buy cloth from state trade agencies; in doing so, they were told, they helped consolidate the worker–peasant alliance that was the foundation of the new state.Footnote 6 State planners, in short, agreed with rural people that the circulation of cloth and cotton had the potential to create communities, but they wanted these communities to be national, not local, in scope.
The remainder of this article proceeds in three steps. The first part will discuss discourses on textiles and women's work during the Qing dynasty (1644–1911). These discourses are relevant to my argument because Qing period thinkers were aware of the social and economic centrality of women's work in a way that few modern writers are, and because the rejection of the late imperial order by the reformers of the late nineteenth and early twentieth century set the terms in which women's work was perceived in later years. Next, I will look at the cotton–textile–female labor nexus under the People's Republic and try to explain why CCP policies did not produce the intended results. Finally, I will zoom in on two counties in north China where I conducted field research, and discuss how women experienced changes in their working lives from the 1940s to the 1980s. In this final section, I will also address the question of how representative my fieldwork observations are for China as a whole.
Women's work in Qing China
A concern with a proper gendered division of labor goes back to the formative periods of Chinese civilization. The Chinese distinction between a female “inner” [nei] and a male-dominated “outer” [wai] sphere bears some resemblance to the Western opposition between private and public spheres, but the idea that the family and the household stood somehow outside politics would have appeared absurd to Confucian thinkers. As Susan Mann has pointed out, the principle of gender separation was:
[…] invoked to stress that wives and mothers inside the home embody the moral autonomy and authority on which husbands and sons must rely to succeed outside. All are part of a family system that constitutes a seamless, unitary social order centered on the home and bounded by the outer reaches of the imperium.Footnote 7
It was the duty of the ruler and of Confucian-educated scholars to ensure that men and women in rural and urban households fulfilled their separate yet complementary tasks. For commoners, these tasks involved the production of food and clothing, the twin necessities that “people cannot go without for even a single day”.Footnote 8 Farming and weaving were often paired rhetorically: men farmed, making sure that everybody had enough to eat; women span and wove, making sure that nobody remained unclothed.
“Women's work” [nü gong] had a prominent but ambiguous place in these discourses. It was understood primarily as textile work; in fact, the character gong was often written with the radical for “silk” to denote the specific nature of female work. More than in the case of men, women's work was believed to be a moral as well as an economic necessity. Work was considered one of the “four womanly virtues,” and work with spindle, loom, and needle exemplified the wisdom, frugality, and industriousness that women were expected to possess.Footnote 9 Yet women's work had also more problematic connotations: written without the silk radical, the character gong assumed the meaning of “craft”; written with yet another homophone, it signified “merit” or “value”. These variations, Francesca Bray suggests, reveal the ambiguity that Confucian moralists felt about women's work: such work was meritorious as long as it produced use value for state and household; if, on the other hand, it produced private profit, it was potentially disruptive.
From one perspective, textile work was the female equivalent of farming, part of the “fundamental” [ben] occupations that the state wanted to strengthen. From another perspective, textile work could be seen as craft and thus as part of the “secondary” [mo] pursuits of artisans and merchants.Footnote 10 Different types of work occupied different spaces in this spectrum. Sericulture enjoyed special status because the classics paired it with agriculture [nong sang] as the basic occupations of the realm. Cotton came to China later and did not have the same canonical status; it was therefore more easily seen as a distraction from “proper” farming.
Late imperial administrators and ideologists envisaged a gendered economy in which “men farmed and women wove” [nan geng nü zhi] and in which almost all work took place in the confines of the small producing household. It is safe to say that almost all women in late imperial China, from farmers to palace ladies, performed textile work – “genteel” embroidery in the case of elite women; spinning, weaving, and sewing for most others – and that textile skills formed an important part of women's identities, a measure of their worth, and sometimes also a source of pride and a means of self-expression.Footnote 11 The identification of women with textile work was so strong that writers in the Qing period referred to women who did not spin or weave as “having no occupation”, even if they worked in the fields or in handicrafts.Footnote 12 Susan Mann provides an example of this rhetoric: Qing official Zhou Kai (1779–1837) commenting on the customs of poor rural people in Xiangyang, Hubei:
I especially pity the women of Xiang. They have no way to develop specialized work of their own and thereby affirm their commitment as faithful wives. […] Once when I was walking out in the country, I saw some poor women. They were at work breaking up clods of earth with a hoe, and they were even ploughing in pairs (za gengyu).This shocked me deeply, for when the Odes refer to women in the fields, they speak only of women who bring food to their husbands […]. I was certain it would not be long before these same wives would be suing their husbands, and husbands their wives. […] All human beings have “feelings of shame and dislike” [xiuwu zhi xin]; can it be that these women's hearts are inhuman [shuren]? Truly the cause of their behavior is that they have no specialized work of their own to do, and they have no commitment to wifely fidelity. As a result, their labor is not sufficient to provide for themselves. All of this is due to the fact that they grow no mulberry.Footnote 13
Actual divisions of labor in the household were more diverse than this rhetoric suggests. Xu Dixin and Wu Chengming estimate that on the eve of the Opium War (1839–1842), almost one-half of rural households wove cotton cloth and that slightly more than one-half of this cloth came on the market.Footnote 14 Ramie and hemp, common textile fibers until the early twentieth century, gradually lost ground to cotton; silk, while important economically and culturally, never accounted for more than a few per cent of total textile output.Footnote 15 Little is known about regional variations in labor practices. The only region that has been studied in some detail is the Lower Yangzi (Jiangnan) region, where weaving for the market was widespread and divisions of labor in the household were relatively pronounced. Even here, the ideal of “men farming and women weaving” was strictly observed only by wealthy families, while poor farmers often adhered to a pattern of “husband and wife working together” [fu fu bing zuo]. Kenneth Pomeranz has recently suggested that it was only relatively late that cloistering of women became the norm in most parts of the Chinese heartland.Footnote 16
Advanced weaving techniques, and with them patterns of strict gender segregation, spread from Jiangnan to other provinces in the seventeenth to nineteenth century. Initially, textile industries in North China were hampered by the fact that cotton is difficult to spin if the air is dry, but in the 1640s, people in southern Hebei learned to spin cotton in underground cellars, where warmth and humidity could be preserved during the winter spinning season.Footnote 17 By the eighteenth century, most northern provinces had become self-sufficient in textiles; in the nineteenth century, this process of “import substitution” spread further to the middle and upper Yangzi regions.Footnote 18 The outward spread of improved cultivation and manufacturing techniques intersected in complex ways with changes brought about by growing imports of foreign yarn and cloth since the late nineteenth century.
Imagining a new National Economy
If Ming and Qing officials saw the economy as organized primarily around gender divisions in the ideally self-sufficient household, late nineteenth-century reformers envisaged an economy organized around exchanges between producers and consumers. In this new vision, women's home-based work disappeared from sight. In 1898, the leading reformer Liang Qichao complained that “out of two hundred million women, every one is a consumer, not one is a producer. Because they cannot support themselves but depend on others for their support, men keep them like dogs and horses.”Footnote 19 This is an extraordinary statement, understandable only if we assume that Liang imposed on China a normative, Western-derived model in which productive work takes places in offices and factories, while households serve as places of consumption.
In reality, only a tiny fraction of even the male population worked outside households at that time. Officials, clerks, and professionals earned individual salaries; in a different way, the urban proletariat in the treaty ports (minuscule at the time of Liang's writing) could be said to consist of self-supporting, independent economic agents. However, the vast majority of Chinese worked in household units – farms, shops, family enterprises – that compensated members with a share in a common income stream rather than a wage. Liang's identification of men with productive work and women with consumption would have made no sense to thinkers of an earlier generation, who saw men's and women's work as equally rooted in the household.Footnote 20
Early twentieth-century debates on women's place in public life centered primarily on women's rights to education and on the ways in which middle-class women could contribute to the modernization of China. Reformers envisaged roles for educated women as teachers in girls’ schools, doctors in women's hospitals, and even soldiers in women's battalions. Women's work in clerical positions, as typists and secretaries, was considered problematic because it exposed respectable women to unrelated men who were thought likely to abuse them.Footnote 21 Many commentators recommended domestic textile work as being more compatible with conservative class and gender standards. Wives and daughters of the new urban middle class were advised to engage in sericulture, needlework, and other textile crafts – occupations that allowed them to stay at home and raise their children as future citizens of the nation.Footnote 22 The men and women who extolled the virtues of revitalized nü gong in the Shanghai press did not, however, extend their analysis to the working lives of the majority of Chinese women. Household-based textile work by non-elite women was rarely discussed, despite the fact that rural women in most parts of China continued to spin and weave and that new export-oriented industries – silk embroidery, lace-making, knitting etc. – sprang up in many parts of coastal China. What had once been the normal and normatively right type of work for all women was now increasingly considered irrelevant to the social and economic life of the nation.
This view was reinforced by the widespread perception that manual textile industries were collapsing under the onslaught of foreign imports. Recent research suggests, however, that Chinese textile handicrafts held out remarkably well in the face of imports and domestic factory competition. The three-way battle between import cloth, the products of Chinese cotton mills, and tubu cloth has been charted in great detail.Footnote 23 The story begins in the 1870s, when imports of machine-made yarn and cloth – initially from Manchester, later from Bombay and Osaka – took off. Hand-spinning, which was hopelessly inefficient compared to mechanized spinning, soon declined: already by the turn of the twentieth century, half of all yarn used in China came from factories. The advance of machine yarn slowed after 1900, largely because weavers took to combining machine-spun warp threads (which have greater tensile strength) with homespun weft (which is thicker and warmer than machine yarn). By 1913, the percentage of factory yarn had crept up to a level of about 75 per cent, where it stayed throughout the 1920s and 1930s.Footnote 24 Hand-spinning survived in cotton-growing areas, where families used the cheap labor of girls and old women to spin for household use. Even some commercial weaving centers continued to use hand-spun yarn: the Dingxian handloom industry, for example, continued to use homespun yarn in combination with machine-spun yarn, despite Dingxian's proximity to Tianjin's cotton mills.Footnote 25
The impact of factory competition on handloom weaving was quite different. In relative terms, handloom cloth declined as factory cloth conquered a larger share of the market. At the same time, a growing population and rising living standards may have led to increased overall demand, and the availability of cheap factory yarn led to the dispersion of handloom weaving from its old centers in the lower Yangzi and the North China plain to central and western China.Footnote 26 New commercial weaving districts, specialized in imitation “foreign” cloth, emerged in places with good access to machine-spun yarn, such as Nantong near Shanghai and Gaoyang and Baodi near Tianjin.Footnote 27 In 1934–1936, handloom cloth still accounted for 70 per cent of China's cotton textile output.Footnote 28 In these years, 24 per cent of Chinese rural households span and wove. Unsurprisingly, handloom weaving was most common in provinces where cotton was grown: in Henan, about 60 per cent of rural households produced cloth; in Shaanxi, 50 per cent; in Hebei, Shanxi, and Shandong, about 40 per cent. Handloom weaving was comparatively rare in southern, south-western, and north-eastern China, where little cotton was produced.Footnote 29
Gender, Textiles, and Cotton Under Socialism
When the Chinese Communist Party came to power, it inherited a number of assumptions from its republican and late imperial predecessors. CCP leaders saw little moral or economic value in household-based textile work; instead, they associated women's work at the spindle and the loom with backwardness and underdevelopment. Like their predecessors, they assumed that rural handicrafts had been practically wiped out by foreign competition, and would have to be phased out even if they survived.Footnote 30Tubu cloth in China did not have the association with cultural authenticity that khadi had in India.Footnote 31 The fact that China's handloom industries had held out relatively well in the face of mechanized competition was a cause not for celebration but rather for concern: as long as millions of rural households dressed in handmade cloth, China could not develop a strong cotton industry – which was seen as a necessary first step in a trajectory that started with cotton mills and proceeded to railroads and heavy industries.Footnote 32
Interestingly, CCP leaders had embraced manual textile industries before their victory in the Civil War. From 1941 to 1949, the CCP – then entrenched in the Shaanxi-Gansu-Ningxia border region centered on Yan'an and embroiled in an undeclared civil war with their Nationalist rivals – mobilized tens of thousands of rural women (and many men, including soldiers and party leaders) in a mass spinning movement. Faced with a Nationalist embargo, CCP leaders promoted hand-spinning and hand-weaving as a way to improve rural living standards, raise political awareness among women, and strengthen the economic independence of the border region. In an area that did not have a strong native textile tradition, 150,000 to 200,000 women produced yarn and cloth in private households or loosely knit collectives.Footnote 33 After the victory in the Civil War, however, the socialist government discontinued this policy. As a 1949 article explained, the priority now was to develop a modern cotton industry; hand-spinning and hand-weaving were to be phased out in the next two to three years.Footnote 34
Ideologically, the CCP espoused the Engelsian notion that women's liberation hinged upon their full participation in paid, public work. Chinese law defined work as the right and duty of all citizens capable of work, and rights to political and social inclusion in the new workers’ state were tied to participation in productive work. Women who did not participate in such work were considered “household dependents” [jiashu] and accused, in both popular and official discourse, of being unproductive “parasites” [jishengchong].Footnote 35 While state policies in the 1950s accepted that “under the present conditions, housework and some domestic handicrafts are socially indispensable types of work”, they called for the creation of conditions in which all rural women could fully participate in “primary agricultural work”.Footnote 36 Revolutionary rhetoric depicted women's liberation as a process of stepping out of the darkness and loneliness of the feudal family, into the sunshine of public life and public work.Footnote 37 “Stepping out of the home” [zouchu jiamen] was described as a “democratic right” demanded by the majority of women,Footnote 38 but internal Women's Federation documents show that many rural women felt apprehensive about participation in male-dominated farm work and needed to be “educated” about the nobility of collective labor, so that they could overcome their “narrow and selfish” orientations inherited from the old society.Footnote 39
Like their Nationalist predecessors, the socialist government operated under very tight resource constraints. Chinese cloth consumption was 6.8 square meters per capita, one of the lowest in the world.Footnote 40 Early PRC planners aimed to increase production to about 10 running meters of cloth per capita.Footnote 41 However, even this modest target was not reached until the early 1980s: cloth consumption remained below 7 meters in most years until 1978.Footnote 42 The dilemma for the nation was the same as for individual households: land used for cotton could not be used for grain; more warmth and comfort on the skin meant less food in the stomach. An aggravating factor was that China – a longtime net importer of cotton and cotton goods – became a net exporter in the 1950s.Footnote 43 Cloth exports, mostly to the Soviet Union and south-east Asia, were about 10 per cent of total output in the years between 1956 and 1973; in 1962, when per capita rations had dropped to just 1 meter, China exported 24.5 per cent of its cotton textile output. It was cotton cloth, more than any other commodity, that paid for Soviet technological aid and for the grain imports that saved the urban population from starvation during the 1959–1961 famine.Footnote 44
Cotton was also central to the state's aim of rapid capital formation. Nicholas Lardy summed up the mechanism:
The sale of homespun yarn and handwoven cotton cloth in rural markets has been and remains legally prohibited […]. Because its major input was underpriced […] and its products at least initially probably overpriced, the cotton textile industry has been one of the most profitable branches of industry and the single largest source of state revenues, providing in the late 1970s about 10 percent of state budgetary revenues from all sources.Footnote 45
In fact, cotton procurement prices were set so low and textile retail prices so high that cotton factories built in the early 1950s earned their investment back within the first year of operation.Footnote 46 The cotton sector, in short, fulfilled the tasks assigned to it by state planners: it accumulated capital, subsidized heavy industry, and earned hard currency for needed imports. What it did not do was to improve the material life of the people. Statistically, the average Chinese person had slightly less cloth (and consumed slightly fewer calories) in 1960–1983 than in 1924–1931 – although a more equitable distribution may have ensured that the median person had slightly more clothes to wear in the 1960s than in the 1920s.Footnote 47
The introduction of “unified purchase and marketing” [tonggou tongxiao] of cotton and cotton products in 1954 formalized a de facto state monopoly that had been gradually introduced since 1949. All cotton had to be sold to the state; cultivators were only allowed to retain one kilogram of cotton per capita to use for quilt and garment padding. This “self-retained cotton” [ziliumian] ration became the main source for manual spinning and weaving under the collectives. Regulations were also issued for the supply of the population with cotton products. All cotton and mixed cotton cloth was to be distributed through state channels. Garments, blankets, and other cotton goods also came under state monopoly; such items as mosquito nets, hats, knitwear, etc. were included in later years. To ensure that people bought only as much cloth as they were entitled to, the state distributed ration coupons which could be redeemed only in the place of issuance. In theory, these coupons could be gifted to friends or relatives but not be sold; in practice, however, they were widely sold on the black market.Footnote 48
Unified purchase and marketing ensured that most cotton ended up in the state's hands, but it did not give the state full control over the cotton harvest. In 1954, the national commodity rate [shangpinlü] for cotton – the share of the harvest purchased by the state – reached 73 per cent; for most of the collective period, it fluctuated between 85 and 95 per cent.Footnote 49 Yet control was uneven: in 1962, for example, about one-third of the cotton crop of Henan, Tianjin, and Shandong (all of them major production areas) remained in the hands of farmers.Footnote 50 Moreover, a 100 per cent commodity rate only indicates that the state cotton company procured all the cotton that it considered extractable at a reasonable cost. Cotton bolls ripen over a period of several weeks, and at the end of the harvesting season, a cotton field may still contain unopened bolls. By removing the stalks from the field, drying them, and picking them over carefully, farmers could obtain fibers that were of no value to the state. Moreover, we know that each level of the administrative hierarchy concealed part of land under their control, leading to cumulative under-reporting. Vaclav Smil found that the official total of China's farmland was almost 50 per cent lower than the real figure, suggesting that large parts of the harvest were concealed from the state.Footnote 51 Cotton, which does not grow well on marginal land, may have been more difficult to conceal than other crops, but it seems safe to assume that the farmers’ share of the harvest was larger than the statistics suggest.
In 1965, a conference convened by the national Supply and Marketing Cooperative (SMC) and the Ministry of Trade identified three sources of tubu: household weaving on the basis of ziliumian rations; collective workshops that obtained reject and low-grade cotton through gray or black channels; and “relief” or “self-reliance” weaving, organized with government help in disaster-stricken areas. Of the 566 million meters of tubu cloth produced in these ways, 366 million were sold on the black market, while 200 million were retained for household use.Footnote 52 The rationale for commercial handloom weaving is not difficult to understand: in order to ensure the profitability of its cotton mills, the state set cotton procurement prices low and cotton textile prices high. In this situation, turning cotton into cloth for the black market was vastly more profitable than selling cotton to the state.
Xu Jianqing estimates that cloth made from one kg of cotton earned a profit of 24 yuan, while the same amount of cotton fetched only 2 yuan if sold to the state. Since it took seven to ten days to spin yarn and weave cloth from one kilogram of cotton, a spinner/weaver could earn up to 3.4 yuan per day, a very good rate for that time.Footnote 53 The scale of rural weaving can be seen from a report from Hebei Province, which states that more than 1 million persons in 68 counties produced a total of 33 million meters of tubu cloth.Footnote 54 Not even the highest authority in the country, the Party Center, could stop the tubu trade: a 1960 “Directive to immediately end the hand-spinning and hand-weaving of cotton” was reissued twice, apparently to no effect.Footnote 55
According to Supply and Marketing Cooperative data, tubu cloth accounted for 23 per cent of all cotton cloth produced in China in 1955, and 21 per cent in 1956. In 1964, four years after the ban on hand-spinning and hand-weaving, it still accounted for 12 per cent of total cotton textile output.Footnote 56 The 366 million meters of tubu cloth produced in 1964 added 0.75 meters to the yearly cloth supply of the average Chinese person, or 1 whole meter if we include cloth that was produced from substandard or recycled materials – not an insignificant amount, a time when the average cotton cloth supply was 4 meters per capita. Assuming that it takes 1 to 1.5 days to spin and weave one meter of tubu cloth, the 1964 output of tubu represents 366 to 549 million workdays.
Cotton and Gender in Guanzhong: a case study
The remainder of this article is based on fourteen weeks of fieldwork, carried out in five separate stints between 2005 and 2011. We conducted about 200 interviews with more than 80 villagers, mostly men and women born between 1920 and 1945, in Gedatou village (Zhouzhi county) and Danbei and Zhangli villages (Xingping county).Footnote 57 All three villages are located in the Wei river valley of central Shaanxi, a region also known as Guanzhong, “the land between the passes”.Footnote 58 Guanzhong comprises most of the flat and fertile land of the province and is relatively prosperous, at least in contrast to the mountains of southern Shaanxi and the dry loess plateau of the north. However, Shaanxi is a landlocked province far from major urban centers, and few villages in the area have industries or sidelines. Average household income in the area is slightly below the national average, and has been so for the past fifty years.Footnote 59
No single village can represent all of China, but I am confident that the places we studied were not outliers in any major way. Discussions with Chinese colleagues who grew up in the countryside or lived there as sent-down youths confirm my belief that these villages were not atypical: I have heard reports of hand-spinning and hand-weaving in the collective era in rural Shandong, Henan, Hebei, Jiangsu, Zhejiang, Hubei, and Sichuan. I have begun comparative research in other places (Jiajiang and Pixian counties in Sichuan, Dantu county in Jiangsu) and plan to do more interviews and archival research in Shandong, Hebei, and Jiangsu provinces. My hypothesis is that patterns similar to those I found in Guanzhong exist in areas that had an earlier tradition of household weaving, i.e. in the cotton-producing provinces of northern and eastern China (Hebei, Henan, Shandong, Shanxi, Hubei, and Jiangsu) and to a lesser extent in Hubei, Anhui, and Sichuan, but not in southern or north-eastern China, areas that produced little cotton.
Cotton production in Guanzhong dates back to the early Ming dynasty and underwent a major revival in the 1930s, thanks to the linking of Shaanxi to the national railroad network and the promotion of cotton cultivation by the provincial government.Footnote 60 Cotton was potentially very profitable, but also risky and expensive to produce. Farmers rarely put more than one-third of their land under cotton; a common plot size was one to two mu, the size of a large garden.Footnote 61 Weaving was widespread: it was commonly said that “all families [in Guanzhong] spin and no one sells cotton; all villages weave and no one sells yarn”.Footnote 62 Weaving for distant markets in Gansu and Qinghai was concentrated in Xingping county; neighboring Zhouzhi county participated in this trade and also sold some cloth to mountain villages in the Qinling Range. While not a road to riches, weaving for the market was not as badly paid as is often assumed. Depending on the relative prices of grain, cotton, and cloth, a textile worker in the 1940s could earn enough to sustain herself and one to three other household members. Old people in Xingping maintain that in the years before the Revolution, a diligent weaver could feed a family of four even if she was the only earner in the household.Footnote 63
Most girls in Guanzhong learned to spin at age seven and to weave at age ten to fourteen, depending on the type of loom they used. Families that did not need the labor of their daughters in the fields groomed them for marriage by cloistering them soon after they had learned to spin. One woman I interviewed was told to stop visiting friends when she was nine; with the exception of trips to nearby temple fairs, during which she and her sisters watched village operas from behind the closed curtains of an oxcart, she did not leave her parent's home until her wedding day.Footnote 64 Cloistering was not always that strict, but most women I interviewed remembered a childhood spent indoors at the wheel and the loom. This regime was relaxed on the seventh day of the seventh month, when girls and young women came together to pray to “Seventh Sister” (qi jie, also known as qi xian, the Seventh Immortal) for needle skills. This was an all-female festival in which girls and unmarried women met to sing, dance, and pray; married women were only allowed to watch from behind a screen. The preceding weeks, in which girls met to produce an effigy of Seventh Sister and practice songs and dances, were periods of intense socializing between girls and young women who were usually confined in their homes.Footnote 65
Gail Hershatter has pointed out that norms requiring women to work indoors at all times were so strong that they skewed their memories, making them downplay the extent to which they worked outside the home.Footnote 66 Many of the women I interviewed remembered participating in farm work before the revolution: weeding, picking cotton, in one case even driving carts and plowing fields. However, most of them remembered outdoors work as an (albeit common) exception from a moral norm, and associated it with poverty, danger, and isolation. For a woman, skill in planting crops or handling animals did not bring recognition: her proper work was indoors, and she could earn praise for herself and her family only through domestic work.
Almost all important interactions in rural Guanzhong involved the exchange of textiles. Births, cyclical festivals, and visits to the natal family were occasions for giving textile gifts, and no marriage could be concluded without gifts of cloth and cotton. The customary standard for a bride price in Zhouzhi was “two bundles of cotton and four bolts of cloth” [liang kun mianhua, sige bu]; enough for two padded winter suits and two padded quilts or, if spun into yarn, for ten lined suits of clothes.Footnote 67 During the first few years of her married life, a woman was expected to produce gifts for her in-laws; if her marital family allowed it, she would also take textile gifts home to her natal family. Gifts of cloth were also mandatory when children were born. Even the dead had to be clothed, and since they wore their clothes for the length of their afterlife, these had to be of the highest quality.Footnote 68 Preparing grave clothes for oneself and for close relatives was seen as an appropriate and dignified end of a woman's working life. Textile work, in short, shaped all aspects of women's lives, in ways that could be both oppressive and empowering.
Early PRC politics were more concerned with reforming the feudal family than with changing women's work. Freedom of marriage and divorce and the right to resist cruel treatment in the family – these were the slogans that attracted many women to the CCP. Yet from early on, participation in productive work, always understood as work outside the household, was portrayed as the final aim of women's liberation. Women I interviewed in Zhouzhi and Xinping remembered little change in daily life and work routines during the first years of the PRC. After collectivization, however, women's participation in collective agriculture became the norm. The transition from a situation in which most women worked indoors to one in which they worked in the fields alongside men (though usually in separate teams) was extraordinarily rapid. Women's Federation records from Shaanxi indicate that women's participation in agricultural work tripled from 30–50 labor days in 1955 to 140 days in 1956.Footnote 69 A 1955 report from Zhouzhi shows that the average woman performed only 9.6 days of farm work a year.Footnote 70 Four years later, women's labor participation had risen to the same level as men's.Footnote 71
The Great Leap Forward with its large infrastructural campaigns further intensified labor demands, as men and women were mobilized to level farmland and to build roads, reservoirs, and canals. Participation in these projects gave new visibility to women's work, but women who wanted to achieve public recognition needed to outperform men in labor competitions and out-suffer other women in public shows of self-denial. As Kimberley Manning has shown, women activists at that time were often indifferent or even hostile towards aspects of CCP policy that stressed the reproductive role of women – health programs, sanitation, protection from overwork and disease.Footnote 72 In Guanzhong, too, young female activists demonstrated their disregard for the reproductive body and embraced physical suffering as the route to liberation. Not surprisingly, most women refrained from activism and continued to devote their time to reproductive work in the household, even if such work gave them no public recognition.
Next to the large infrastructure projects of the Great Leap Forward, it was cotton cultivation that added most to women's workloads. Cotton is a labor-intensive crop at the best of times; it was especially so in Maoist China, where weeds and caterpillars were removed manually and cotton was picked by hand. Because cotton yields tend to deteriorate after a few years of cultivation, teams had to mobilize more and more labor simply to keep output at constant levels. Traditionally, cotton cultivation in Guanzhong had been men's work, but it became feminized after collectivization. Tens of thousands of women participated in so-called “Silver Flower Contests”, competitions to raise cotton output through improved – but often vastly more labor-intensive – cultivation techniques. As Gail Hershatter and Gao Xiaoxian have shown, these campaigns had a transformative effect on female activists, some of whom rose to national prominence.Footnote 73 However, women who did not aspire to labor model status often experienced these contests as exhausting and badly remunerated production drives.
While women's participation in the fields increased dramatically, little was done to reduce women's textile work. In theory, the introduction of rationing in 1954 should have ensured that every person received an adequate supply of factory cloth. Cloth rations differed widely, but rural rations were consistently lower than urban ones, by a ratio of about one to two.Footnote 74 Average rural rations in Guanzhong dropped from around 5 square meters in the early 1950s to less than 1 square meter in the crisis years of 1960 and 1961, and then gradually rose back to about 4 square meters.Footnote 75 Based on interviews, I estimate a yearly subsistence minimum of 3 square meters of tubu cloth, or 4 square meters of factory cloth (which wears out more quickly). This is enough for one suit of summer clothes, one suit of winter clothes, and a quilt and blanket shared between two, to be replaced every three to five years.Footnote 76 Twice that amount (6 square meters of tubu or 8 square meters of factory cloth) would constitute a social minimum – the amount needed, according to local standards, by a person who is poor but not destitute. Ration supplies, in other words, were sufficient to keep people covered, but not to keep them decently clothed according to the very frugal standards of the time.
However, rural people in Zhouzhi did not use their rations. All my informants agreed that only village cadres (who were expected to wear “urban” clothing when they attended meetings) and young married couples used factory cloth. Most other people sold their ration coupons on the black market and continued to wear tubu. The reason was cash scarcity: high quotas and low state procurement prices for wheat and cotton depressed collective incomes and reduced the amount of cash and grain that was distributed to team members. In Zhouzhi, the value of a male labor day was around 0.5 yuan, with substantial variations between teams; women earned about 0.4 yuan a day. Assuming 300 workdays for a man and 250 workdays for a woman, a family with two earners received 250 yuan a year. Since most of this income was retained by the collective to compensate for grain consumption, disposable income was often as low as 50 yuan – barely enough to purchase salt, vinegar, school books, and other necessary items. By making cloth at home, households not only saved the cost of buying ration cloth (between 5 and 15 yuan per person) but also earned an additional 1.2 to 6 yuan from the sale of their coupons.
As mentioned above, the main source for homemade cloth was the “self-retained cotton” [ziliumian] ration, set in most teams at one kilogram per person. One kilogram of cotton yields about 4 square meters of tubu cloth – more than the absolute minimum. However, only part of the ration could be used for spinning: in an area where few people owned any wool or leather garments, warmth in the winter came from cotton padding. A warm winter suit contained about one kilogram of cotton padding; a large quilt required 1.5–2 kilograms. If we subtract padding as well as losses in spinning, we arrive at a yarn availability of 590 grams, equivalent to 2.5 square meters of cloth – somewhat below the subsistence minimum.
People without additional sources of cotton would have reached the end of the collective period with their clothes not only patched (as indeed most rural people did) but in complete tatters. How, then, did people manage? Part of the answer was pilfering: while farmers I interviewed often maintained that taking cotton from the fields was too dangerous, former team leaders said that the practice was widespread and joked about women “returning from the cotton harvest with big bellies”, because they had stuffed so much cotton wool under their belts and jackets.Footnote 77 Teams hid cotton and sold it on the black market to finance necessary purchases of fertilizer, pesticides, and fuel. Many teams also maintained a social fund to help families defray the costs of marriages and funerals. Young men who were about to marry received 1–2 kun (5–10 kilograms) of ginned cotton to help them pay their bride price; some teams also gave 1–2 jin (0.5–1 kilograms) of cotton to help families clothe their dead. At year-end meetings, when teams distributed cash and grain income, many teams also gave out “secret” shares of cotton – sometimes as much as one extra kilogram per person.Footnote 78
Another source of cotton was the black market, fed by cotton stocks that collectives concealed from the state. Cash-strapped farmers rarely bought cotton for consumption needs, but in a common pattern, women bought cotton and transformed it into cloth, which they exchanged for more cotton. By doing this repeatedly, they could earn a cash income or accumulate cotton for a dowry. Profits in black-market weaving were relatively high, especially if one made the long trek to the cattle fairs of Dianzhen and Mazhao, where people from the Qinling Mountains sold livestock and forest products and purchased grain and cloth. Several interviewees, including a former chairwoman of the village women's federation, admitted that they sold one or two pieces every year; one old lady even claimed that she sold about ten pieces every year for several consecutive years. Most people, however, said that weaving for the market was too risky, and that in any case they could not withdraw labor from the team to the extent necessary for commercial weaving.Footnote 79
Rural women under the collectives found themselves caught between conflicting demands. On the one hand, state and collective leaders expected them to work in the fields; on the other, husbands, children, and parents-in-law demanded decent clothes. Because of pervasive scarcity, the choice of one option was usually a choice against its alternative: one more hour of work in the field meant one less hour at the loom; more work points earned in agriculture meant fewer winter clothes; tidy clothes and the respect they brought came at the expense of reduced cash and grain income and perhaps conflict with the team leader. These were not just material considerations – full stomachs versus warmth and comfort – but choices between conflicting moral obligations. The leader of the women's agricultural team was not a remote functionary but a neighbor and perhaps a relative; her demand that each able-bodied woman worked three shifts a day could not simply be shrugged off.
Many women I interviewed also felt a deep debt of gratitude towards Chairman Mao and the Communist Party – not for improving their material lives, which remained impoverished, but for raising them above the status of a commodity that could be bought and sold. Old women still remember their price in cotton and endure teasing remarks by men about how precious or how cheap they were at the time of their marriage.Footnote 80 Against this debt of gratitude stood an obligation to provide clothes for the entire family so that they looked respectable in the eyes of the neighbors, to find good spouses for one's children, and to participate in the ritualized give and take that knit communities together. Openly or tacitly, husbands and in-laws encouraged women to side with family and community against state and collective: to glean cotton from the fields, to shirk collective work, and preserve energy for the long hours of domestic work that began after dark. However, when women were caught stealing cotton or shirking their collective duties, they alone were blamed – as they alone were blamed, and blamed themselves, for struggling with conflicting duties: “Who could I blame for being overworked? It was my own fault that I had so many children!”Footnote 81
For many women, the only way to square the circle was to work ever longer hours, up to the limit of physical endurance. Most women's working days started before sunrise, preparing breakfast while men and children were still asleep. Fieldwork shifts lasted from 6.30 to 8.30 am, 10.00 to noon, and 2.30 to 6.30 pm, or, in busy periods, until sunset – a total of eight to eleven hours. Meals were short and hasty affairs, followed by a nap or a card game for men and household chores for women. After the end of the third shift, women cooked dinner and prepared food for the next day, washed dishes, and put the children to bed; by the time they were done with these tasks, it was close to midnight. It was then, when men and children were asleep, that most textile work was done.
Most women learned to spin at candlelight, in the moonlight, even by the light of an incense stick. In busy times, it was not uncommon for women to sleep only three to four hours every night. They kept themselves going by taking one-minute catnaps with their heads resting on the spinning wheel. Several women I interviewed said that the only real rest they ever had was when they fell sick from exhaustion, and even then they returned to work before full recovery. Pregnancy did not provide much of a reprieve either, because women worked into the eighth month and returned to heavy household chores (though not necessarily fieldwork) one month after birth.Footnote 82
As Stevan Harrell, an anthropologist of China, has pointed out, all revolutions are in a sense about the division of labor – about “what is work and what is not; about the proper kinds of work for men and women, educated and uneducated, rich and poor; and specifically about the differences between household-based and non-household based labor”.Footnote 83 The Chinese Revolution – here understood as a long process beginning in the nineteenth century and involving social, cultural, and economic as well as political transformations – replaced one division of labor, one world of work, with another.
In the old world of work, discredited among intellectuals but still very much alive at mid-century, almost all work took place in domestic arrangements based on gender and generational hierarchies but also on mutual interdependency between men and women. In the new world of work, labor was divided at the scale of the nation, between city and countryside, industry and agriculture, producers and consumers. Household-based textile work, for centuries the archetype of women's work and one of the two fundamental occupations of the realm, became insignificant and invisible: not being paid and public, not contributing directly to national wealth and strength, it could no longer be considered work. To the extent that it was perceived at all, it came to be thought of as part of physical reproduction: necessary perhaps, but repetitive, circular, and ultimately futile. This was true in particular under socialism, which glorified future-oriented work in factories and fields, work that remade the face of the earth, created beautiful new things, and moved the nation forward.
Textile work did not produce future; it reproduced life as it was, in its physical and social dimensions. The metaphor is tired, but women who span and wove did indeed reproduce a social fabric – in their labor at the wheel and the loom, in the exchange of cloth, and in the everyday use of clothing. These dimensions of rural women's work were only dimly perceived by a state that kept its eyes on distant goals and saw little value in the complex social ties that held village society together. Consciously or not, the state undermined local social reproduction: by banning showy ceremonies and excessive ritual expenditure, but more importantly by directing cotton and other raw materials towards “rational” uses, and thus denying rural people the material basis of their social and ritual life. It is interesting (but perhaps not surprising, given the many ways in which state socialism accepted the basic premises of the system it aimed to replace) that the Chinese socialist state played a role elsewhere left to the market: that of a disintegrative, disembedding force that disrupts local circuits of exchange and inserts goods and people into larger flows that are better adapted to the production and extraction of surplus value.
But of course socialism did not replace household-based textile work; it perpetuated it. The ironies are palpable: the effort to construct a textile industry that would enrich and embellish people's material lives and liberate rural women from the loom instead led to the entrenchment of the very industries that it sought to replace. This was not simply a survival of an outdated industry, kept alive by a sentimental attachment to homespun cloth or to traditional gender roles. Women continued to spin and weave by hand not in spite of but because of the industrialization of the textile sector – because of policies that, on the one hand, left little cotton in the countryside, but on the other kept rural people so undersupplied that they were forced to produce cloth with whatever scraps of cotton they could find. State planners could have attained their dual aims – mobilizing all rural female labor for work in agriculture or infrastructure and ensuring that all cotton was processed in factories and contributed to accumulation – by raising farm incomes or by lowering the price of factory cloth. However, this would have reduced profits in state industry and slowed down capital accumulation.
In other words, as much as state officials criticized backward and inefficient manual textile industries, as much as local governments tried to minimize the amount of cotton left in farmers’ hands, there was no serious attempt to make rural people switch to factory cloth. Instead, planners tacitly assumed that rural people could make do with very little clothing and that rural women would somehow manage to provide for the most pressing needs, despite the fact that state and collectives left them with little raw materials and labor time. Unpaid and invisible as it was, women's textile work underpinned socialist accumulation, as much as it underpinned the reproduction of village life.